Jump to content
Sports Interactive Community
Kamakazeee

Mike Ashley meets Moneyball and Sabermetrics

Recommended Posts

rsz_moneyball_blog_1.jpg

Mike Ashley meets Moneyball and Sabermetrics

Introduction

For a while I've wanted to take a real life club model and try to implement it to its truest fullest possible state on football manager. Whilst I know that the game is limited in certain areas and it will be difficult to import realism to its fullest effect, I do believe with outside sources of help such as excel spreadsheets coupled with in-game data that I can implement Money-ball to great detailed level. Whether this model will be effective in game the rather than going to sign all the known wonder-kids and players about on the forums remains to be seen. But, this game isn’t about going onto to world domination as soon as I can, it’s about implementing a real life model and gauging the success I have coupled with the amount of realism in game I have managed to get out of the system. I also want to make people aware in advance that the version of the game I will be using won’t be the most recent version of football manager, I just can’t get into the game for some reason and I much prefer my games on football manager 2014, so I hope that clears up any confusion before we start the read.

What is Money-ball?

The new craze amongst the nerds and statistical geniuses among sports. Money-ball, it has seen a bigger rise to fame within sport over the past few years than Robbie Savage quiff featuring on match of the day on a weekend. Money-ball, is the name of the book on the end of most people’s tongues associated with sports around the world, wrote by Michael Lewis in 2003, Soccernomics. There is also an film adaption of the book starring Brad Pitt, both the book and film tell the story of American Baseball team, the Oakland A’s, who’s general manager Billy Beane, in some people’s opinions and others not so much, revolutionised player recruitment. He began to use the depth of information widely made available by today’s technology and took a greater interest in the statistics made available on players and used such information to cast the final verdict on whether certain players were feasible as the best option for the limited budget he had to improve the team. The approach paid off for Beane, as his team the Oakland A’s, romped to the playoff in two consecutive seasons, all whilst operating on the third smallest budget in Major League Baseball.

So you ask how did this happen? Well this I will go into greater detail later, stating the rules and guidelines that encompass Money-ball to allow the method to work effectively, but the term sabermetrics began to fly around and became the new buzzword along with Money-ball. Is it effective? Will it work in football? Is it a bird? Is it a plane? Sabermetrics is the statistical use of data relating to the performance of players in such a way it will justify the transfer of the player. It uses wealth of statistics, the player’s circumstances and the level the player is playing at etc, leaving no stone unturned in creating the big picture to determine whether the player is best option furthermore the best financial option.

Real Life examples

So, we know the basic outline of what Money-ball tries to envisage, but how does it figure in real life? Are there any success stories in football of this philosophy working? Well they are few and far in-between, but there are some cases in which Money-ball is deemed the model followed, but first let me start you off with a quote from a former professional and how the old guys of football see the Money-ball model. You know the type, proud enough to clean their own boots, teas at 4 o’clock, the bread and butter of which footballs roots.

“They want a head coach and mathematical modelling … Ha, ha, good luck with that.”Micky Quinn, Former Newcastle United player.

Not good huh? Well football is an ever changing world and clubs are having to adapt all the time to new ideas or they are running the risk of being left behind in the modern era. However, I don’t want to run the risk of going off topic, so I’ll get back to real life examples of the Money-ball in football.

Matthew Benham, the owner of Brentford in the Championship and FC Midtjylland in Denmark, has recently come out and stated that Mark Warburton and Brentford will be parting ways regardless of whether there current league campaign ends in promotion to the Premier League and instead opting for a continental approach, typically introducing a head coach rather than a manager, a sporting director to oversee the running of the club and the use of a mathematical model on the topic of player recruitment. Obviously, this hasn’t been taken well by the more old fashioned people linked to the game, which accounts for Micky Quinns opinion which actually directly related to the scenario that panned out in Brentford. Following on from this, Brentford’s future is actually already been panned out in the form of the other club Benham owns, FC Midtjylland, reports say he rescued the club from obsecurity in July 2014 investing £6.2Million into the club and instating Rasmus Ankersen into the Chairman position of the club. Ankersen is 31 year old former player who clearly buys into the Money-ball philosophy Benham wants to imprint into the DNA of the clubs he owns and if you go buy the success of his club in Denmark, quite rightly so, as they lead there league by eleven points, on course to win the first trophy of the clubs history, all by challenging the conventional ways in which a football club is run by backing Money-ball.

Benham also holds a detest for the registration in England, as this obviously limits his options when looking at players under the Money-ball principles, whereas a team on the continent such as his team Denmark aren’t limited as such to domestic based players. However, saying this sounds like he wants to just have a worldwide based team in his registration of players, but he does see the future in investing in the academies of the clubs he owns and played a massive part in getting the Fc Midtjylland academy off the ground and coupled with Sabermetrics and Money-ball they have a thriving academy at the club, more notably producing the much sought after in Europe, Demark youth player Pione Sisto. The same can be said for Liverpool who I will discuss below, as there academy is producing some more than decent players at this very moment, Jordan Ibe and the emergence of Raheem Sterling to name two.

Interestingly Benham’s teams are also share a strong link with E4Talent, which claims to be “the future of scouting and performance analysis”. E4Talent is a client based programme which offers its services to clubs and allows the access to the wealth of statistics there business acquires. The main method of E4Talent in acquiring the statistics is they used a self-invented term, MPI, Match Performance Indicators, which set up as a indicator to players performance in comparison to a peer playing in the same position.

Another high profile use of Money-ball and sabermetrics being used, albeit I don’t agree to the degree Benham proposes his clubs go by is Liverpool, who use it with much more financial clout. John W Henry, the owner of the Boston Red Sox liked the system so much he introduced it to his own baseball team.

Once he took over Liverpool it was a given that he wanted to implement the system into the club he had just bought and after a while, post Dalglish, the signings thye are making you can see have a great Money-ball flavour, examples would be Emre Can, Mahamdou Sakho, Daniel Sturridge and Phillipe Coutinho. Now none were big names in their own right at the time, more unpolished gems, who had potential, the wrong side of twenty struggling to reach the potential there ability previses them to have. They had all been decent throughout there career, but not world class, maybe not even to the level you would agree they deserve to play for a club at the level of Liverpool, who at the time needed great investment into the playing squad, however with some much need TLC the players are turning heads and the recruitment team behind the transfers are getting there just rewards. Sturridge has turned into one of the most feared strikers in the league, Coutinho one of the best playmakers in the world, Emre Can has formed a great trio alongside Skrtel and fellow Money-ball signing Sakho at the heart of Liverpools defence. All the signings are at the right age so that Liverpool can enjoy the best playing years of their careers and all offer great resale value on what their value actually was when they first signed.

However, Liverpool do have examples of the not so down sides of Money-ball but showcases the risks involved in taking some players. Fabio Borini will be my first example, hes a decent player, hes had success at the clubs hes played at throughout his career. Roma, he enjoyed his break out season in which rewarded him with his transfer to Liverpool due to sabermetrics deeming he was a much viable option financially and statistically and at the age he was at the time, he was young and had potential and proved he can play well, compare this to say a more well known striker who was available at the time Van Persie who signed for Man Utd, he was at the age where no resale value would be offered, yes he met all the statistics Money-ball encompasses, but he didn’t fit the rest of the criteria. Borini has since had injuries problems, he enjoyed a decent yet unspectacular loan at Sunderland but has been unable to nail a first team place in the Liverpool squad. A wasted transfer? Well not exactly, signing him at the young age, he still has the chance to turn his career around and still has the reputation for being a more than able goal scorer abroad due to his feats at Roma.

A further case of Money-ball being active in Liverpool’s transfer policy which shows a completely different side to the model than anything wrote above is the transfer of Mario Balotelli. Here we have a player who has travelled club to club, he has problems at all the clubs, doesn’t come with the best image, but he does have a decent record where ever he has been, is it worth the risk? Well Money-ball would say yes! A feature which Money-ball provides is taking a player who has shown the quality and potential to be a more than decent player, he is at the right age, his statistics have been good enough to flag up some interest, but the problems surrounding him have had an adverse effect on his career. Money-ball states that this player would be great choice, as the club will bring the player in, offer an around the shoulder sort of situation and get the best out of them. Albeit this might not have worked with Balotelli yet, he has shown quality in parts this season.

So we have a basic idea of what Money-ball is and how sabermetrics are calculated into making this policy work, we know a few of the details covered by this model and how real life clubs have taken an interest in implementing this into the running of their clubs. I will go into the ins and outs of the model further at a later time, but above we have working and not so working examples but it clearly states that Money-ball is active in the football world.

Football Manager and Money-ball

So, we have outline, we all have a basic understanding of Money-ball is, it’s now the opportunity to devise a plan on how we envisage this model being used in game. Firstly, we need to get the basic principles of Money-ball, which I have acquired from the book mentioned above “soccernomics”, which I would recommend to anyone from any sport as it’s a great. Below I will list the basic “rules” this game will abide by and an explanation of how I plan to implement them into football manager. I also have had the pleasure of reading the book “The Numbers game” by Chris Anderson and David Sally, from that book I have decided on two rules I will add additionally to this game as I fell they accompany the vision of Money-ball perfectly.

Yearly wage spend should be considered more than yearly transfer spend.

I think this pretty self-explanatory, however I will be implementing a financial prudent model to the club in question, but I will explain more on this in the correct section.

Give the old boys a chance – Don’t take over a club and rip the heart out of it and replace every player – The New Manager Syndrome.

To accompany this rule, I will not be making any transfer dealings within the club within the first season, and will try to keep the original squad as together as much as I can, but only in correspondence to the other rules surrounding Money-ball and the principles I am going to put in place.

Don’t buy players who looked Gucci at international tournaments: they’re likely over-valued and past performance is no indication of future performance, especially when they’re playing with a different team – there are different incentives and a different tactical set-up at tournaments, and it’s a super small sample size.

