Thumbnail image prepared by Lucas De Luigi: www.behance.net... Dropbox links Test League Editor file: www.dropbox.co... Test League Savefile: www.dropbox.co...
I reckon the leagues not being full detail helps as the amount of factors that can go wrong when developing a player, things like sackings, coaches, injuries, morale, ineptitude by the AI, etc. are largely glossed over and CA growth is applied more cleanly. Certainly a thing I'll keep in mind for my saves. With regards to the League Rep tests, I think it would be interesting to see more relevant spreads of reputation used, as 200 and 100 are worlds apart, essentially comparing the Premier League+ to the Norwegian second tier (4.5 stars vs. 2.5 stars in game) , and the former value probably being impossible to achieve organically? Not to mention the fact that if your wonderkid can get gametime in the Premier League, he's most likely gonna be fine. It would be more interesting to essentially compare different data points equivalent to in game leagues, maybe to see if the effect is linear or eventually peters out. I really appreciate the work you're doing, but sometimes I think you need to focus less on IF something has an effect, and more on finding the point of deminishing returns that are more relevant to FM players, though to your credit you did find great data on the Mental attributes in this video.
Hey bud, yup I agree 1/200 league reputation is extreme. I did feel the need to use extreme parameters in my tests to make the differences as visible as possible. If I tested using leagues with similar reputation eg. 170/200, 175/200, 180/200, the difference will be too minor and may lead me to incorrectly conclude that league reputation has no effect on player growth.
My thinking is that nonloaded leagues are sort of averaging out all the factors and thus make the progress more smooth and steady potentially leading to slightly bigger gains.
@@Draddar I think so. As I showed in the video, inactive/view only leagues don't have any coaching staff so maybe the game engine just assumes perfect quality coaching.
I like that the benefits of coaches and training facilities seem to be non linear. Going from 1 to 10 is way more important (and easier) than from 10 to 20.
You're making amazing videos and analysis man, keep it up! I cannot wait for you to take all your findings and knowledge of both FM22/23 and creating the ultimate training schedule to put everyone else's to shame... although hopefully balanced to include keepers and defenders into the growth since they were both neglected a bit in 22 I found.
The last point is really surprising. As I understood the patch notes, CA progression was really significantly worse in non-loaded leagues before and they fixed that. Supposedly to make it equal to playable. Possibly overdone it? I have been noticing quite decent growth from my players loaned to those leagues lately, hard to quantify as there are so many factors in a "normal" save, I'd say anecdotally not worse than playable leagues and maybe even better. So your experiment kind of confirms this, but I'm still shocked.
Yup, in the Training Analysis Part 1 video I noticed how Endrick in a non-active Brazilian league didn't grow as well as I anticipated. This was back in v23.1. So maybe now the devs fixed it but possibly overdone it.
Interesting results, thanks for the time you put into making these. I think some of your conclusions need more testing: - For natural fitness, you're looking at players with 1 injury proneness, playing a roughly equal number of games across 1 season. I've been doing some alternative testing (analysis of a 20 year sim with default database) and it suggest natural fitness does have some effect. I think it's likely that it increases the amount of game time a player is available for over the course of their career, which would allow greater opportunity to develop & retain ability. - As you touched on, your goalkeeper sample size is small. I think this probably accounts for some of the differences without outfield player development, I'm not convinced GKs are treated differently when it comes to determination, ambition etc. - A minor one, but determination & ambition effect on development appears to be a curve that levels out rather than a cutoff at a given number. As you found, very high ambition/determination aren't much better for development than average values, but there was still a small correlation with growth in my tests. This could also be down to player attribute distribution, i.e. better players tending to have higher determination from the start. - You recommend loaning to inactive leagues, but also identify that league rep, facilities & staff quality are fairly important for development. In normal gameplay, inactive leagues are likely to be lower quality ones (depending on db setup). I think this warrants further testing to understand how these factors interact.
