Moneyball Strategy?

I was just wondering has anyone tried the moneyball method and if so how has it worked out for you, im intrigued to see if it works because i may try it myself.

Care to explain what moneyball is

I don't apply full-on "moneyball" as such but I do look out for players that have unusually-good scoring records or average ratings at international level (as this is more trustworthy than club football in a lot of ways) or players who score a lot of goals from non-striking positions.

For example, a former trainee of mine got 20 goals for another club, playing in central midfield 80% of the time and 15% of the time on the wings, although around 7 goals were penalties. I'd already known he was a very good striker from his youth so it was easy to extrapolate that he would be at least 20-goal striker in a superior team (mine). And he got 30+ goals when I bought him back.

I do look out for the odd unusually-high average rating but I can never depend on it because I'd never end up signing anyone, since these are rare.

Instead, I generally take a look at their attributes and if there are no glaring issues with solid key attributes, I'll keep an eye out.

Sikker: have you not watched the film

It isn't all that relevant for football as football is a much more complicated game than baseball. Also, Liverpool have exposed one flaw, which is chance creation. They signed Adam, Downing and Henderson as they were three of the leaders of the "chance created" statistic last season. What passed everyone at the club by was that Adam was the sole set piece taker at Blackpool and the other two took a good share of them at their clubs. Obviously all three couldn't take at Liverpool, particularly with Gerrard and Bellamy around too.

Sikker: have you not watched the film

The book which came out 9 years ago is so much better than the crappy film. Football isn't a fit for the A's strategy because it isn't a game based almost purely based on statistics, which baseball is. There's the odd example but not across the board with things like WHIP, ERA, Ks etc..

The book which came out 9 years ago is so much better than the crappy film. Football isn't a fit for the A's strategy because it isn't a game based almost purely based on statistics, which baseball is. There's the odd example but not across the board with things like WHIP, ERA, Ks etc..

whats WHIP, ERA and Ks

whats WHIP, ERA and Ks

Walks and Hits per Inning Pitched, Earned Run Average and Strikes. All standard baseball terminology. There are tons of other stats too. It's a mathematician's wet dream. Completely the opposite to football which is based on interpretation of styles and skills rather than numbers.

whats WHIP, ERA and Ks

WHIP: Walks and Hits per Innings Pitched

ERA: Earned Run Average

It is basically ratio- and statistics-driven transfers.

The Drowning and Adam ratios should have really been taken as a proportion of goals/assists/chances created as a proportion of the entire team. But then again, where do you stop when it comes to including and excluding things in ratios? You can go on forever... (Which is why I don't think it works in football, unlike baseball. There are many styles of football, and really only a few ways of playing baseball.)

Nope... Just watched the trailer and baseball doesn't interest me the slightest bit

In general I want to make money with every transfer, or buy a player I know is good enough to start (rare), and will play at a high level for a couple years. A successful team is a couple of stars, surrounded by great role players, IMO. Sure, you could have a eleven stars in the starting XI and 7 more on the bench, but its not likely, is it?

So I try to sign youth who could AT LEAST be good role players in the future. I make sure they have high determination (15+), their potential is rated highly, no glaring weaknesses, although I make allowances for players who have low bravery, natural fitness or aggression, as I don't rate those as the most crucial attributes (and natural fitness is more important when they are 31 than 21!)

Players who aren't CBs can have no aerial game as long as they are good at everything else, strikers with no jumping are OK if they have pace.

Each position has attributes I look for, but thats the general guidelines.

I'm not really sure what the 'moneyball method' is, in FM it's pretty easy to be very successful just by being smart with your money. It's not hard at all to bring in players cheaply or on a free purely to generate money. Buy player from Argentina that scout thinks has a good future cheaply. Loan out to feeder clubs for 2 years. Sell on for massive profit.

It isn't all that relevant for football as football is a much more complicated game than baseball.

I disagree with this. A lot.

I disagree with this. A lot.

and whats your reasoning for disagreeing

As Moneyball is essentially signing under-valued players, everyone tries to do it, and Newcastle have done it particularly well this year.

For a team so publicly trying to mimic it, Liverpool have been showing exactly what not to do, in signing over-valued English players.