Again pretty self-explanatory, but by going against this rule it would go against the visions of Money-ball anyhow, seeing as all signings need to be certified by sound stats over the course of a season which justify the transfer.

Take into consideration more well-known/rising countries are overrated, England, Holland, certain South American countries.

Again simple to understand really, I also think this is something that all fans can agree on, particularly in England, where it’s a well-known fact that English players cost is at a premium. However, this is not to say I won’t sign players from these regions, I will just keep into the DNA Money-ball outlines in getting the very best for the best financial option.

Sell players at the right time, don’t miss out on resale value.

One of my favourite rules, this option ensures that there is no place for sentiment and that all players will have to be moved on at some point to ensure that the money spent on the player over his career at the club is recouped to some extent.

Use the wisdom of crowds: ask all your scouts and a Director of Football if you have one.

I don’t think this needs too much explanation, it clearly states that all transfers need to be judged by numerous means within in the club, which in this case will be a few of my best rated scouts, as I won’t be including a Director of Football in my game.

Buy players at the right age, sensibly early twenties, eliminating the risk of teenagers not developing.

I think this ties in nicely to what I said in my introduction, in that all players signed will need to have the statistical data to falsify the transfer rather than just the name, reputation, attributes and potential, in fact you could probably count the first two out and whereas the last two will play some part, as I will need certain types of players to fit systems. Potential may not be too important, as since I am not signing teenagers and more players in the early twenties they should be more or less near there full potential, but statistics backing the player up will be priority. More will be explained in the section where I explain the transfer strategy.

Centre-forwards cost more than they should, develop your own where possible.

Centre-forwards are the celebrity of any team being the primary goal scorers and because of the stardom their role attracts they will command a excessive transfer fee. So, where possible the club/I will try to develop my own forwards if the quality is within the club, if not the club will try negotiate the best possible deal for the best possible player, backed up by statistics of course.

Sell any player if a club offers more than they are worth and try to replace them before they are sold.

Self-explanatory, any offers for any players that exceed the worth for the club set by primarily, but I am going to set a 35% limit here. So basically if I don’t want to sell a player and I deem him worth more than the offer being made, but the offer is 35% higher than the game value of the player, the offer needs to be accepted. Keeps a sense of regularity in the game.

Don’t buy players for the sake of it: the ultimate goal is to develop a youth network and try to develop your own players.

Again a financial prudent transfer policy will be put in place and only signings made out of necessity will be made, however before signings are made there will be the option of replacing the player or gap in the squad in house.

The Bloody Clough effect.

Take a player who has had multiples issues at his parent club, drinking, drugs and gambling etc, and make him feel wanted and take him under your wing to battle his problems with the aim of getting his talent out onto the pitch for you, basically a undervalued player, this may be hard to replicate on football manager,but we could use players who want away for there clubs, rotting in the reserves etc.

The Numbers Game rules.

The first is the theory that the best way to improve a team is by identifying and replacing the weakest links, rather than by splashing out on making the best links even better.

This ties in nicely with the rule above and not making signings for the sake of it and also ties in nice with Money-ball as the game is all about statistics and any player not making the grade from a statistical point of view will be viewed as a weak link within the team.

The other idea that I like, especially as a newly promoted side, is that a clean sheet is worth just over two goals scored in terms of points across the course of a season.

Another rule easy to envisage as being simple to implement into football manager, we just have to ensure we have a defensively stable team and the first port of call for transfers will be fixing the weak links in defence, which ties in nicely with the rule about new managers, as the first season will give me chance to analyse the squad.

So, after that exhaustive passage of writing we have our principles for the game and the next few pieces will explain how all this will tie in with the game in terms of what team we will be managing. What systems we will use to implement Money-ball. Tactics and training. Transfer policies. How we will manage data and statistics. Squad Management etc. So still plenty to get on with. I must also stress that isn’t going to be a fast paced thread with updates flying out every day as it’s a very time consuming way of playing and for the first few updates at least will be very in-depth.

Anywho, I hope you have enjoyed the read so far and I haven’t bored you. I hope this gets some great discussion on the go.

Share this post


Link to post
Share on other sites

rsz_moneyball_blog_1.jpg

Mike Ashley meets Moneyball and Sabermetrics

Following on from the opening post I am now going to go on and explain the transfer strategy at the club I have chosen to implement this model upon. But first I will outline the club I have chosen and why. Newcastle United are the lucky club who I will take charge and revamp to ensure that the implementation of Money-ball is able to operate to its most efficient state.

688.png

So why Newcastle you may ask? Well, they are the club I support, as with depth and time consuming nature of this game, it had to be a team I was actually really interested in, something which could hold my interest for a prolonged amount of time. I also feel that operating in a fairly obscure league in Europe whilst in the short term would offer a great challenge, but in the long term once the infrastructure is in place, I feel the game could become a tad uncompetitive domestically and I always find these games lose their interest after a while. Whereas, managing in the premier league and constantly having to cycle players in and out of a squad playing in the best league in the world would offer a great challenge. I also know the club inside and out, which will play a massive part in the development of the structure needed to implement this model, more so for the first few seasons that the latter area of the game. This game is also got a touch of a personal flavour, due to my absolute detest for the way the club is run and the boards seeming reluctance to invest in the playing squad in real life. I was also curious to see how my Football Manager game talents could effect this game, in terms of being able to keep a demanding club competitive in one of the toughest, most relentless leagues in the world. A league which is swamped with money, a league where clubs aren't afraid to spend. How would Money-ball fair against the riches of the top end clubs in the premier league, would implementing this model hinder or progress the competitiveness of the team I'm going to manage. Well, to some extent we can’t really answer that question till I actually get into the game, but what we can presume before the game has started that this method of game play and running of a club will have a positive effect on the finances of the club. It can also be a fair assumption that the club will no doubt have a solid defensive base, due to the rule from the numbers game stating clean sheets are of the utmost importance and the more favourable investment towards the defensive structure of the game. Another fair assumption is that the infrastructure and facilities of the club will no doubt be classed among the world’s which I currently believe stands in the league’s best, this again though is due to the rules outlined in the first post, in regards to the favouring youth when and where possible.

Newcastle United – The Chosen One

So a little bit on the club in question, the board expect a top half finish, something which the current squad should find no problem at all, but in progression of the save, I think it’s fair to presume that if Newcastle aren't challenging for domestic cups, European places and even the league at some point in the save then you have failed in your quest at managing the club. Newcastle also offer a solid base to start with, they easily have a competitive squad which comfortably sits in the top half of the league, whilst also offer a challenge in getting the clubs finances in order to keep within the strict financial prudent model I am aiming to introduce to the club. The club also has a few players reaching the back end of the age model Money-ball states as needed to shift out to ensure resale value is kept at an absolute maximum without hindering the clubs performance on the pitch. Newcastle also have many players who are of great interest to many teams throughout the team, notably in other games I have seen the one of if not all of the trio consisting of Davide Santon, Cheick Tiote and Hatem Ben Arfa, transferring clubs. So the transition period of this club will be very enjoyable yet challenging in ensuring the club doesn't take a step back in the progress it will no doubt take. The club also offers a great base in terms of the facilities already on offer and also the reputation of the club worldwide. The training facilities currently stand at a rating of superb and the youth facilities are deemed as great, so in the short term not much to be done here. However, they will need to be improved upon in the future to ensure this model reaches its optimum efficiency, but from my past experiences the Newcastle board are very tight and convincing them to improve the facilities is a challenge within its self. On the youth recruitment side of the club, it seems to me they are at decent level in regards to the rest of the league, but definitely enough scope for improvement, with currently both the Junior Coaching and Youth Recruitment rankings both standing at average.

The financial state of the club is actually doesn't look to bad, they have decent amount in the bank amounting to twenty-six million and offer a wage budget of Eight-hundred and ninety thousand, currently spending fifty thousand under that and a transfer budget of six million. Debts at the club currently amount to seventy-two million as net debt and a gift loan from Mike Ashley amounting too one-hundred and twenty-nine million, which is repayable by 26/6/2028, at a rate of just over five-hundred thousand a month but only when the club is running at a profit which is quite handy. The club also starts with two kit sponsors, one lasting four years amounting to eight million a year and the other two years amounting to two million a year. But, undoubtedly the clubs great stadium St James Park is the primary source of income holding a capacity of over fifty thousand and over thirty five thousand season tickets being sold and a very impressive average attendance of near capacity every game, again amounting to over fifty thousand. However, the clubs finances I will go into a little more detail when I explain the financial model that is going to be implemented and how.

Newcastle_Information_Facilities.jpg Newcastle_Information_General.jpg

Screen shots of Newcastle United Football Manager information screens.

Aims and Objectives

Now before I continue, I am going to outline the objectives, aims, phases whatever you want to call it, but these will be the aims of the save and what I want to achieve in each phase of the game.

1: Domestic Consolidation and Foundation Laying

• Top half domestic finishes

• Qualify for European competitions

• Begin to improve the youth infrastructure among the domestically ranked best.

• Clubs Financial model will be implemented and churning out results.

• Money-ball will be having an actual visual representation within the club and its operating modus.

• A viable and productive scouting network in place.

2: Domestic Dignity

• Consistent title challenges year-on-year

• Steadily improving youth production and starting to see the dividends.

• Beginning to focus more on English talent, with the aim of being represented in the national team.

• Club ranked in the top fifteen richest in the world.

• A domestic cup should have been won.

3: Ranking amongst the best

• Improved performances in continental competition each year.

• Viable youth academy and system. A constant channel of youngsters floating through the club.

• Facilities of the club should now all be ranked amongst the best possible.

• Money-ball should be now in full effect, and if its this far proving to be more than decent operating model.

• Domestically should be ranked amongst the top three in the league.

4: Continental Dominance

• Year-on-year challenges in European football

• Flourishing youth academy, 25% of English squad developed at Newcastle.

• Richest club in the world.

• 60% bias towards incoming transfer funds than out, over the due course of the game. Evident Money-ball as a transfer strategy is successful.