You raise valid points. - It appears that NAT is more to do with recovery rather than growth/decline. Therefore it is possible that players with high NAT will on average be able to play more matches throughout his career which will translate to more growth during teenage years and less decline during 30s. This way, NAT contributes indirectly towards a player's growth/decline. - I agree there's no reason why the GKs would be treated differently to outfield players by the game engine. In this video there were 2 perceived differences between outfielders and GKs, ambition and training intensity. It is possible that the relatively small sample size of GKs produced statistical noise and the effects of ambition / training intensity on GKs were not represented accurately by the line graphs. - Yup, the effects of determination & ambition on growth are probably logarithmic rather than a 'cut-off'. eg. 20/20 determination may be a tiny bit better than 10/20 determination. - Absolutely, there are multiple factors that should be considered when choosing a club to loan out your players, including facilities, staff, league reputation etc. And yes the inactive leagues tend to be lower reputation leagues, so all these factors should be considered when making your decision.
Hi Max, great video as always. With the loan test, I think the real test is whether it is better to loan them out to play regular football at a lower level or to stick around not playing but training with good facilities and coaches.
Hey bud, I didn't test this but my guess is that it is better to go out on loan and get playing time, since the benefit of regular match exposure > harms of inadequate facilities & staff. This is assuming that the loaning club has bad facilities & staff. If a top club like Bayern Munchen wants to loan your player with promises of regular appearances, why not?
For the staff attributes experiment, wouldn't it be a good idea to see how the star rating of the assignments affect player growth, not just the accumulative attributes of the staff. I understand you partially did test this, as the better stats the higher the stars, but with enough coaches it should be possible for every coach to have five stars, and then test this against a team with four stars, three, etc. Maybe even link it up with the attributes improved, as explained when you hover over the assignment in the edit coach assignment screen. This also brings to question the focus the Staff page puts on accumulative staff attributes, when it compares you to the rest of the league. Would a team with only ten coaches, each coaching one assignment to 2.5 stars (maybe with one or two extra 1ca coaches assigned to everything just to lower the workload) have the same result as ten coaches with perfect stats, assigned to all ten categories, where the star rating is watered down to 2.5 stars (or whereever it plateaus).
Hey bud, valid points. It's not just the coaching attributes but the stars of the individual coaching assignments that count. Assessing the impact of the coaching assignment rating will be a bigger job however since it'll involve looking at the growth of different attributes and not just CA. For now I just wanted to keep it simple and look at the staff attributes. In the future I'll make a video dedicated solely on staff so I can delve deeper into the coaching assignments.
@@ebfm I do have a test idea that can probably be quickly assessed at CA level. Namely if we add a bad coach to reduce coach workload does it have an impact. I always do this to bring the workload to average, and the game tells us A. that the quality isn't impacted (stars stay the same) and B. that lower workload (or at least not too high) for the coaches gets better results. Both of which are I think highly questionable. The B part might require a bit more work, but for A just taking your 20/20 coaches test and adding an additional 1/20 coach and assign him to help with everything and you can easily compare it to 20/20 baseline. Perhaps even adding multiple of 1/20 coaches and assigning all of them to everything to first of all actually reduce workload visibly in-game and also potentially "dilute" the work that the good coach does. I know you are swimming in propositions for various kinds of tests so no pressure 😉 I just love this kind of content.
@@Draddar Interesting suggestion mate, kind of similar to what Mads Larsen has suggested (the impacts of coaching assignment ratings). For now I'm thinking more about topics like: what happens when you place O23 players in U23 team / O18 players in U18 team / U18 players in senior team. But after that I can certainly take a look into coaching assignments.
I was not surprised by the loaned out test as it was run. However, the point of a loan is to get the player playing time. If I am considering loaning a player then odds are he is not going to get much playing time at my squad. You showed in your last video (or perhaps the one before that) that playing time is much better than just training sessions, and full game playing time is better than coming in as a sub, for player growth. Loans are used to get the player the playing he would not get a home. The better test would be to test how a loaned player, playing full time at loaned club, compares in development to a player staying at his hime club and just training, playing as a sub, or playing in the youth team.