You can definitely take elements of the A's strategy and put it into practice, obviously the stats you use will be different as it's a different sport and you will have different types of player which will adjust to your own personal style and interpretation of how football should be played. For example, if you want a midfield general you won't focus on passing, instead you will focus on his tackle & interception success rate, where his successful tackles are made, how far he runs during a game, how many bookings he picks up.

Another element would be noticing where a player is better suited to another position and re-training them to make better use of their attributes. I once found a 24 year right midfielder, who had awful attributes for that position but had stellar attributes for a central defender, this was in the lower Italian leagues and he ended up getting sold for a huge profit. To be fair this happens quite a bit in real-life but with perhaps less drastic positional changes.

I disagree with this. A lot.

If you disagree with the relevance, I'd appreciate further clarification.

If you disagree with how complicated the games are relative to each other, then we probably have a terminology problem. Perhaps "chaotic" is a better word than "complicated". Baseball- a ball is pitched, a batsman must hit it, the batsman must run around the bases before the ball gets around. Football- there are many different ways of scoring a goal, players have much more choice, matches are much more chaotic and less regimented. A scale would go like this: darts-chess-baseball/cricket-rugby-football-war.

A baseball player who hits a home run 50% of the time is probably as good as another who hits a home run 50% of the time. Sure, the quality of pitcher may vary, or the direction of the wind, or the size of the ground (I'm unsure how strict the regulations are in baseball), but those factors should balance out in the long run. Likewise pitchers and fielders. Football is much more chaotic. You can't compare Leon Britton to Paul Scholes and conclude that Britton is a better passer because he completes 98% of his passes whereas Scholes completes 90% (for example). Britton's role is to play simple passes with players near to him. Scholes does that, but he is also tasked with difficult forward passes to forward attacks and split defences. Some players dribble through the centre of crowded pitches, and so might be tackled more often despite being better at dribbling than a pace merchant winger who just takes on a full back out wide. Some players play in teams that only create awkward chances, but still stick away quite a few, whereas others play for teams that create "easy" chances and still miss most of them.

Baseball is a stats driven game. Contracts and selections are based on stats. Football is not - contracts and selections are not purely based on stats. i.e a 45+ goal scoring striker in the BSN is not always guaranteed to be brought up to the Premier League and earn a contract like Rooney is gonna earn. Football is more than stats. I moved to the US last year and when watching SOCCER (MLS) over here they try to add stats to the games as much as possible. The thing is football will never be as stat driven as the game of baseball. I actually like both sports but there is such a difference and therefore I feel it would be difficult to implement a method or club policy like with the Oakland A's Moneyball method, for a football club. On the other hand you would have to be naive and narrow minded to think football management and coaches couldn't learn or implement maybe something else like training techniques or mind management techniques from other sports, including baseball. Its always good to explore new techniques. I just dont think the specific MONEYBALL theory can be applied in football - in my opinion.

We could simplify what moneyball is with LFC example:

Sell Torres and Babel for 56 million GBP

Buy Carroll and Suarez for 57,5 million GBP

= total amount spent is 1.5 million GBP

Probably their combined wages are around same amount which in the end leaves the club within the budget.

Point of moneyball is to get value without overspending cos in baseball you trade players and with trading you also accept their contracts from previous clubs. So you can see trades of 1 big name for 3 hot prospects and some benefits like drafting rights.

Billy Bean used Sabermetrics as a way of determining who is a good player, while old school thing was to watch players like it's common in football nowadays, with focus on goals and assists or home runs for example.

So Comolli used statistics in a way like Sabermetrics value player. Thing is that none of us know which data he collected and gave preference to. It's still early days for deep football stats but I believe it's a matter of time when some genius will figure out which overlooked stat is showing something important.

That's so weird I was just thinking about a moneyball strategy a few days ago and about posting a thread.

and whats your reasoning for disagreeing

Don't know what his reasoning is, but baseball is quite complicated. I don't know if it is fair to say that it is more complicated than football. However, baseball sorts itself well into discret events which can be measure and tabulated. Football not so much. However, if someone were to figure out a good way to measure things in football, they would gain a significant advantage--at least until they let someone write a book about them.

A baseball player who hits a home run 50% of the time is probably as good as another who hits a home run 50% of the time.

This made all us Americans laugh and laugh.

And baseball park factors do not balance out in the long run. Recognizing this was one of the key insights in sabremetrics.

As Moneyball is essentially signing under-valued players, everyone tries to do it, and Newcastle have done it particularly well this year.