5: World Dominance

• Ranked best team in the world, Money-ball being the primary factor.

Right now we know the club, we know the aims of the save and where want to be in the somewhat flexible contents of phases in the clubs development I have outlined above, I am now going to outline the transfer strategy that will be used and what process will be used, but firstly I am going to start with a few real life examples of what players would and would not suit a Money-ball routine of signing players.

For the purpose of these examples I am going to use strikers as I feel they have much easier understanding of what the statistics mean and much less of a case for an argument of what they are designated to do on the pitch. I have then gone onto choose three players and explain how the statistics or Sabermetrics of these players are used to negate whether this player would make a viable transfer option under Money-ball principles. Now I chose these positions only because it will be easier to present the analysis of their statistics and I know they all play roughly the same role and position, admittedly I wanted to showcase midfielders but as there is many roles in the midfielder, I went with wingers. Baring in mind this plays no significant part other than just displaying a few real life examples and how we can implement this method and transfer strategy into the game. (For the purpose of this I will be using the website WhoScored, Squawka and TransferMarkt for the values of the said players).

First up we will take a look at the strikers. A less concerned player about statistics may just look at the amount of goals scored compared to the games and come to conclusion whether the player has played well or not, now that isn’t a bad way as after all a strikers responsibility is to put the ball in the back of the net. However, by playing Money-ball you need to take a deeper look into the players stats, for example, a player may have played 21 games, scored 21 goals. Looks great on paper, but what if 16 of them were penalties? What if 16 goals came in 3 games? And the other 5 over 16 games? But we must also take into account what he does other than score, so here is a list of the metrics that will be involved in the process of buying players.

Non-Penalty goals - For reason I said above, hardly paints a good picture for open play contribution if the only time a player in question can score is from the spot.

Shots – Well we want them in and around the box getting shots off

Shooting% - Are is shots on target? Again this ties in with the metric above, we don’t want a wasteful striker who needs dozens of attempts to score a goal.

Passing% - Is there passing of an acceptable standard?

Assists – Are they contributing to team play rather than just playing a selfless game?

Key passes – Same as above really, you don’t want a one man team player, we need a team game with every player contributing to some extent.

Int+tackles – Just because he is a striker doesn’t mean he can away without putting his foot in, strikers are the first line of defence.

Successful Dribbles – We need players who are adequate on the ball, we don’t want him turning at goal and just letting the ball run away from under is control.

Average rating – A striker will get a good rating when he scores, but what the games when he doesn’t? Is he still contributing enough to be deemed having a good game? More for a football manager point of view if I’m being honest.

Goal Conversion% - How many of his shots are actually resulting in a goal? Like I said above we don’t want a wasteful striker.

1408789650127.jpg

Strikers are typically only judged on there goal scoring feats, Money-ball is different.

Right now we have the Sabermetrics our players are going to be judged on, are there any other metrics that our players are going to be judged on? Well yes, if you read the rules above there are quite above these include,

Nationality – In the rules and the book it outlines that certain nationalities are over-priced.

Value – Remember Money-ball is getting value for money and also ties in with age.

Contract – Another useful piece of information would be when the player’s contract is up, a player on a longer contract will be more expensive and vice versa for a player on a smaller contract will be cheaper. Wage may also come into this on football manager.

Age – He has to be at the right age, too negate the worry of not developing and offer resale value.

Type of player – This will refer more to football manager than for the purpose of these examples.

Club and League – Gives a good idea of what level the player is playing at.

The first player up for ridicule is a striker from Blackburn,

Name: Jordan Rhodes

Value: £8.8M

Contract: June 2019

Age: 25

Nationality: Scottish

Club and League: Blackburn Rovers, Championship, England

Non-Penalty goals - 0.53

Shots – 3.3

Shooting% - 45

Passing% - 73

Assists – 4

Key passes – 1.05

Int+tackles – 1.01

Successful Dribbles – 0.1

Goal Conversion% - 16

Now we have all the pitch statistics we need, but now what do we do with it? It’s hardly presentable, it isn’t really in a format where we can easily draw comparisons with other players, this is where I owe a great thanks and credit to fmcoffeehouse.wordpress.com and for Ed who writes on this blog for him allowing me to use his radars. There perfect to present data and make it easily comparable. Here is the template for the base radar used for strikers and attacking midfield strata.

Radar_Example.png

Radar used for strikers.

As you can see it basically represents a spider diagram, and on the left a input option where I can simply input data and it then allows we to draw a diagram based on the minimum and maximum values outlined. Below is the radar/diagram drawn using the example statistics we gathered on Jordan Rhodes above,

Name: Jordan Rhodes

Value: £8.8M

Contract: June 2019

Age: 25

Nationality: Scottish

Club and League: Blackburn Rovers, Championship, England

Non-Penalty goals - 0.53

Shots – 3.3

Shooting% - 45

Passing% - 73

Assists – 0.11

Key passes – 1.05

Int+tackles – 1.01

Successful Dribbles – 0.2

Goal Conversion% - 16

Rhodes_Radar.png

Jordan Rhodes radar.

Firstly lets look at his personal biography, you can see Rhodes is at a great age to offer resale value, albeit his value being a tad high and his contract having plenty to run. Going by our Money-ball principles we can also gather he is from an over-rated nation playing in the UK and being Scottish.

However his stats look much more presentable than a bunch of numbers listen next to titles, it is presented in a much clearer fashion and more able draw comparisons and come to conclusions. So, what can we see from the radar created on Jordan Rhodes, it is evident he has having great season in front of goal, with him scoring over 0.5 goals every 90 minutes, which doesn’t include penalties. We can see he isn’t shy from getting shots away when the opportunity arises, furthermore his accuracy and goal conversion rate isn’t too shabby either and backups the visual representation of his form in goal.

The visual representation of the statistics for Jordan Rhodes also show some glaring downfalls in Jordan Rhodes’ game notably the shocking representation of his dribbling, it clearly shows that isn’t the strongest part of his game and he’s much more of penalty box player. It is also clear to see, touching on what I said about defending from the front, that he isn’t the type of player who would be strong in making challenges and intercepting the ball as it’s played from defence. Where his passing seems quite decent and he averages a decent amount of key passes per 90 minutes, he struggles to create assists for his team mates.

Time for player two,

Name: Carlos Bacca

Value: £17M

Contract: June 2018

Age: 28

Nationality: Colombia

Club and League: Sevilla FC, La Liga, Spain

Non-Penalty goals – 0.39

Shots – 3.1

Shooting% - 73

Passing% - 67

Assists – 0.21

Key passes – 0.8

Int+tackles – 0.6

Successful Dribbles – 0.6

Goal Conversion% - 17

Bacca_Radar.png

Carlos Bacca radar.

Again looking at his personal biography first you can see Bacca maybe a tad on the old side for the principles of Money-ball, and is value and contract put may up there for a mega bucks transfer which really isn’t the philosophy of Money-ball.

The visual representation of Carlos Bacca’s sabermetrics clearly show he has an outstanding rate of accuracy when striking the ball at goal, and with that amount of accuracy you could say he would be a tad disappointed to not score more than he has done. Now, should you take that into consideration? Well in my opinion no, I want my striker to finish, not get praise because he can hit the target, he may blame world class goal keepers? Well good, but I want him to still finish. But, there is potential there, and that would be a big green tick on his name if I were considering him for a transfer. Now again, it’s easy to see his shooting is his strength, now I know I said he may be disappointed to not score more, but you can see his conversion rate is more than adequate and which will explain his decent record of non penalty goals over 90 minutes. His assisting record isn’t alarmingly bad for a striker either, same case with his shots per 90 minutes also.

However, the disappointing aspects of his game are clear to see, his dribbling is non-existent, his forward defensive play, again non-existent. Contributing factors? Determine the role he is being asked to play at his current club, however, this really isn’t the Money-ball game. Money-ball is about how good the player is on paper for the role you want him to play.

And finally the last player of the example I have used,

Name: Nico Lopez

Value: £3.9M

Contract: June 2015

Age: 21

Nationality: Uruguay

Club and League: Hellas Verona, Serie A, Italy

Non-Penalty goals – 0.55

Shots – 1.8

Shooting% - 85

Passing% - 73

Assists – 0.13

Key passes – 0.8

Int+tackles – 1.3

Successful Dribbles – 1.6

Goal Conversion% - 19

Lopez_Radar.png

Nico Lopez radar.

And finally the personal biography of Nico Lopez, his age is at a fantastic number, as he is sure enough to offer opportunity to recoup the initial outlay and more in the future. His contract and value are also well within reach, with his contract up in 6 months, his value with decrease due to this and lastly his origin and country of play don't match any of the countries outlined as being over-valued in the book soccernomics.

Nico Lopez, a much more balanced looking radar in terms of his attacking play, his goals, shooting and Passing more than likely among the best in Europe at this current time. You can see he is marginally better dribbling and forward defensive play in his radar than the other two examples. Now im not going to go to in depth with this player, I’d much rather do a conclusion paragraph of sorts and compare the three examples and explain which player would make the more suitable Money-ball signing.

Okay, we have three very impressive players above all in their own right. So how do we decide which player is the best value for money using Money-ball principles? Well we have our rules laid out above, so let’s remind ourselves of them and ask ourselves some questions.

A scenario? What type of player are we after? Well let’s paint a picture, again I think it would be appropriate to use Newcastle as the example, I must stress this by no means has any actual relation to transfer activity in my game, its just examples to give you a overview of how the system works, there will be updates on my game in due course. So, back to the pretty picture, Cisse, the number 9 of the club, has just completed his final season, hes 28, we need to recoup the money we spent on him. Werder Bremen have offered £14m in January, due to his great first half of the season where he has scored 17 goals and assisted 4. A judgement call was made that a offer as great as this may not reoccur, now is viable opportunity to replace the player. We have done out statistical analysis above, and three players have turned up as replacements for the departing striker.

Pappis Cisse, played the number 9 role, he was an advanced forward, who lead the line main source of goals.