Hey mate, absolutely the main purpose of a loan is to get your players more playing time. In this video I wanted to find out whether the act of loaning out a player itself had any effect on the rate of growth.
@@ebfm i think a good test to simulate if it would be good to loan youngsters is to do test 1 where the player would not play matches and do an optimal training schedule, such as yours or ZaZ, and test 2 where the player is loaned out to a team where he will play every match but have suboptimal default training schedules
@@ebfm If test 1 would prove to be better than test 2 we could have more grown at club players and influence their training rating(if it actually matters) and help their hidden attributes, this would also give us more control on what attributes the player could develop, and their familiarity with different positions
Hello mate, thanks for time you've spent on this research. Amazing work overall but i wanted to add a bit of criticism: 1) Since FM16 i am watching how people try to decide which hidden attribute affects player growth the most. And honestly for me it is a waste of time. Because in game you are not working with one or another attribute - you see player's personality as a whole. And personality is some kind of a summ of these attributes. You will never find a player with 1 Det, 1 Amb and 20 Pro, and you will never find a player with good personality (i mean with good pro attribute) with low numbers in det and amb. So for me even a Balanced guy with, let's say 12 determination, and Reserved/Evasive media handling style worth keeping in club, cause he already have these hidden attributes which will allow him to progress further. 2) Natural fitness allow player to recover faster after match. It means that he will return at high training intensity faster, what at the same time lead to better growth. I am now playing in Iceland league where whole tournament (22 games+cup+eul) being played in 4 months. From july to september team have 3 matches per week and i definetly see how nat fitness works. Guys with low attribute don't train at all and are a subject to rotation :D 3) During FM20 times my computer melted and i started playing on crap laptop with little number of leagues. And i had the same question - what if i send youngsters on loan to inactive league, will they grow somehow? And i saw an official answer on SI forum (can't find it right now sorry), where one dev confirmed that there is no actual difference between active or inactive league at this case. Because what matters - type of contract (part time/full time), match ratings and league reputation. And i wonder how you interpret the result of your experiment. And what exactly you did. Because if you sent players on loan from worst playable league with part time contracts to a test league with full time contract and better reputation with condition as regular player, even if test league is inactive, i have no question why they progress better then in wales. And i dont speak of adap attribute. You see there are more factors here. A mind breaking question for me here appears when i have to decide wether i send a player on loan to 4rth division but at a team wich will won competition OR i send him to 3rd division at a team wich will fight against relegation (regular starter in both cases ofc). So i am choosing between lower rep but higher ratings AND higher rep and lower ratings. I hope you get what i mean. Sorry for my english, hope you understood what i tried to say. :3
Hey mate, appreciate the feedback. 1) Of course there won't be any real players with 1 Det 1 Amb 1 Pro. But in my tests I had to use these extreme values in order to isolate the effects of individual attributes and evaluate their effects. 2) NAT may allow players to recover faster after a match, which will allow players to return faster to training which will accelerate growth. This is something that I did NOT test in this video. It is possible that in my tests, there are 30 matches/season which is not too busy (about 1 match/week), so all players trained about the same amount regardless of their NAT, resulting in similar amount of growth regardless of NAT. 3) Sure, there will be more than 1 factor that affects the rate of growth. My test results suggest that players develop faster in non-active leagues, all other factors being equal. Not sure what the devs meant back in FM20. I do know that in the FM23 v23.2 patch the devs made some adjustments to player growth in non-active leagues, which may have changed things.
@@ebfm actually ,speaking of personalities, i was pretty wrong. Just for fun had a look at newgens in my save (2031 year), and found few guys with professional/fairly professional personalities with extremely low determination 1-7. For me this is nonsence because i was sure that even fairly professional has 10-14 det attribute. But all in all pattern for newgens seems to be the same - high professionalism goes hand in hand with at least good det attribute. It is what i see since FM16.