For a team so publicly trying to mimic it, Liverpool have been showing exactly what not to do, in signing over-valued English players.

This in a nutshell. It's a perfectly valid strategy for football management.

Edit: Also agree entirely with the Liverpool comment. Can't quite fathom how they could compare their transfer activites with the "Moneyball" strategy. It's the exact opposite.

This made all us Americans laugh and laugh.

And baseball park factors do not balance out in the long run. Recognizing this was one of the key insights in sabremetrics.

In both those cases, relative to football.

50% of balls faced being a home run is pretty much unheard of, right? That would 7 home runs every 8 times he steps up to the plate. Surely anyone who did that would be world class? (assuming they are batting at the highest level). As I have shown, I don't know a great deal about MLB.

We could simplify what moneyball is with LFC example:

Sell Torres and Babel for 56 million GBP

Buy Carroll and Suarez for 57,5 million GBP

= total amount spent is 1.5 million GBP

Probably their combined wages are around same amount which in the end leaves the club within the budget.

Point of moneyball is to get value without overspending

Whilst Liverpool probably got value on those sales, they didn't on their purchases. Saying they are "within budget" neglects to mention that Fernando Torres was (and still is) worth a lot more as an asset than Carroll. They also sold Babel (an asset they'd valued at £11.5m when they purchased him from Ajax) for a £3.5m loss.

Look, everyone tries to buy under-valued players, and has since the dawn of professionall football. The point of Moneyball is that you buy under-valued players according to statistical measures other teams don't appreciate the significance of. This is very hard to do in football because it is a fluid game not as easily reducible to statistical measures as baseball is.

Look, everyone tries to buy under-valued players, and has since the dawn of professionall football. The point of Moneyball is that you buy under-valued players according to statistical measures other teams don't appreciate the significance of. This is very hard to do in football because it is a fluid game not as easily reducible to statistical measures as baseball is.

This is true for real football. But I think it may work for FM. In the end, the way any result is calculated in FM can only be through pure mathematics surely, with a touch of randomness.

Even if moneyball could be applied to football, it's not necessary in Football Manager since we are giving a direct, objective representation of player ability and intelligence in the form of attributes. There is no need to use complex statistical methods to determine whether a player is underrated.

Additionally, since CA doesn't really tell you anything about PA, Football Manager's system of player development makes it resistant to moneyball analysis, which assumes you can roughly predict a player's growth. In FM, a high CA youngster may have already nearly maxed out his PA, even if he is great for an 18 year-old. Likewise, if a youngster has 100 CA and 180 PA, there's nothing to tell you that he's a high potential youth other than Scout report star ratings that simply reflect his PA (which is purely an abstraction and is not stat-based).

I agree with the post above. The moneyball strategy is really of not much use at all in FM as it provides you with clear indications of what each player can and can't do via attributes. In the real world you don't really have those and so you are left relying in statistics much more to get a feel for what a certain player is like. In FM you can actually ignore all statistics for any player and simply focus on their attributes. Having said that, moneyball doesn't actually work in real life football either as stated in several other posts before mine. The nature of the game is considerably different than that of baseball.

...Completely the opposite to football which is based on interpretation of styles and skills rather than numbers.

A tad ironic, since we all love a game that is based around a database of numbers!

In both those cases, relative to football.

50% of balls faced being a home run is pretty much unheard of, right? That would 7 home runs every 8 times he steps up to the plate. Surely anyone who did that would be world class? (assuming they are batting at the highest level). As I have shown, I don't know a great deal about MLB.

Whilst Liverpool probably got value on those sales, they didn't on their purchases. Saying they are "within budget" neglects to mention that Fernando Torres was (and still is) worth a lot more as an asset than Carroll. They also sold Babel (an asset they'd valued at £11.5m when they purchased him from Ajax) for a £3.5m loss.

50% percent homerun would be completely impossible at the major league level--or pretty much any level except perhaps little league (ages 8-12) but would be extraordinarily rare even there. But even if a 50% home run hitter existed in major league baseball he would not exist for long because he would simply be walked everytime he came to bat so he would turn into a 100% on base guy who never got the swing the bat. And just for the record two guys who hit homeruns half the time would probably not be equally good. Setting aside park effects, what do they do the other half of time. A guy who strikes out half the time would not be nearly as good as one who get other hits some of the time he doesn't hit home runes.