The first thing we look at is the statistical analysis, how did they perform? Well they all have decent scoring records, they only marginally differ from each other. They all have a great conversion record of the shots on goal and there accuracy is decent too, more so with Lopez and Bacca. However Rhodes and Lopez offered the more defensive play up front, where as Bacca and Lopez were they better passers. Bacca and Rhodes were more capable of getting shots on goal then Lopez. So what were we replacing? A goal scoring striker, then they all satisfy goal scoring sabermetrics as long as bring their own personal strengths to the fore. We must now move on with all three still possibility of a transfer.

Next age, the player needs to offer some sort of resale value, the ideal time for selling would be from 28-30. Carlos Bacca is 28, would he be the best option? Well possibly no, obviously these rules can be flexible and 30 isnt the deadline once they hit, the player is gone. However, at 28 he would be pushing the guideline and he would have to be looking to move on rather quickly into his career at the club to get the value we need to redeem. Rhodes is 25, at least 3 good years in him before again, eyes need to be cast away from the club and a buying club needs to be found to ensure in the coming years we get the maximum amount of money back into the club. Lopez, 22, young, plenty of time at the club. Plenty of time to get him playing to the best of his ability and to drive the reputation and value of the player up. So age has been concluded, what can we draw from this, Carlos Bacca is too old, Lopez and Rhodes fit this criteria. Bacca is now struck off the potential transfer list.

We are not left two Rhodes and Lopez, we now know they both are of the correct age, they both are having good seasons which a replicated by our radars and showing that the statistical analysis side of the home work backs the transfer. What next? Value and Contract. Well say you have been alerted by your scouts that one of the players, Lopez, has his contract nearing the end, with only 6 months left, whereas Rhodes still has till 2019 to run, there values differ greatly, with Rhodes valued at double the price of Lopez.

What do we know now? Well they both are marginally similar in terms of statistics, they are both are similar in terms of offering a sell on fee. However, when it comes down to the potential cost, Blackburn hold all the cards seeing as Rhodes’ value is high, higher if you count the season he is having and had in the past, his contract still has considerable amount of time to run, whereas Lopez is cheaper, his contact is up soon, so Hellas Verona will not want to risk losing him for nothing and also Lopez is younger.

The Moneyball signing would be – Nico Lopez.

Well I hope that post gives you a basic outline of how transfers work, in the next post I will be giving you the run down on how this is all applied to football manager and how I plan to track the performance of my players using statistics.

Share this post


Link to post
Share on other sites

Really interesting this. I have a Wrexham save in 2025 and thinking of starting a Roma save. Will look to supplement my approach with this.

Share this post


Link to post
Share on other sites

As with the other moneyball thread currently running, always interesting to read and can see a lot of effort put in. I have outlined before the reasons that a sabremetric approach cannot be fully applied to Football (which is not to say that in depth statistical analysis doesnt have a place - Man City have had an entire department for years doing this).

Couple of observations / questions i would make, comparing "Moneyball" with your approach

- THe view and approach to age is interesting, and in many ways somewhat different to the moneyball approach taken by Billy Beane and others who use this approach in Baseball. If the goal is to find value, you most also consider that ageing players are often massively undervalued. There is a tendency in football, and in baseball, to view certain players as "over the hill" with no real statistical basis. If you look at the pioneering season where the Oakland A's started to use this approach, they often targeted older players (David Justice being one of the big examples - An ageing player cast aside by a big team). It is somewhat down to the "transfer" differences between baseball and football - In baseball there is no re-sale value, only potential trade value which, in the case of veterans, is not much of a consideration.

- Much of the principle of moneyball is about finding value by identifying people who are good at one specific thing which contributes to winning. So take the A's again - the first focus was players with high OPB. There were a number of players who looked "poor" by traditional analysis methods, but had a high OPB (Giambi, Hatteburg). This meant they were afforable. I suppose some what of an analogy would be goals and assists for a striker. A striker who scores 20 goals and "Creates" 5 will cost you a huge amount. A striker who scores 10 and assists 15 would likely be valued much lower. Yet in terms of contribution to the team success, the outcome is the same - 25 goals (that is simplifying things - you could factor in the relative difficulty of each task if you were doing it properly). What i would be interested in here, is your view that the poor dribbling performance from Rhodes was something to mark against him. Dribbling is unlikely to be a key factor in a striker's succes, particularly an AF(a). Its like the A's accepting that speed was not an important factor for their ballclub, since power/contact/patience delivers significantly more runs than speed on the bases. So would it not make sense, in some ways, to narrow what you look at for each role?

- IN terms of specific nationalities being over-valued, how are you applying that to player nationality v League? You mention Lopez (who is rubbish by the way, i had him for a season at Sassualo :D) has a positive as he is from Uruguay. However he plays in Serie A, and has done for almost all his career (6 career appearances outside of Italy).

All said thought, lovely depth of writing and explanations. Great to see people putting such levels of thoughts in and will follow with interest :)

Share this post


Link to post
Share on other sites

I very much enjoyed reading this last night, though admittedly, I'm generally skeptical of the idea that moneyball can be applied to football in the same way it was applied to baseball. I do think there are definitely good squad building principles that you can derive from it, but there are some ideas that I think could end up causing a club to shoot itself in the foot, namely:

The first is the theory that the best way to improve a team is by identifying and replacing the weakest links, rather than by splashing out on making the best links even better.

This is definitely true in baseball because you can more easily quantify the aggregate contribution of the group of players with whom you're replacing a top player you just sold or decided not to sign. But in football, a star can have a more dramatic "magnifying" effect on the players around him. Your example of Borini (as well as the example of Lamela) is relevant here as it was a case of having a player like Totti who is able to bring the best out of a far more limited, underdeveloped talent.

Beyond that, a big difference between baseball and football is the regulation of club/franchise finances. In European football especially, the financial situation is far more cutthroat, so there's a greater risk that a moneyball policy is just a sexier way of saying you're a feeder club.

In any case, excellent thread and I'm looking forward to seeing where you go with this.

Share this post


Link to post
Share on other sites

Mosfine

Kam, a question. Why Max Shooting% = 60 and not more?

The values on the radars were gained from data distinguished over 3 seasons in top leagues, the top 5% and lowest 5% of the data were used as the values. Suprsingly though, I am thinking of lowering the amount to 55% just to give a better overall feel to the data, as before I am going to start the game properly I will review the values.

Jambo98

THe view and approach to age is interesting, and in many ways somewhat different to the moneyball approach taken by Billy Beane and others who use this approach in Baseball. If the goal is to find value, you most also consider that ageing players are often massively undervalued. There is a tendency in football, and in baseball, to view certain players as "over the hill" with no real statistical basis. If you look at the pioneering season where the Oakland A's started to use this approach, they often targeted older players (David Justice being one of the big examples - An ageing player cast aside by a big team). It is somewhat down to the "transfer" differences between baseball and football - In baseball there is no re-sale value, only potential trade value which, in the case of veterans, is not much of a consideration.

I do agree with you here surprisingly, however I do believe Moneyball does have a place in football. I think anyone would be naive to think that principles from one sport, even though it worked to great effect could be copied down to the last letter to another sport. Things need to be adapted, like you said in baseball there is no resale value (You might embarrass me here, as far as my knowledge of the how the game is placed is that a bat and ball is used, bit like rounders), so they wouldn't really worry themselves about recouping the outlay. I am also daring to say that football is a much more intense sport played over 90 minutes with constant running over that period. I think its a fair assumption that older players do lose there ability to keep up there performances they did in there peak, apart from a few special players. How many times have we seen people discuss that a old overpaid player is just sitting on the bench picking his wage up? To ensure money-ball is efficient, some thing i like to refer to as the "cycle" of players, money needs to be recouped to be reinvested in the squad or club its self. I guess the conclusion i came too, which would be different for everyone, such is the beauty of the game that everyone will amass there own opinions and ideas, is that money-ball for baseball and money-ball for football whilst keeping the same principles sabermetrics etc, will be somewhat different, the metrics each player is graded on will be different. There two different sports with different demands, i hope that makes sense?

I must also stress that i did say the rules outlined are by no means the be all end all and can be flexible if needs be, its not a case of one player hitting a age and being shipped out or not considered, it all comes down to the options available, you have 3 players? All basically the same in terms of statistics, but the only thing differentiating them is age? I think its clear to see which would be the choice, in terms of value you for money.

Much of the principle of money-ball is about finding value by identifying people who are good at one specific thing which contributes to winning. So take the A's again - the first focus was players with high OPB. There were a number of players who looked "poor" by traditional analysis methods, but had a high OPB (Giambi, Hatteburg). This meant they were affordable. I suppose some what of an analogy would be goals and assists for a striker. A striker who scores 20 goals and "Creates" 5 will cost you a huge amount. A striker who scores 10 and assists 15 would likely be valued much lower. Yet in terms of contribution to the team success, the outcome is the same - 25 goals (that is simplifying things - you could factor in the relative difficulty of each task if you were doing it properly). What i would be interested in here, is your view that the poor dribbling performance from Rhodes was something to mark against him. Dribbling is unlikely to be a key factor in a striker's success, particularly an AF(a). Its like the A's accepting that speed was not an important factor for their ballclub, since power/contact/patience delivers significantly more runs than speed on the bases. So would it not make sense, in some ways, to narrow what you look at for each role?

Without running the risk of repeating myself a tad from my last message replying to your first point, football is a different game, the principles of money-ball will need to be adapted to suit the game, i dont think you would be anywhere near successful in looking for only one key area in a players repertoire, footballers need to be more all rounded. Players on the football pitch have a lot more responsibilities to adhere too. I think it comes down to the level of football being played. In the football a striker maybe forced wide, have to turn run at a deep defence etc, i guess yes it does come down to your interpretation of the role being played, again the beauty of football as it draws so many different opinions. I guess the lower the level of football you play, you could then start to narrow your search for certain metrics which suit a more limited role in the team. But yes, you are right that it all comes down to the role the player is asked to perform within the team, i was just using the above as an example of how transfers are mulled over and what the radars have the potential to show and aid in choosing a player. Hope that makes sense, it should after my next post when i explain my transfer process in game.