The more I've fiddles about with training the more I think it's intrinsically tied to game time, which no one's really tested in depth yet. Just like when you do a match practice session and they use it to practice their individual training, actual matches seem to work exactly the same way. I'm increasingly getting the impression that whatever you have set in individual training is boosted massively by match time. For players that are actually regular starters, for me, I'm seeing that general training doesn't seem to matter much at all, more so the individual training. I'm managing a team that's now in 2044 and wins everything so my players don't really get much time training due to rest time, but their improvements very much follow what is set in individual training.
Hey mate, I haven't tested the effects of individual training yet so can't say for sure. It is a frequently requested topic though - hopefully I can get to that topic sometime in the future.
A small criticism : players of high age are the best in my opinion ! I'm already mourning the fact that I won't be able to get Zlatan in my team next year :( but FM isn't just for tryharding, it's also a bit of fun and an old legend like Buffon, Neuer, Rammos etc. is always great in the team.
Goalkeepers not being affected by training intensity makes sense when you remember they arent the ones doing dribbling or technique drills with the rest of the squad, theyre just stopping shots in team drills, or off doing their own drills, and I doubt goalkeeping specific drills are all that intense
I think the Error in "Playable vs View Only vs Inactive" is caused by the Devs overtuning a fix, as it would make sense that players in those leagues missout on some aspect of training, be it their games played, their training, or their coaches, and so SI had to introduce some buff to those leagues in order to not have players from say Japan, in FM 23, left behind in the world stage. They likely had to set some constant bias to make up for it that they simply set too high
On the loaned out part, I think a better test would be to see if the player stayed at his own club but is not playing, against going to another club with lower reputation but regular playing time
Hey bud, for the scenario you proposed, the factors in play are 1. Playing at a lower reputation league = lower growth 2. More playing time = more growth Factor 2 is more decisive than factor 1. So going on loan to get more playing time will definitely be more beneficial than staying at your own club and not getting any game time.
It's probably not the best place to write this but quick thought for you: what if age doesn't impact that much the growth of the players, but the attribute score does? The higher the score, the slower the potential growth possible. In reality it makes sense and it's exponential, you usually reach a plateau from which it's really hard to improve. This would mean that a 25 years old player with the same attributes scores as a 15 years player would grow more or less the same given the same circumstances. Have you tested this before? And also to consider, is there a sort of plateau in attributes scores? Like after a certain value, the growth it's so slow that it's better to focus training on other aspects?
Hey mate, the answer is that players growth slows down due to 2 mechanisms. 1. I showed this in the Part 1 video (ua-cam.com/video/zZU8zu_2Xx4/v-deo.html) - there are 2 obvious 'steps' in the rate of growth - the first at 19 and the second at 27. 2. This is exactly what you've suggested - the closer a player's CA gets to his PA, the rate of growth slows down. This was also shown in the Part 1 video.
@@ebfm So is it possible that without matches, with everything else identical between 1st team and U20 (coaches, structure and training), the league reputation doesn't have any impact? Which would mean, training in the 1st team isn't really useful unless players play matches? (well, it's always useful in a real situation because of mentoring and probably better coaches, but sometimes the heavier schedule reduce the amount of training, so it would be interesting to compare that real case situation...)
First of all, thanks for your effort, your content is really helpful to understand some dynamics. I wonder if you are planning to investigate Technical based training’s effects on team’s performance. As I see, in FM 23 team performance is not only based on player attributes but also players’ familiarity to the gameplay. For example, when I’m playing with a gegenpress mentality tactic, team performs much more better if I train them Transition - Press session.
hey mate, I am planning to go through the individual training sessions one by one. Not straightaway though, it will probably be covered in Part 6 or 7.
I'm guessing Natural fitness has more of an effect on match sharpness rather than how long attributes are held based on your results. I can't picture another effect of the attribute other than that if it's not a bug
I think the loaned out player conclusion may be a bit of a red herring because you would loan out a player to get first team football, so the better comparison would be would loaning a player out be better than them staying at the parent club and not receiving first team football, and secondly would a player playing regular at a team with worse facilities grow better than a player not playing at the parent club with better facilities.