I would think a money ball strategy could be used in FM but it might detract from trying to win the champions league every year. It is perfectly possible to find high potentional young players for under $2million, develope them, and sell for a good profit. But if you focus on doing that how do you get and hold onto the top notch players needed to win. Of course, you can sell good players at 29 or 30 before they drop in value and buy young players to replace them. You could probably fund all your transfers out of current sales once you get it going.

Just watched the film. Not read the book. Interesting thread... someone mentioned Charlie Adam and I thought Ian Holloway showed a bit of Billy Beane about him when he signed CA from Rangers.

Isn't the point not to buy cheap >>> develop >>> sell high. The message I got of moneyball was to use the strengths of a group to perform to the very highest standard without paying an absolute fortune i.e. make the QF of the Champions League with a £250k a week budget, for example. Has anyone done that? Or will even less?!

EDIT: And I mean in a major league such as England / Spain / Italy etc, not Hungary where the contracts will be generally lower anyway.

I agree with the post above. The moneyball strategy is really of not much use at all in FM as it provides you with clear indications of what each player can and can't do via attributes. In the real world you don't really have those and so you are left relying in statistics much more to get a feel for what a certain player is like. In FM you can actually ignore all statistics for any player and simply focus on their attributes. Having said that, moneyball doesn't actually work in real life football either as stated in several other posts before mine. The nature of the game is considerably different than that of baseball.

The fact that stats arent sufficient today to use a Moneyball strategy in football only means that proper statistical analysis hasnt evolved sufficiently. Better statistical analysis is coming to football. It just hasnt been invented yet. Many Moneyball stats in baseball are recent inventions.

The fact that stats arent sufficient today to use a Moneyball strategy in football only means that proper statistical analysis hasnt evolved sufficiently. Better statistical analysis is coming to football. It just hasnt been invented yet. Many Moneyball stats in baseball are recent inventions.

True, but some were around for decades, but the experts refused to use them. Something similar may well be happening right now in football. After all, not every manager who has a clever idea on how to outwit the opposition will tell a writer about it. Many A's fans think that Billy Beane should have kept his mouth shut. Since I am a Giant fan who likes Michael Lewis books I am glad he talked.

By the way, as far as I can tell, the closest think to a Moneyball book about football is Soccernomics by Simon Kuper and Stefan Szymanski. If anyone can tell me about a similar book I would be quite pleased to hear.

Never thought I'd find an argument over sabermetrics in this forum...

I think that its a bit more difficult to use it with football as with baseball they play a much longer season and thus have a much larger sample size statistics-wise.

As far as the 'moneyball' strategy is concerned, I think the whole point is identifying which attributes and statistics are the right ones to consider when looking at a player's worth.

I've had no problem doing that in FM>

If you disagree with the relevance, I'd appreciate further clarification.

If you disagree with how complicated the games are relative to each other, then we probably have a terminology problem. Perhaps "chaotic" is a better word than "complicated". Baseball- a ball is pitched, a batsman must hit it, the batsman must run around the bases before the ball gets around. Football- there are many different ways of scoring a goal, players have much more choice, matches are much more chaotic and less regimented. A scale would go like this: darts-chess-baseball/cricket-rugby-football-war.

I'll just fix this for you:

darts-chess-baseball-cricket-rugby-football-war-AFL.

With football it is also more about fit, your looking at guys who play a certain type of way, do you play wingers/IF on the wings the stats you need will be different for both (but the rating from ASS man/scout will be rated as if they are the same)

Parts of the Moneyball idea are possible in FM, I think. Judging on stats / certain attributes and maybe considering a different position for that player (like Billy Beane did) can make a worthless player great again.

Also, the idea of buying cheap, selling high is one already put in practice by a lot of FM players.

two players...

a baseball player who scores 100 home runs in the minor leagues gets picked up by one of the majors.

a football player in the conference who scores 50 goals and gets picked up by a prem team..

i wonder who is more likely to do better.....

was just a thought about the complicity of the two games really..

ive always thought a decent player in baseball will be decent wherever he plays

not so true with football

If anybody had ever managed to apply a scheme like that to football, then he would be an eternal hero.