I do agree with your point about the 25 goals example, but in the language of money-ball, 20 goals and 5 assists is 25 goal contributions. 10 goals and 15 assists is again 25 goal contributions. From a sabermetrics point of view, 25 goal contribitions is 25 goal contributions and id expect they would cost the same, but the whole world of football doesn't operate in the way money-ball does, so the prices way differentiate, i hope that makes sense as it did my head :p

IN terms of specific nationalities being over-valued, how are you applying that to player nationality v League? You mention Lopez (who is rubbish by the way, i had him for a season at Sassualo ) has a positive as he is from Uruguay. However he plays in Serie A, and has done for almost all his career (6 career appearances outside of Italy).

Im sorry i should have stated that those positive and negative remarks should be taken with a pinch of salt as they are by no means the be all end all on my opinions, i was just trying to take examples, using real life data from this season.

But in reference to your question, again i think this is a flexible rule which will be bent and twisted in so many ways. For example, English players in England cost a premium to buy. Dutch players from holland are marginally cheaper, but once they move to a more competitive country they cost a lot more, i guess you could call it exposure. A great example would be Arjen Robben, he moved to chelsea after a great season in holland for £15m, now i dont think what he did at chelsea in terms of statistics was too much different to what he did in holland, but because of the exposure the premier league his value soared to almost double in his short time in england. Something which is a common trend around the more fashionable countries mentioned in the opening post. But again, it will be variable on the metrics much like many of the money-ball principles, as there a lot of contributing factors within football compared to other sports.

The Hand of God

I very much enjoyed reading this last night, though admittedly, I'm generally skeptical of the idea that moneyball can be applied to football in the same way it was applied to baseball. I do think there are definitely good squad building principles that you can derive from it, but there are some ideas that I think could end up causing a club to shoot itself in the foot, namely:

The first is the theory that the best way to improve a team is by identifying and replacing the weakest links, rather than by splashing out on making the best links even better.

This is definitely true in baseball because you can more easily quantify the aggregate contribution of the group of players with whom you're replacing a top player you just sold or decided not to sign. But in football, a star can have a more dramatic "magnifying" effect on the players around him. Your example of Borini (as well as the example of Lamela) is relevant here as it was a case of having a player like Totti who is able to bring the best out of a far more limited, underdeveloped talent.

Beyond that, a big difference between baseball and football is the regulation of club/franchise finances. In European football especially, the financial situation is far more cutthroat, so there's a greater risk that a moneyball policy is just a sexier way of saying you're a feeder club.

In any case, excellent thread and I'm looking forward to seeing where you go with this.

This some that is really hard to differentiate from on football manager, as i do agree with your point that players like Totti are equally as effective for there influence on players aswell as there efforts on the pitch. Metrics which is cover this in the game could range from many things, there relationship to the club? Are they a long time servant and on the icon, favoured personal list? Are they club captain? Loved by the players? All of the latter would effect the club morale if they were transferred out and equally effect the players if there playing with favoured personel or there icons. Same could be said for Gerrard at liverpool, who dragged certain players through games, a few finals spring to mind.

I would disagree a tad with your statement of being a feeder club, as technically my aim is to get the best out of the players, move them on for a price that recoups some of the outlay on the player and get someone of equal ability and able to perform to the statistics of the outward bound player. If its implemented right, the club in question should see a constant cycle of players coming in of great ability and outward ones whos career had there best spent years at our club, whilst the finances of the club are greatly cared for within the method imposed.

Your first point though about the effects of players has given me somehitng else to consider. Cheers.

Thanks for the comments guys, always nice to discuss topics and listen to other views and contrasting opinions, always gives you soemthing to think about. :)

Share this post


Link to post
Share on other sites
I very much enjoyed reading this last night, though admittedly, I'm generally skeptical of the idea that moneyball can be applied to football in the same way it was applied to baseball. I do think there are definitely good squad building principles that you can derive from it, but there are some ideas that I think could end up causing a club to shoot itself in the foot, namely:
The first is the theory that the best way to improve a team is by identifying and replacing the weakest links, rather than by splashing out on making the best links even better.

This is definitely true in baseball because you can more easily quantify the aggregate contribution of the group of players with whom you're replacing a top player you just sold or decided not to sign. But in football, a star can have a more dramatic "magnifying" effect on the players around him. Your example of Borini (as well as the example of Lamela) is relevant here as it was a case of having a player like Totti who is able to bring the best out of a far more limited, underdeveloped talent.

Beyond that, a big difference between baseball and football is the regulation of club/franchise finances. In European football especially, the financial situation is far more cutthroat, so there's a greater risk that a moneyball policy is just a sexier way of saying you're a feeder club.

In any case, excellent thread and I'm looking forward to seeing where you go with this.

I agree with this in general. Baseball is a sport where a player's statistics are derived mostly from his own ability and tell us so much about him. Football statistics are extremely context-dependent. Look at Jozy Altidore's production at Alkmaar vs. Sunderland, for example.

Somewhere in between is another of my favorite sports, basketball, and I think there's a lot to be learned from how NBA teams have attempted to apply Moneyball principles to a game that's a little bit more like football. One of my favorite articles ever talks about how advanced stats led the Houston Rockets to sign Shane Battier - a player who contributes very little in terms of traditional statistics but has that "magnifying effect" on his teammates:

Battier’s game is a weird combination of obvious weaknesses and nearly invisible strengths. When he is on the court, his teammates get better, often a lot better, and his opponents get worse — often a lot worse. He may not grab huge numbers of rebounds, but he has an uncanny ability to improve his teammates’ rebounding. He doesn’t shoot much, but when he does, he takes only the most efficient shots. He also has a knack for getting the ball to teammates who are in a position to do the same, and he commits few turnovers. On defense, although he routinely guards the N.B.A.’s most prolific scorers, he significantly reduces their shooting percentages. At the same time he somehow improves the defensive efficiency of his teammates — probably, Morey surmises, by helping them out in all sorts of subtle ways. “I call him Lego,” Morey says. “When he’s on the court, all the pieces start to fit together. And everything that leads to winning that you can get to through intellect instead of innate ability, Shane excels in. I’ll bet he’s in the hundredth percentile of every category.”

To me, that should be the biggest goal of a moneyball type of approach - finding the players who make the whole unit better, whether it's a hard-working defensive midfielder like Makelele (essentially the football equivalent of Battier) or a brilliant playmaker like Totti. How to do that? I'm not exactly sure.

Anyway, good thread and I'm also interested to see where it goes.

Share this post


Link to post
Share on other sites
I agree with this in general. Baseball is a sport where a player's statistics are derived mostly from his own ability and tell us so much about him. Football statistics are extremely context-dependent. Look at Jozy Altidore's production at Alkmaar vs. Sunderland, for example.

I dont think that is entirely accurate, but i accept that it probably depends on a bit of an understanding of baseball in depth. A baseball players statistics can be extremely context dependent. Some of the things which frequently impact a baseball players stats above and beyond his own ability:

- The type of pitching he faces (left v right). Large platoon splits are common in baseball (eg, players generally hit better against opposite handed pitching)

- Defensive shifts can be huge. A players effectiveness can be limited by a the defensive shift. I suppose the equivalent might be double man marking.

- The ballpark. The sizes and shapes of baseball fields vary, and it will massively impact a players stats. Moving from San Diego to Yankee Stadium, or from Kansas City to Texas, will have a huge impact on the stats

- The league. This is a bit complex, but in essence for pitchers the American League is generally much harder than the National League, due to the different rules

- The weather / atmosphere. The "coors field effect" is quite well known. In essence Colarado play at much higher altitude than other teams, meaning the air is thinner and the ball travels further. Likewise in Chicago its often windy. In Toronto and Tampa they play in a dome so there is never a weather factor.

Now i realise that is all a bit boring and baseball related, nothing to do with Football, Football Manager or indeed people might ask whats it to do with this thread. The context, for me, is that whilst Moneyball is often admired by those outside of Baseball and outside of the US, at times, with the greatest of respect, it is not properly understood. That can lead to it being difficult to translate.

To me, that should be the biggest goal of a moneyball type of approach - finding the players who make the whole unit better, whether it's a hard-working defensive midfielder like Makelele (essentially the football equivalent of Battier) or a brilliant playmaker like Totti. How to do that? I'm not exactly sure.

Anyway, good thread and I'm also interested to see where it goes.

Is that not the goal of any approach, at any team? I wouldnt characterize that as being the "moneyball" approach. Making the team better is a common approach in most (not all) teams. Other than at Real where you sign on individual talent and star power, clubs look at players in the context of how they improve the team and squad already.

Rather moneyball was about using non traditional methods to judge player performance, and in doing so, crucially, find hidden value. Billy Beane did not choose to go the moneyball route just because he fancied it. It was forced on him by being a very small market team trying to compete with large market teams. A traditional approach would mean they could never find players good enough to out perform their payroll status, so Beane used sabremetrics to identify players ignored by other teams because they did not perform well in "traditional" stats or didnt "look good". In football terms it was finding the free transfer keeper discarded because he was too short. Or the unwanted reserve striker who doesnt get a game because he never scores. The challenge comes in that football there are only 2 ways to succeed - Score goals, and prevent them. If a striker scores 5 a season and assists 5, in football terms, it is unlikely you can find a non traditional stat argument for why he should be good for your team.

The biggest irony of "Moneyball", is that it didnt actually suceed. The A's never made the World Series, let alone winning it. They only made the playoffs 6 times in the 12 years since 2002, and only once made it past the first round. People talk about the Red Sox, but they didnt play Moneyball. They embraced Sabremetrics (they hired Bill James afterall) but when they won in 2004 they had the second highest payroll in all of baseball.

