st4lz

I have been asked an interesting question about how to judge the keepers and I happened to have a terrible defending performance this season, so I decided to give it a go. I divided the scope of play into three areas: shot-stopping, box threat prevention, and build-up play. 1. Shot stopping 1.1 Expected Save Percentage Overperformance Expected Save % is about the percentage of shots we would expect a goalkeeper to save based on post-shot xG of shots faced, and we have it available as "xSv %". The simple formula is save_ovp = save_pct / xsave_pct, where results > 1 should indicate that the goalkeeper is performing better than average. To my surprise, no keeper in the league I play achieved that, but I try to explain that there is much disparity between the offensive and defensive players in the Turkish league. I wouldn't be too shocked if the model hasn't been calibrated well enough, too. 1.2 xG Prevented Per Save xGP is only useful if we know how busy the goalkeeper was. Usually, goalies from the underdog teams have more opportunities to drive this number up, as they are defending much more often. Calculating this is quite tricky because we don't have the number of shots faced, so we have to derive this number from other stats. The solution is to get this number from goals conceded and save pct, like this: shots_faced = goals_conceded / (1 - save_pct). Then, we can use: xGPrev/Shot = xGP / shots_faced. The two above methods are highly correlated with each other, as they are probably based on the same model, which is good as it confirms the validity of this one. 1.3 Save distribution There are a few save types (handled, tipped, parried) that have been available on FM for some time, but I never thought about them much, until now. If we sum them up together with goals conceded, we end up almost exactly at the shots faced (unplayable leagues have a little more, playable a little less, but who cares why). What we definitely can do, is to have some kind of league average, and then compare the keepers to that average. Someone who can handle balls more often than others is more skillful. I'd say a tipped save is better than parried, as parried may go straight into the opponent's foot. Conceding more is worse. Alternatively, I created a scoring model when handled_pts = 3, tipped_pts = 2, parried_pts = 1, and conceded_pts = -3. After that, calculate a weighted average of percentages: save_dist_score = handled_per_shot * handled_pts + ... You can see all three dimensions in the chart, the bigger the dot, the better the save distribution is (this is the 2026/27 season, so some names here might be unusual). 2. Box threat prevention This section covers: interceptions/shot, tackles/shot, mistakes/shot. The bigger the dot, the more mistakes per shot faced (no dot - no mistakes). As you can see, there isn't much difference for a sweeper play, at least when it comes to box prevention, and the data distribution seems random. Also, the goal mistakes are too sparse to be meaningful, even on the full-season data. 3. Build-up play This is the least useful for scouting, as it is affected by the team tactics a lot. There are two types of teams, those who focus on retention and play from the back, and those who just hoof the ball forward from goal. Unfortunately, the effectiveness of the latter is just terrible, and I wonder why most of the teams play this way. There is the top of the league with a high Pas %, all the rest just waste possession, while they should protect it even more when defending. Again, I compared 3 areas: pass_pct, passes_attempted_to_progressive_ratio, and passes_wasted_to_progressive_ratio for goalkeepers. The higher the dot, the less progressive passes are made compared to attempted. I made this analysis in Python, let me know if anything is incorrect, missing, or not clear enough. The next step should be taking whatever is useful for scouting, reduce to one simple value with some kind of scoring model, and figuring out if there are any missing gems to buy to prove this idea makes sense. I could potentially add this functionality to the app if it works.

Of course, the formulas should be easy to apply to Excel for anyone. It is so simple, that I even doubt if it is correct.

I mentioned this idea before, this is how I'd evaluate keepers defensively. My idea is to derive shots faced from conceded and sv %, as the sum of tipped, parried, and held doesn't seem to add up into total. Any per 90 metrics for goalies makes no sense to me.

I wanted to do it on my website, but if you want to do the app yourself, I won't spoil the fun for you. I'll figure out my own. Just don't forget to pin me like this time when you are done @milenec11 Save % is indeed irrespective of the chances faced. xG-Prev doesn't say how many chances you had to increase the number. A fair comparison would be xG-Prev/ShotsFaced, as it takes both quality and quantity into account. The optional stats for me are either Pas % for short distribution (but mostly to eliminate bad passers rather than reward the best ones), or KeyPasses/90 if I would want the goalkeeper to initiate counter-attacks (1-2 per season may be luck, but seeing more would be interesting).