However, I can't see that irl nor would I know how that strategy could look like in FM.

two players...

a baseball player who scores 100 home runs in the minor leagues gets picked up by one of the majors.

a football player in the conference who scores 50 goals and gets picked up by a prem team..

i wonder who is more likely to do better.....

was just a thought about the complicity of the two games really..

ive always thought a decent player in baseball will be decent wherever he plays

not so true with football

Was he hitting home runs in single A, double A, or triple A?

If anybody had ever managed to apply a scheme like that to football, then he would be an eternal hero.

However, I can't see that irl nor would I know how that strategy could look like in FM.

It would be hard to visualise because it would likely be some insane multi-variable equation.

As an example, a player's performance has to be "benchmarked" against things like the team's average performance and the league's average performance. So you could in theory derive an equation to plot a curve (or in reality, a >4D hyperplane) for average performance per team. You then look for players who are (substantially) above that curve given the quality of the squad they are in.

Of course, how would you derive performances? You could derive it from long-term analysis by creating a model that is similar to perhaps SI's match engine rating system (but it has to be more advanced as it has to consider things like injuries, fitness and form).

So the ideal moneyball would be to get, say, ProZone data and stick it through this giant black box to identify good players. It would be like solving a 100+ variable equation, essentially. And a model that complex will of course be questioned...

two players...

a baseball player who scores 100 home runs in the minor leagues gets picked up by one of the majors.

a football player in the conference who scores 50 goals and gets picked up by a prem team..

i wonder who is more likely to do better.....

was just a thought about the complicity of the two games really..

ive always thought a decent player in baseball will be decent wherever he plays

not so true with football

There are a number of things wrong with your baseball analysis here. This is not intended in anyway as an insult. You didn't grow up with baseball just as I didn't grow up with football (well I grew up with American football but that is irrelevant). First, and least important, you hit home runs you don't score them. You score runs, but they occur in a variety of ways besides home runs--although every home run will score at least one run.

Second, minor league baseball players are already owned by a major league team so there is no picking up of minor league players in the same sense as in football. Further, there are severe contractual limitations on minor league players. Basically, they are not paid nearly as well as major league players. But when they are brought up to the major league a clock starts to run and the major league team has (I think) six more years of ownership before the player can become a free agent. Moreover, the player remains very cheap during the first three years. A critical part of moneyball in baseball is finding ways to effectively use these cheap years and then deciding which payers are worth paying more later, which ones to trade for value while you still can, and which ones to simply let go as free agents.

Third, home runs are not the most critical stat in evaluating a young player. Since I am not an employee of a major league baseball team up-to-date with the latest trends I can't say for certain what stats are but most likely if they could use only one stat it would be OPS which is on base percentage plus slugging percentage.

Fourth, park effects are absolutely critical. To take just one example, the Colorado Rockies, because they play at a high altitude, tend to have much higher scoring games. If you just look at the stats their pitchers look much worse than they are and the batters look much better. Likewise the shape, size, and prevailing winds make one park very different from another. I do not believe there are any effects similar in significance to this in football--recognizing that not all pitches are the same and they do affect the games.

Fifth, the difference between playing in major league baseball and one level down in AAA baseball is far greater than the comparable differences in football. Much of the difference is, in my opinion--I can't prove this of my own personal knowledge--arises out of the difficulty of hitting a major league curve ball and the much greater speed of the major league fast balls.

We may get an interesting test of relative difficulty on different levels of football next season. According to FM, Ricky Lambert of Southampton is totally inadequate as a premier league striker even though he has been scoring like crazy in League One and the Championship. Can he keep it up next year?

Fourth, park effects are absolutely critical. To take just one example, the Colorado Rockies, because they play at a high altitude, tend to have much higher scoring games. If you just look at the stats their pitchers look much worse than they are and the batters look much better. Likewise the shape, size, and prevailing winds make one park very different from another. I do not believe there are any effects similar in significance to this in football--recognizing that not all pitches are the same and they do affect the games.

Fifth, the difference between playing in major league baseball and one level down in AAA baseball is far greater than the comparable differences in football. Much of the difference is, in my opinion--I can't prove this of my own personal knowledge--arises out of the difficulty of hitting a major league curve ball and the much greater speed of the major league fast balls.

We may get an interesting test of relative difficulty on different levels of football next season. According to FM, Ricky Lambert of Southampton is totally inadequate as a premier league striker even though he has been scoring like crazy in League One and the Championship. Can he keep it up next year?