Anyway, i do think its important to really understand moneyball to then see how much it translates to football.

All that said (god i sound like a negative ****!!), i really like the thread and the debate :)

Share this post


Link to post
Share on other sites

Is that not the goal of any approach, at any team? I wouldnt characterize that as being the "moneyball" approach. Making the team better is a common approach in most (not all) teams. Other than at Real where you sign on individual talent and star power, clubs look at players in the context of how they improve the team and squad already.

Rather moneyball was about using non traditional methods to judge player performance, and in doing so, crucially, find hidden value. Billy Beane did not choose to go the moneyball route just because he fancied it. It was forced on him by being a very small market team trying to compete with large market teams. A traditional approach would mean they could never find players good enough to out perform their payroll status, so Beane used sabremetrics to identify players ignored by other teams because they did not perform well in "traditional" stats or didnt "look good". In football terms it was finding the free transfer keeper discarded because he was too short. Or the unwanted reserve striker who doesnt get a game because he never scores. The challenge comes in that football there are only 2 ways to succeed - Score goals, and prevent them. If a striker scores 5 a season and assists 5, in football terms, it is unlikely you can find a non traditional stat argument for why he should be good for your team.

I think you're missing part of Sparty's point, especially re: Shane Battier. There are players that contribute to a team's performance in less than obvious ways -- a "Moneyball" approach would identify the statistics that best represent those less obvious contributions. For football, maybe that's a focus on bullet throws (Stoke) or corners (Atletico). In reality you can work to exploit the holes in conventional thinking (RBIs/HRs vs OBP), although I think FM statistics aren't really in that realm yet.

In your striker example, maybe the player feeds the player making the assist. Maybe he has a high "effective" shooting percentage. Maybe he never scores a goal but his movement and defensive harrying mean his +/- is off the charts. I'm converting statistics from other sports here, but you could tailor football specfic stats. For example, someone had posted that most goals take less than 8 seconds (I could be mis-remembering the actual number) from winning possession -- maybe you value players who have the ball out of their feet and moving forward in under 2 touches.

The biggest irony of "Moneyball", is that it didnt actually suceed. The A's never made the World Series, let alone winning it. They only made the playoffs 6 times in the 12 years since 2002, and only once made it past the first round. People talk about the Red Sox, but they didnt play Moneyball. They embraced Sabremetrics (they hired Bill James afterall) but when they won in 2004 they had the second highest payroll in all of baseball.

It really depends how you define success. I think they outperformed their budget by a significant margin, which I'd count as a win. The funny thing is that any edge from "Moneyball" tactics vanishes as soon as assets become properly valued. You have to keep finding new statistical edges to even churn water.

Also, I really enjoyed the discussion so far and thanks to the OP for his thoughts that kicked it off.

Share this post


Link to post
Share on other sites
I dont think that is entirely accurate, but i accept that it probably depends on a bit of an understanding of baseball in depth. A baseball players statistics can be extremely context dependent. Some of the things which frequently impact a baseball players stats above and beyond his own ability:

- The type of pitching he faces (left v right). Large platoon splits are common in baseball (eg, players generally hit better against opposite handed pitching)

- Defensive shifts can be huge. A players effectiveness can be limited by a the defensive shift. I suppose the equivalent might be double man marking.

- The ballpark. The sizes and shapes of baseball fields vary, and it will massively impact a players stats. Moving from San Diego to Yankee Stadium, or from Kansas City to Texas, will have a huge impact on the stats

- The league. This is a bit complex, but in essence for pitchers the American League is generally much harder than the National League, due to the different rules

- The weather / atmosphere. The "coors field effect" is quite well known. In essence Colarado play at much higher altitude than other teams, meaning the air is thinner and the ball travels further. Likewise in Chicago its often windy. In Toronto and Tampa they play in a dome so there is never a weather factor.

Now i realise that is all a bit boring and baseball related, nothing to do with Football, Football Manager or indeed people might ask whats it to do with this thread. The context, for me, is that whilst Moneyball is often admired by those outside of Baseball and outside of the US, at times, with the greatest of respect, it is not properly understood. That can lead to it being difficult to translate.

I agree, baseball stats can be context-dependent too. I guess what I was getting at is that it's much easier to put things in context in baseball. You can divide a player's stats into home/road, day/night, lefty/righty, flyballs vs. groundballs, etc. and get a pretty good idea of what he might do on a different team in a different park in a different league. It's very hard to do something similar in football.

Is that not the goal of any approach, at any team? I wouldnt characterize that as being the "moneyball" approach. Making the team better is a common approach in most (not all) teams. Other than at Real where you sign on individual talent and star power, clubs look at players in the context of how they improve the team and squad already.

That's true. The hard part is quantifying it all, though. I see stuff in baseball articles all the time like "Signing Scherzer changes the Nationals' expected wins this season from 88.4 to 94.2" or whatever. What I'd love to figure out is how to get as close to that concept as possible in football. We can watch a match and know that a great DM has a huge impact on a team, but what's that it really worth in terms of goals prevented, possession gained, and ultimately results? And how does it compare to the impact of a great striker? Because to me, before you can decide if a player's overvalued or undervalued, you have to be able to put some kind of number on his value. Then you can compare that to the financial cost and make Moneyball-type decisions.

Share this post


Link to post
Share on other sites

Well in my eyes that totally lies down to the fact what system you are playing. In mourinho terms, gelid system makes it necessary to have a "destroyer" type of midfielder in his DM strata. So surely he would be able to signify to himself what worth he would put on a player in that position. He would know what he expects of a player in that position, breaking up play and laying it off to a more creative player in the team. So with that in mind, interceptions? Tackles attempted? Tackles successful? Interceptions? Duals won? Aerial duals maybe? Passes over a short distance? Etc... Looking at mourinho's system and how his system is perceived to rely on that type of player to structure the defensive aspects of his game, surely that guy would be first player in his mind when choosing his team or if he can improve that position in the transfer market of statistics show it as being weaker area than he expects more from.

I think in football carrying moneyball over in my mind is a much complex motion, seeing as footballers have various roles to supplement on the pitch, I.e defensive or attacking roles, transitions. And in them roles there more roles such as like mentioned above for the "makele role" in mourinhos systems or the carzola role in arsenal, even how important Henderson has become in how Liverpool play.

If I stay with the team I'm using and use a real example of newcastle play in real life, I think it's a fair assumption that sissoko is vital to newcastle. But his role is paramount and has many distinguishing features, he tackles, runs with the ball, scores, passes, creates etc etc... So typically a box to box midfielder yeah? A box to box midfielder must be decent at both areas of the pitch, so metrics for this would be more complex as a suppose to the way route baseball used in moneyball and picking out a individual statistic and basing there signing on that, baseball is a completely different game, the model needs to be flexible to fit certain roles. One shoe doesn't fit all as the saying goes, I think moneyball is much more complex than being plug and play method. In football I don't think it's possible to pick out certain metrics to see if a player is viable for transfer under moneyball, your selling yourself short in getting the bigger picture as a manager. Every player has his role and a support role, Diego Costa is a striker, strikers score goals, his support role for the team would be bringing others into play. He is also the first line of defence up front, with all that In mind would you just concentrate on the metrics in regards to his scoring feats? Course not, you'd want to see his all round play. In terms of what was discussed earlier, you would need to know what would his effect be on the team, is a leader, shy, a role model etc... You need to balance the team, as 11 leaders are likely to clash on field.

Sorry I did plan on more in depth reply here but I'm on my phone, I shall reply more in full when I get on my laptop. Great discussion though guys I'm enjoying it.

Share this post


Link to post
Share on other sites
To me, that should be the biggest goal of a moneyball type of approach - finding the players who make the whole unit better, whether it's a hard-working defensive midfielder like Makelele (essentially the football equivalent of Battier) or a brilliant playmaker like Totti. How to do that? I'm not exactly sure

One of the more interesting cases of a pre-moneyball moneyball signing with that "magnifying effect" was Terry Pendleton signing for the Braves in '91. He had all the looks of a money ball player (stout, 5'9", no star appeal, etc.) and none of the stats, but he brought a cool, experienced head to a young, unsure team loaded with rising talent and then ended up MVP that season. In football, that psychological component is that much more important, so it's something that would be at the top of my list of qualities to look for, especially if you're filling out a squad already loaded with inconsistent 20 year-olds.

Share this post


Link to post
Share on other sites
One of the more interesting cases of a pre-moneyball moneyball signing with that "magnifying effect" was Terry Pendleton signing for the Braves in '91. He had all the looks of a money ball player (stout, 5'9", no star appeal, etc.) and none of the stats, but he brought a cool, experienced head to a young, unsure team loaded with rising talent and then ended up MVP that season. In football, that psychological component is that much more important, so it's something that would be at the top of my list of qualities to look for, especially if you're filling out a squad already loaded with inconsistent 20 year-olds.

I echo this, I think a vital part is like I mentioned above, your effectively building 2 teams. A playing team and personality team? You want players to play well on the pitch, so you need talent, you need the numbers to add up to have fluid team capable of fulfilling what you ask them to do on the pitch. Secondly you need a balanced team in terms of personality, 11 leaders would clash. 11 shy players would get steam rollered with no supportive voice on field. You need a inspirational figure such as Totti, something which metrics on the game maybe hard to impose, wel soon see.

Share this post


Link to post
Share on other sites

As an American who knows baseball much better than i know "soccer", i would say that you cannot really qauntify player ability via stats like you can in baseball. In baseball, there is some coaching, and the manager needs some ability to set lineups and such but the reality of it is that you will have 9 guys go bat in the same order and field in pretty much the same spot no matter what team they are on and who their manager is. In football, the playing style and formations have such an effect on what a given players stats will look like. I guess i have a hard time thinking about how they could ever have a WS/WAR like statistic in professional football. Although i see using statistics like Brenham does to determine what kind of strategy/philosophies to use to win games to be exactly what you would use advanced stats for. Thats why Billy Beane caring more about on-base-percentage, walks and not allowing his players to steal bases, is why his strategy worked. If you want to truly learn whY "Soccer" can never be analyzed like Baseball, go look on baseball-reference.com and any player and look at all the stats and how complicated a lot of the stats are to determine. Also Front Offices like the A's go even further and are now recording bat speed and how hard the ball is is hit and where it lands. Teams like the Devil Rays use defensive shifts hundreds of times a season based on where players typically hit a ball. Just the amount of data that can be recorded in baseball is unmatched.