I speak Polish, so for me Czech is not only beautiful, but also funny Isn't Save Ratio the same as Shots per Conceded, but inverted? Overall, I think what you picked is very reasonable and well thought. My question is, where to go from here? Do you think it is possible to create a ranking for each stat per position and combine all the results together using weights for [P, S, O] category? We could end up having a list of best performers for each position per league. Would the results make sense and you trust the order?

Are the match ratings in real football much better than those in FM, by the way?

@steve.battisti You can take a look at my website https://moneyballfm.com. You can get some ideas there. You will probably have the most success with the free youth players released by the bigger clubs and loans. Use trials whenever you can to explore free agents. There are always hidden gems available in different ways, it's good to check all the possibilities and decide what works for you. The bigger your club is, the transfers with fees will make more sense as your scouting team will be better and the available talent pool smaller. Good luck!

Looks awesome! Clearly, a great effort was put into this. I like all types of analytics people use while playing the game, but appointing the wage structure is something I can't fully grasp. Firstly, when I want to sign a player, I negotiate hard. Even if I plan to use the player as a first-team member, I don't hesitate to give him an Impact Sub, if he agrees to. Regarding wages, the transfer fees often surpass the yearly wage multiple times, so the wage is usually secondary. If I can sign a player for 500k and I know his value is around 5M, I can easily give him 500k p/a like other First Team players in my team, even if I know he wouldn't play regularly. Also, the problem with pre-defining key players is, that if I see they are underperforming, they immediately sit on the bench. I don't care who is the Important Player and who is Fringe, if they earn their chances, they play. This may change a couple of times throughout the season, and I am not too attached to it. If a player is unhappy about the playing time, well, everybody will have to leave eventually, if a good offer comes at the right time I sell, no matter how unsettled he is. I am worried the wage budget would put too many constraints on the way I am managing clubs. Don't you feel it is too much corporate management for a football club?

Dribble failures to be precise, successful dribble may lead to other actions, like a shot. Would be awesome if some kind of passes received and passes aimed were added to the game, we could analyze how good a player is as a pass recipient (the first touch impact). It would be very useful to counter gegenpress. We should add possession won to it and have touches. But I doubt possession won/lost is a reliable stat now, I think it may be broken. The same for crosses received/aimed for separating offensive headers.

@dunk105 The raw stats are nice to have, but they are not too useful when used as is. Let's consider the scenario of looking for a new Pressing Forward. The natural process of selecting a good PF is to take stats and order by Pressures Attempted per 90 or Pressures Completed per 90. But, I can guarantee, if you do that you will come up with poor results. It is really hard to say if the number, let's say 8.4, is much better than 7.6. The amount of pressures is highly correlated with the amount of time the team is in possession, meaning the teams that have only 35% average possession have much more chances to make pressures, than teams that have 65% possession. You may find out that players on top of this list are not even the PF, but other roles that low-possession teams are playing. To balance that, you could discount the pressures amount by the average team possession stats. The problem is, the only way the data is aggregated in FM is per 90 minutes. There are other approaches in football analytics, and one of them is to present data per touch, which takes into account possession. touches = passess_attempted + crosses_attempted + headers_attempted You can aggregate touches per team per 90 to know the average team possession. Then, you can look at the pressures distribution by possession, and decide how much more pressures should be valued per 1% possession increase. There are a lot of calculations to get there, but the output might still be not satisfactory, as you could also factor in Opponents PPDA, to make sure the player is really outstanding with pressures, and it is not influenced by the team's tactical focus. My point is, that when looking for a forward position, we should primarily care about the main KPI which is the ability to score goals (F9 and similar creative forwards should be rather judged as attacking midfielders IMO). Then, if we isolate ~5 best options, we should factor in the pressing tactical requirement. It is a completely different approach than comparing stats of two strikers in your own team. If they play enough amount of time, the level of opponents is similar and their teammates are the same (if there is the backup team playing only cup games or low tier opponents, it is still hard to compare with first team). You can compare not only the pressures output, but most importantly, what is the impact of the pressures on the overall team performance. All of the above lead me to conclusion, that recruitment KPIs and performance KPIs should be different. It would be awesome if you put a v4 update, I would gladly take a look at what changed.