There are major differences between pitches in football. Famously, Ecuador play at altitude, and their players are better adjusted to the low-oxygen conditions so they perform better. Another example is Rotherham. The pitch is on a slight slope, and the prevailing wind blows up the slope. Traditionally, if the Rotherham captain wins the toss, he usually chooses to play downhill in the first half and uphill in the second half. Actually, I think someone might have been winding me up there, but it is definitely noticeable when you are playing uphill, or into the wind, or on a pitch that is 30m shorter than the one you played on last week.

Rickie Lambert is reflected fairly accurately in FM, certainly not as a player who is "totally inadequate" in the PL, he's rated as a good PL striker. Look at the number of 14s! He's a very good penalty taker who is good at free kicks, and very strong and good in the air. Southampton base their entire game around creating chances for Lambert, their midfield sits deep so the full backs can get forward and hit deep crosses for Lambert to head in. In the PL then more clubs will have an answer to that, and Southampton will probably win less penalties. Grant Holt would probably have been a better example.

I just watched MoneyBall..

Works great for Baseball it seems, but I can't see a way to implement this is Football..

The book which came out 9 years ago is so much better than the crappy film. Football isn't a fit for the A's strategy because it isn't a game based almost purely based on statistics, which baseball is. There's the odd example but not across the board with things like WHIP, ERA, Ks etc..

Ha, I read the book nine years ago, and I thought it sufficed for anyone that hadn't read it. I think Bennett Miller did a great job.

To answer OP, as others have stated, football's statistics do not correspond with the game like baseball does.

Second, minor league baseball players are already owned by a major league team so there is no picking up of minor league players in the same sense as in football. Further, there are severe contractual limitations on minor league players. Basically, they are not paid nearly as well as major league players. But when they are brought up to the major league a clock starts to run and the major league team has (I think) six more years of ownership before the player can become a free agent. Moreover, the player remains very cheap during the first three years. A critical part of moneyball in baseball is finding ways to effectively use these cheap years and then deciding which payers are worth paying more later, which ones to trade for value while you still can, and which ones to simply let go as free agents.

The baseball setup is more like if a Premiership team had a feeder club in the Championship, League One and League Two and loaned players to progressively higher level clubs depending on his performance.

Third, home runs are not the most critical stat in evaluating a young player. Since I am not an employee of a major league baseball team up-to-date with the latest trends I can't say for certain what stats are but most likely if they could use only one stat it would be OPS which is on base percentage plus slugging percentage.

To tie this back to what the sabremetricians figured out, on-base percentage is more important than batting average. A player may have a .270 batting average, but if he never draws a walk then you don't get extra opportunities to have him on base and able to score a run. A sabremetric analysis would possibly prefer a player with a lower batting average if he draws enough walks to get on base a percentage of the time than the guy who actually has the better batting average.

Fourth, park effects are absolutely critical. To take just one example, the Colorado Rockies, because they play at a high altitude, tend to have much higher scoring games. If you just look at the stats their pitchers look much worse than they are and the batters look much better. Likewise the shape, size, and prevailing winds make one park very different from another. I do not believe there are any effects similar in significance to this in football--recognizing that not all pitches are the same and they do affect the games.

Colorado isn't quite the launching pad it used to be since they installed the humidor.

A better example now would be San Diego and the dimensions of Petco Park. That park is so expansive that pitchers tend to look better than what they really are, because it's much more difficult to hit a home run. Ball stays in the park, more opportunities to record outs.

Fifth, the difference between playing in major league baseball and one level down in AAA baseball is far greater than the comparable differences in football. Much of the difference is, in my opinion--I can't prove this of my own personal knowledge--arises out of the difficulty of hitting a major league curve ball and the much greater speed of the major league fast balls.

Yeah, it's usually the breaking pitches that end up deciding who becomes a major league player and a guy destined for life in the minors. Plus major league pitchers don't just have a curve, they have 3rd or 4th pitches they can go to, like a slider or a circle change. The velocity on a fastball is a factor, but hitters are going to tee off on even 98-mph fastballs if they don't have any movement on them.

My own sense, btw, is that cricket would be a more accommodating sport than football when it comes to statistical analysis.