But with all that being said, i am a huge stat-head and love the idea of this post. I would say in FM the fact you have a 1-20 rating for each attribute is used by most in a money ball like way to determine what players to buy, and thats why i was drawn to this game despite initially having zero knowledge of the sport.

Share this post


Link to post
Share on other sites
As an American who knows baseball much better than i know "soccer", i would say that you cannot really qauntify player ability via stats like you can in baseball.

You're absolutely right that baseball is one of the easiest sports to quantify because it occurs in mostly discrete steps and those events are quasi-independent (relative to play in other sports). Like I mentioned above, we're not yet living in the age of effective football statistics but you can see where the future is headed -- basketball. Cameras around NBA arenas are able to pinpoint each player's location on the court as he shoots/dribbles/rebounds/etc. This in turn leads to things like shooting percentage statistics which adjust for the difficulty of a shot from each point on the floor or the proximity of the nearest defender. It's still a nascent field that does still have some holes (Bruce Bowen posted terrible PERs because his tenacious defensive contributions weren't captured in steals or blocks); we're not far from seeing the same burgeoning revolution in football.

Share this post


Link to post
Share on other sites

Players in basketball take hundreds of shots a season and you can quantify defense in basketball to a degree, they even have stats determining offensive and defensive win shares that are pretty good indicators of ability. PER is a really good marker in basketball, something i don't think that soccer can ever get close to. I think that you could compare striker contributions but determining a defenders value seems almost impossible. I suppose turnover ratio would be something that can translate in soccer as long as you factor in pace and what not. I should add i am very interested to see how this play through goes.

Share this post


Link to post
Share on other sites

I think moneyball is cool etc. but don't understand trying to apply it to FM. On FM you have individual stats for players, eg their 1-20 stats, why would you delve into their game stats in much detail?

Share this post


Link to post
Share on other sites
I think moneyball is cool etc. but don't understand trying to apply it to FM. On FM you have individual stats for players, eg their 1-20 stats, why would you delve into their game stats in much detail?

Beautiful piece of irony that :) You likely dont realise it, but you just explained Moneyball pretty well....The entire need for the moneyball approach in MLB is because if you you just ranked players based on the type of info that everyone else does, then you will never ever find good enough players for your low budget teams.

A lot of people miss the point of moneyball - And that is not a critiscm of anyone, but people watch a Brad Pitt film and try and apply it based on Wiki research. I would urge anyone who wants to understand it to properly read the Michael Lewis book. There is a huge amount more to the approach than the film shows, and there is a huge amount more to it than just using advanced statistics.

Its all about finding value where others cant. It is far harder to apply to football for many reasons, but the core idea of finding value where others dont translates. Ironically for this thread, you could argue that Mike Ashely did it. He looked at the french league and saw value. He saw that you could take a player in France, who if he was playing in England, Italy or Spain, would cost £15million and get that same player for £5million.

It is just much harder to do in Football. There is no equivalent to Chad Bradford - There arent really players judged on "how they look" in football terms.

Share this post


Link to post
Share on other sites

I have read the book, but I don't think you're grasping what I'm saying. Why are you looking at a whole bunch of in game stats when we can decipher how a player will play from his ratings screen, plus ppms, etc. unless we try to decipher his hidden stats, which we know there are better alternatives to do.

in life we don't have these ratings, so statistics can help us make sense of this.

Share this post


Link to post
Share on other sites
I have read the book, but I don't think you're grasping what I'm saying. Why are you looking at a whole bunch of in game stats when we can decipher how a player will play from his ratings screen, plus ppms, etc. unless we try to decipher his hidden stats, which we know there are better alternatives to do.

in life we don't have these ratings, so statistics can help us make sense of this.

Except thats not true at all?

Ratings = talent.

Talent does not always = performances

Also in life you do have the equivalent of these ratings. Every baseball player has a 20 - 80 rating in about 10 different categories.

The 2 things are not connected as such, your missing the point of moneyball. If the Oakland A's knew the exact rating of every single player available to them, they would still have had to apply moneyball to find those who are undervalued. Because those with the equivalent of an FM "20" rating are not going to be available to the smallest market teams.

Share this post


Link to post
Share on other sites

I don't think sabremetrics have a place anywhere much less soccer. The A's clearly blew their load last year when they used such a method. They broke that team in half in the trading deadline.

Share this post


Link to post
Share on other sites
I don't think sabremetrics have a place anywhere much less soccer. The A's clearly blew their load last year when they used such a method. They broke that team in half in the trading deadline.

Every team in baseball uses sabermetrics to some extent. The A's don't have a monopoly on them just because of a movie. In fact the team the A's made probably their worst trade with, the Cubs, are run by Theo Epstein, a well-known numbers guy himself.

A couple bad moves by one team is a stupid reason to throw out the whole concept.

Share this post


Link to post
Share on other sites

I've used a sort of money-ball approach in my current save. Like the OP, I started by using stats to rate each player. I couldn't find a way to make it work, though. You want to be able to use the rating to compare players on different teams, in different positions/roles/duties, different formations, different leagues, different levels, etc. Perhaps it is possible to control for all those variables, but I couldn't even start to figure out how.

So I came to the same conclusion as sebs: use attributes, not stats, to rate players. I use a weighted average of attributes, with different weights for each position/role/duty that I use in my formation. It would be hard to come up with the "perfect" set of weights, but a "good enough" set of weights is easy enough to arrive at.

Once you have the rating system in place, you can definitely find good players who are undervalued by the game. With a little help from the "Print to web page" option, you can export masses of player data, and use the rating to hone in on the best players in your price/wage range. I've used this approach for 5 or 6 seasons now, with decent results.

Interestingly, while the stats-based approach will lead you to players coming off a good season, the attributes-based approach will lead you to players who've had a mediocre season (since their price will be depressed). That feels like money-ball, at least as I understand it.

Mike

Share this post


Link to post
Share on other sites
Every team in baseball uses sabermetrics to some extent. The A's don't have a monopoly on them just because of a movie. In fact the team the A's made probably their worst trade with, the Cubs, are run by Theo Epstein, a well-known numbers guy himself. A couple bad moves by one team is a stupid reason to throw out the whole concept.
A couple of bad moves?? Their World Series opportunity is on the line with those moves. I was also using teams like the As as examples. I'm not ignorant in thinking that they are the only ones who buy into this philosophy. Oh, no. I've been watching baseball since 2000, I know better than that :o

Share this post


Link to post
Share on other sites
Except thats not true at all?

Ratings = talent.

Talent does not always = performances

Also in life you do have the equivalent of these ratings. Every baseball player has a 20 - 80 rating in about 10 different categories.

The 2 things are not connected as such, your missing the point of moneyball. If the Oakland A's knew the exact rating of every single player available to them, they would still have had to apply moneyball to find those who are undervalued. Because those with the equivalent of an FM "20" rating are not going to be available to the smallest market teams.

no, it is true. In FM we know everything about a player except the hidden mentals, and even then there are ways of predicting them. they don't have some kind of nefarious random ranking which dictates their performance.

I agree that it's about finding value in a competitive market, but I don't see how looking at random statistics like those mentioned in this thread, such as non penalty goals and successful dribbles are going to help matters, which has been my point since I first posted and you still haven't addressed that point.

players might have some 20-80 ranking, but the point of moneyball is that they thought they were looking at the wrong things. well in FM we can choose to look at whatever we want to!

Share this post


Link to post
Share on other sites
no, it is true. In FM we know everything about a player except the hidden mentals, and even then there are ways of predicting them. they don't have some kind of nefarious random ranking which dictates their performance.

I agree that it's about finding value in a competitive market, but I don't see how looking at random statistics like those mentioned in this thread, such as non penalty goals and successful dribbles are going to help matters, which has been my point since I first posted and you still haven't addressed that point.

players might have some 20-80 ranking, but the point of moneyball is that they thought they were looking at the wrong things. well in FM we can choose to look at whatever we want to!

Ok let me make the point another way. As an experiment, try this:

Look at a full season of stats and note down the top 20 players in the following categories:

- Goals

- Pass completion

- Interceptions

Then note down the top 20 players in the league based on the following attributes:

- Finishing

- Passing

- Tacking

Now compare the two lists...... They wont match. If they did, the game would be pointless. Knowing that a player has "20" attribute in "crossing" does not necessarily mean he is the best winger in the world.

You could argue that all that is needed is to truley understand all of the attributes which impact each action. For example, above i was being a little facitious......we all know that "finishing" alone wont get you lots of goals, but do we know the exact type of attributes which do? Is it composure, off the ball, anticipation, bravery? Its some combination of those (plus likely some hidden stats). You would need to be very smart to understand the absolute perfect mix of attributes which makes the best goalscorers.

So the alternate approach, to finding value at least, might be to look at stats. We can all point to that time in our FM career when the striker who didnt look like sh*t on paper suddenly scored 40 a season for us. The reason for this is likely that we just didnt see the beauty in his attribute balance - Maybe he had 8/10 for finishing and composure, but had 15+ in anticipation, work rate, off the ball........

There was actually a far better moneyball thread i think last year, where a guy did this. He looked at the top 5 cross completion players acros the best 5 leagues in the world, over 3 years, and tried to work out what attributes thost 75 players (less actually since the same players to some extent topped the list each year) had in common. From there you could apply that template to other players who were likely to become great "crossers".