@Rodrigogc Fair point, the stats are useful for finding overperformers, but they won't let you find an excellent player who is for various reasons not performing or simply sitting on the bench. The thing is, you skip them as 1 out of 1000 is useful compared to 1 out of 10 which produces good stats. For the newgens, I would manage the youth team as well. I would play them in the same position for a long time to have comparable stats. Also, I'd use generic CB, DM, CM, AM, etc for youth team tactics, to not emphasize any specific role, so their stats should show what they are good at. But youth stats are quite useless, they are not exportable and it is all about how well they perform in the senior team. That's what friendlies are for, you should watch the games and make your guesses.

I am pretty sure you have success. People are playing attributeless and have success, even though it looks like a huge disadvantage. Paying attention to the data brings a completely different level of understanding to this game. Attributes are often misleading, there are a lot of hidden factors that play a role as well, the game is exploiting very well the biased judgment of our minds. The stats are much more reliable. Also, if you know how to play, the game is not too hard, so I would be amazed if anyone who puts in a certain level of cognitive effort is unsuccessful. This is huge, I found out the same. There is too much data to look at everything. Usually, you want a player to be great in just 1, 2, or max 3 areas, the other areas are either unimportant or he just needs to be around average. I saw people playing "Moneyball" saves and trying to search the replacements with 8-10 stat filters, the random process of the stats distribution and variety of player types, league quality, team performance, tactics, etc makes it very random who is going to match the criteria. I don't trust it myself, I constantly improve it as I see the flaws from time to time. The point is, if there is a wonderkid striker in the league or a really good one well above the others, he is always somewhere on top. You can try it by yourself: https://moneyballfm.com/Strikers I'd like to eventually create a page for every position, but while forwards, keepers, or center-backs seem easy, there is a lot of variance in central midfield. Maybe I will implement the KPIs that are composable instead. I would be happy to hear some ideas

I can confirm it was due to the Scout Feedback being assigned to DoF. After changing to Chief Scout, scouting is progressing, although there are differences from the previous versions. I don't think it is a bug, it seems like a legitimate penalty when some staff member is too overloaded with responsibilities.

I did a lot of research into this and haven't come up with a definitive solution yet. Main issues: * Some of the stats are favoring players heavily in teams that have an advantage - it is almost impossible to beat the average xG or Pas % from the underdog team * Opposite, some of the stats are much better in the underdogs, especially the defensive ones - all of the Tackles Completed, Key Tackles, Interceptions, Clearances, Blocks, Headers, etc are higher for teams that defend often * Lower efficiency doesn't always mean worse performance - ex. Headers Won Ratio for two CB, one is the air superiority corner target, and the other stays back during Attacking Set Pieces. The first one may have 60% HWR with 12 Aerial Attempts per 90, the other may have 80% with 3 Aerial Attempts per 90. HWR is as useless as Conversion Rate or Shots on Target % (vs. xG), you can't just aggregate them and treat all situations equally (you discriminate low probability scorers, like players who score a lot of headers or long shots will have lower). * Some of the stats are dependent on team tactics or playing roles, and can't be compared objectively - it is hard to compare a No-Nonsense CB with a CB when it comes to Pass %, the first one is encouraged to make safe clearances. Or to compare a Pressing Forward with a Trequartista when it comes to Pressures Attempted. * Some of the stats are negatively correlated - there is often a low Pas % that goes along a high Key Passes. Creative players do take more risks, so their mainline Pas % is lower, which doesn't mean their passing is weaker. You need to identify those relations, and not try to put too much weight on Pas % when looking for a playmaker. I generally like to create indicators that are constructed based on available statistics, I like the Tackle Quality (Tackles Attempted / Fouls Made) that you created. I have completed my evaluation model for forwards, and I think the fewer areas you compare the better results you may have. My model assumes 3 main abilities to evaluate a forward: winning space, finishing, and success to waste ratio. Winning space (favors big teams): simply Np-xG/90 Finishing (favors underdogs): This is xG overperformance, but calculated as a ratio, not as a difference - Np_Goals/Np-xG Success to waste ratio (neutral for team strength or tactical bias): Shots on Target, Crosses Completed, Key Passes, Critical Chances Created, Headers Won divided by Shots off Target, Missed Crosses, Wasted Passes, Offsides, Headers Lost I have quite a success in finding great overperforming strikers with this model.

7 players out of the top 25 in the queue are in the youth teams and not playing. I changed the instruction for them to '1 Week' and put them at the bottom of the queue for now. I also noticed my Director of Football was assigned the duty of Providing Scout Feedback, I changed it to a Chief Scout. I will test if any of those changes matter.