The problem for football is that it's a fluid game, where baseball (and likewise cricket) are more a series of static events one after the other (like the frames of a movie). Ice hockey is similar in having difficulty coming up with any kind of statistical analysis. American football (rugby would probably be similar) and basketball are somewhere in between, where you can find enough similar sets of action to devise objective measurements, but not as many as a game of baseball lends itself to.

EDIT: And to get my post to stay on topic, all "Moneyball" is about is finding players who the market undervalues. Michael Lewis (because of his Wall Street background) recognized right away that what the A's were doing was a version of arbitrage.

Because the A's have a budget much smaller than the Yankees or Red Sox (think the difference between Wigan's wage budget and Manchester City's), they needed to find a way to field a competitive team under their wage budget. Sabermetric analysis was still largely scoffed at in professional baseball, but the A's figured they had nothing to lose by trying it. As the movie version explains, scouts will discount players because something about them doesn't look right: a guy has a funny gait when he runs, so he's viewed as a lesser player compared to the guy with the fluid stride, even though his actual performance may be better.

It isn't all that relevant for football as football is a much more complicated game than baseball. Also, Liverpool have exposed one flaw, which is chance creation. They signed Adam, Downing and Henderson as they were three of the leaders of the "chance created" statistic last season. What passed everyone at the club by was that Adam was the sole set piece taker at Blackpool and the other two took a good share of them at their clubs. Obviously all three couldn't take at Liverpool, particularly with Gerrard and Bellamy around too.

But that's also an example of how using the wrong stats leads you to overvalue players or mismanage your squad.

Take for example (to return to baseball) my pet peeve: the save. In baseball, a pitcher gets a save if he's the last pitcher for his team, which wins the game, and he 1) enters with his team winning; 2) has the tying run at bat, on deck, or in the hole OR 3) pitches at least 3 innings.

This has lead teams to designate their best relief pitcher as the "closer," the guy whose designated role is to come in (almost always in the final inning) with a 1-run lead and get the final few outs (typically the closer comes in to start the inning).

With the way baseball teams manage their bullpen now, this means the best closers accumulate 40 or 45 saves and can expect a contract in the range of $200,000/week.

Except. If the closer is your best relief pitcher, doesn't that mean you should bring him in in the 5th or 6th inning, when the other team has, say, the bases loaded with 1 out? That never happens though. The "closer" pitches the 9th inning only (maybe the last out of the 8th as well). Even though if one of your lesser pitchers proceeds to give up a bases-clearing double that now has you behind, you might not get to use your closer at all. You end up taking the ball out of your best relief pitcher's hand.

That's an example of how stats can lead you to mismanage a squad.

Um... technically, the so-called "Moneyball" method of management has already been introduced to football. A good read would be Soccernomics.

It is essentially a method of management where you don't exceed your means. In other words, you estimate a realistic position based on your given resources. The whole "Moneyball" method far exceeds the data and options that we are capable of managing in FM.

The best we can do in applying the methods here is looking for good value buys and good sells. This method, however, revolves around the one thing almost everyone hates doing, selling their stars at overpriced market values. Certain figures in the football world believe that the transfer market is a different game on its own from the football game.

Newcastle actually applied this method in their glory days. They had excellent scouting, excellent management and weren't afraid of selling their stars. However, what they did was sign all the controversial footballers that had excellent technique, and helped those players get through their troubles. By assisting them in whatever ailed them (alcholism, drugs etc), they would buy excellent footballers that were plagued and transformed them through man management, and then sell them if big offers come in.

Now attempting this in FM is difficult due to limitations. We can't suggest Balotelli to see a shrink. in the game (exaggeration).

Personally, I believe this will only work with lower tiered clubs. Buy players that are having conflicts with their clubs but have a good degree of skill and past performances. Avoid buying players who have performed well during big international games since they tend to be overpriced.

To put it simply. Look for cheap unhappy players, improve them, sell them at inflated price, buy better cheap unhappy players, improve them, sell them at inflated prices etc.

Lastly, the one important rule that i remember being written in the book was the fact that you shouldn't sign young stars under the age of 21-22, since bellow that age, their potential is still unpredictable, you carry too much of a risk as a club, whereas by 21-22, you will have a good understanding of his abilities and his potential growth. At this age, they also tend to be cheaper if they are not already established. Lastly, unload players at the age of 29, since that is when the declining officially happens, and its where their value will be the highest.