The point of moneyball is not that they were looking at the wrong things - As a poster above points out, the A's were hardly the only ones looking at Sabremetrics. The point was the A's found ways within this approach to find true value. Chad Bradford is the great example - His stats were always solid, his scouting report would have said he had good stuff and great movement. As crazy as it sounds, teams did back away because he looked "funny". Part of the reason i actually say you cant replicate moneyball fully in footbal, is that situations like that really dont exist in football.

Share this post


Link to post
Share on other sites

I've always found that the players who seem to inexplicibly score tons of goals despite rubbish stats seem to get their value from being in a team that fits them perfectly. Its hard to transplant that same player to another team that plays in a different style.

Share this post


Link to post
Share on other sites

I'll reply tomorrow to this thread, I have got updates to post, I've just been so busy at the moment with my girlfriends 21st birthday and univeristy, but great discussion.

One thing I'd like to add quickly is, I think maybe some people have grasped the idea of this game in the wrong way slightly, particularly people saying signing players on attributes. Yes attrivutes play a certain part defining the type of player required, but the bulk of the decision is made upon sabermetrics/statistics, after all that is the principles of moneyball. Signing a player who has had a great season under AI control, with statistics across the aeason that accompany this, but perhaps there attrivutes don't match? Can I get the same success from the player? Will it help me sustain a level of accomplishment in game whilst playing the moneyball way? I can easily go out and build a great team signing the best rated players by attrivutes, but can I do it using moneyball and picking players based on merit and statisitixs and parameters in the first post? Or will it hamper my success, it's more of an experiment really :)

Share this post


Link to post
Share on other sites

To be fair, i dont think anyone has grasped it wrongly :) People are simply questioning the wisdom of the approach - Why use advanced statistics (a slight aside, but the use of the word "sabermetrics" in terms of football is wrong. The word "sabermetrics" is derived from the Society for American Baseball Research) when attributes can show you a players ability better?

It is not disimilar to the many argument used against Billy Beane - Why use this approach when baseball scouts with 100 years combined experience can tell you better. Obviously the huge difference is that baseball scouting ratings could be subjective, whilst player attributes in FM are absolute. That is why you get the challenge from people as to what you can gain here.

I think its has been pointed out above, but actually moneyball was not at all purely about using Sabermetrics. That was only one part of it. In many ways, it would make more sense to look carefully at the brentford model and the Man City model (Look up Gavin Fleig - He is influenced by Bill James but applies it to football in an effective way).

Share this post


Link to post
Share on other sites

I think statistics have their place but they are highly context sensitive. Context can be things like quality of competition but it can also be the role that's being asked of a player. So that being the case I think an area where statistics can help perhaps even more so than attributes is in comparing players on the same team in the same role. So for example let's say you have a CM(D) whose job is to cover space and break up plays. You have one guy with a little better Positioning, Anticipation and Work Rate, versus another with better Stamina, Aggression, and Tackling. It may not be obvious from looking at attributes who will do the job better, and stats like Tackles + Interceptions can help you decide. The answer could be different even for the same role in a different tactical setup if the player is asked to do the same job but covering more (or less) space. Using statistics to understand what attributes seem to be most important in a role in your own system can in turn help guide transfer and training decisions.

Care is still required in the evaluation though because like I said, quality of competition matters a lot too. If you have prospects that you only play in really easy matches they can end up looking very good in the stat department. So I don't think anyone is going to even approach the suggestion of throwing away attributes entirely.

Share this post


Link to post
Share on other sites

I didn't mean people mis-understood literally, it was just a turn of phrase, my apologies. What I meant was, I don't want people to think I'm just trying to get rid of player attributes, as im not, I'm just exploring a real life model on the game. Statisitcs have a part to play in my game as they are how I determine the type of player I need, as you can't just speak to a scout and say "I want a hardworking midfield player who can pick out a pass", we need the attrivutes to determine the type of players. We then use moneyball to determine between players for transfers, under the guidelines stated in the first post, value, happiness, circumstances, wage, contract and statistics etc ect.

Now moneyball is contextual and everyone will get there own ideas of what it's about as I feel it is totally flexible as everyone will determine there own "rules" or "understanding" of what they will apply too football from the moneyball model in baseball. I merely used soccernomics and some reading on the Internet.

Trepanated: I'm gonna try split your post up a tad but in the hope I cover everything, quality of competition will be massively influenced as im looking for value for money, undervalued players, more obsecure leagues than the premier league will no doubt be a hunting grounds. But yeah competition is considered im going to reasonable and not expect a striker scoring 30 goals in lwague two or one to do that in the premiership, this is where the attribute templates play a part, determining the type of player, the moulding attribute wise, hope that makes sense :/

Sorry if I come across as uptight here as that's kot really the case, it's the wording is hard to get right. I have a more than decent understanding of football And the game, so like in the example above I have devised radars for each position targeting the statistics that match what I want from each player in each role. So it will be easier to distinguish what players are performing well statistical wise for a role im looking for.

Share this post


Link to post
Share on other sites
To be fair, i dont think anyone has grasped it wrongly :) People are simply questioning the wisdom of the approach - Why use advanced statistics (a slight aside, but the use of the word "sabermetrics" in terms of football is wrong. The word "sabermetrics" is derived from the Society for American Baseball Research) when attributes can show you a players ability better?

It is not disimilar to the many argument used against Billy Beane - Why use this approach when baseball scouts with 100 years combined experience can tell you better. Obviously the huge difference is that baseball scouting ratings could be subjective, whilst player attributes in FM are absolute. That is why you get the challenge from people as to what you can gain here.

I think its has been pointed out above, but actually moneyball was not at all purely about using Sabermetrics. That was only one part of it. In many ways, it would make more sense to look carefully at the brentford model and the Man City model (Look up Gavin Fleig - He is influenced by Bill James but applies it to football in an effective way).

it's not the same argument used against Beane. baseball scouts were looking at statistics that were less effective at winning games of baseball and looking at physical attributes that were less relevant.

I don't see the advantage in pouring over statistics to decide whether a certain player would be a good fit for a system, when we have it right in front of us when you look at their profile.

I think there are interesting things to take from moneyball, eg approaches to the game regarding tactics to minimise cost, but focusing on semi-relevant statistics when we have an absolute list of attributes is unhelpful. If people want to do it that's fine, I just don't see how it would gain you an advantage. As far as I am concerned all statistics tell you is how other teams set up their tactics and how effective your tactics are, depending on how you are playing the game.

Share this post


Link to post
Share on other sites
it's not the same argument used against Beane. baseball scouts were looking at statistics that were less effective at winning games of baseball and looking at physical attributes that were less relevant.

I don't see the advantage in pouring over statistics to decide whether a certain player would be a good fit for a system, when we have it right in front of us when you look at their profile.

I think there are interesting things to take from moneyball, eg approaches to the game regarding tactics to minimise cost, but focusing on semi-relevant statistics when we have an absolute list of attributes is unhelpful. If people want to do it that's fine, I just don't see how it would gain you an advantage. As far as I am concerned all statistics tell you is how other teams set up their tactics and how effective your tactics are, depending on how you are playing the game.

But im not trying to replace attributes in the game, far from it. I just want to see if moneyball can be translated into the game and whether studying ingame statistics and finding undervalued players abiding by the rules laid out above, and if it can yield the same results or thereabouts of playing the usual way snapping up the best players.

I personally think there is a lot of scope for this and think it will be a very financial encouraging game to play. Its not for everyone, just an experiment of sorts.

Share this post


Link to post
Share on other sites

Couple of points

Football has so many variables it leaves quite a number of statistics misleading.

Also moneyball system was tried in football, in England. It was in 2011 with the Fenway sports group who made purchases such as Stewart Downing and their "panic" buy Andy Carroll. The reason they were so keen to replace outgoing Torres with Carroll was a pure stats based decision & probably the fact he was English!

Downing had most chances created or most assists & Andy Carroll had most Aries challenges won ect it was something like that but I know they used this system which served their Red Sox so well and as we all know it failed. Terribly.

Share this post


Link to post
Share on other sites
But im not trying to replace attributes in the game, far from it. I just want to see if moneyball can be translated into the game and whether studying ingame statistics and finding undervalued players abiding by the rules laid out above, and if it can yield the same results or thereabouts of playing the usual way snapping up the best players.

I personally think there is a lot of scope for this and think it will be a very financial encouraging game to play. Its not for everyone, just an experiment of sorts.

As one of the people arguing for using attributes over statistics, I definitely misunderstood what you were attempting. I thought that you were claiming that using statistics was the optimal way to play. I now see that you're attempting to follow a real world approach as a self-imposed constraint, regardless of whether or not it is optimal. I think that's really interesting, and I'm keen to hear how things progress.

One thought....have you look at what the CIES Football Observatory is doing? They have a system of rating players, plus a financial model for predicting transfer costs. There's not a ton of detail available, but at a high level they rate players in six categories, and then take different weightings of the categories based on positions.

They define the categories as follows:

  • Shooting: ability to exploit goal opportunities through accurate shooting.
  • Chance creation: ability to putting teammates in a favourable position to score.
  • Take on: ability to create advantageous situations by successfully challenging opponents.
  • Distribution: ability to keep a hold on the game through efficient passing.
  • Recovery: ability to minimise opponents’ chances through proficient interception work.
  • Rigour: ability to minimise opponents’ chances through robust duelling.

Looking back at the list of metrics that you mentioned in your OP, these line up quite well:

  • Shooting: Non-Penalty goals, Shots, Shooting%, Goal Conversion%
  • Chance creation: Assists, Key passes
  • Take on: Successful Dribbles
  • Distribution: Passing%
  • Recovery: Int
  • Rigour: tackles

The site is: http://www.football-observatory.com. The document describing what they are doing is here: http://www.football-observatory.com/IMG/pdf/cies_footobs_eng.pdf

Share this post


Link to post
Share on other sites

I will be updating it in the coming weeks, been swamped with uni work due to it being my final year etc etc.... But I do aim continue this once i get sufficient free time, the next few updates are practically wrote up, just need a few amendments.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...