Hey! Thank you very much for all the testing!! If the players develop more in inactive leagues this means they will develop more if we loan them out to a inactive league than playing in our first team... Not realistic at all... I think is a bug!! Do you know if they try to fix it on the most recent update? (I think there was a update after you done this video). Keep up the excellent work, thank you!!!!
Great video as always. Can you go into more depth with youth coaching? Some would argue that coach's personality, preferred formations & working with youngsters are more important than actual coaching attributes at youth level.
Excellent point, Ahmed. Last week, Dalijt uploaded a video about this topic [ua-cam.com/video/ledh6TAolXM/v-deo.html], "COACHES DON'T MATTER when..". It will be very interesting to see data results about youth teams and how to get better results on developing players.
Could it be possible to do a video about youth intakes? Things like head of youth impact (reputation and attributes wise), facilities, club reputation...
@@ebfm if you do the youth intake study some factors I recommend you check is city attraction,club reputation, state of development,game importance and finally the facilities
@@ebfm Have in mind, that increasing - decreasing parameters for Youth Intake; need some years to have an affect and it's not going to be "visible" in first year.
One thing I noticed scumloading is that even though in December I had a golden generation foreseen, in March the chance to actually get it is below 1% (with exceptional rectruitment and excellent youth facilities in a 5star rep team and league and one of the best HOYD)
@@afasia2341 Interesting. This is also something I've been wanting to test for ages. I totally approve of your scumloading though. It's the basis for scientific observations such as the one you described.
Even though I enjoy your videos and I am waiting for the "best training schedule" I believe you should run your tests in a real world situation and not in a custom database. Also your findings should be discussed with SI but if you are using custom databases SI won't listen.
There are too many unintended variables in a non-custom database. He is trying to isolate independent factors so he has to make sure everything except that factor stays the same.
The findings with training intensity probably align with why more than 30 games didn't have as much benefit. Fatigued and worn down players not getting the same quality of training.
When it comes to the inactive vs active leagues, it makes me curious how it determines matches in terms of number of matches played, minutes, and general fatigue. If it's assuming they get quality playing time and not actually fatiguing them in the process, then they'd get maximum benefit of both experience and training.
Loaning out players to inactive leagues, there might be another factor in that. Adaptability might cause players to grow worse in other countries/ continents. Would be interesting to see if that makes a difference also good to look at when getting wonderkids from south america if they have bad adaptability maybe it effects their player growth
That's actually a great point. I've been neglecting the role of Adaptability, but it may play a part in player growth, especially when you bring in South American wonderkids to Europe. I'll do a test on it in the future.
@@ebfm I think we can all relate finding a great signing only to see lack of adapt / Consistency / injury prone / Perhipral I am really curious to see how these effect performance & growth. Also consistency might not directly effect growth but when on loan to another club they might play em less cus worse performance and thus reducing growth indirectly.
Hey Max. I was just thing about hows players love/hate big games? How do they know it is a big game? This got me thinking about whether reputation affects the outcome of the game? Is something you'd be interested in testing?
Hey mate, it's an interesting question. It probably has to do with the reputation of the competition. The 'Important Match' attribute of the players should also play a part in deciding the player's performance in competitions with high reputation. I'm focusing on the training aspect of the game at the moment, but maybe in the future I can run some tests regarding competition reputation.
Didn't freeze them. But D/A/P stayed the same 99% of the time. Which means growth of D/A/P must be due to factors other than training / matches. Some people have commented that Determination can go up when you punish them after a bad performance.
Something to explore is if moneyball style player picking based on statistical performance is still possible in FM23. Since we have seen the meta tactics this year are not as effective as previous years. I wonder if CA/PA is king or if it's still possible (or ever was) to pick a winning team based on xG/match engine performance separate from attributes and CA ?
CA/PA are pretty reliable indicators or talent. It is said that Physical attributes are more important than Mental / Technical attributes in FM, so prioritise players with high Physical attributes as well.
There's no way of knowing the exact value without using the in-game editor. But the personality of the player can give some hints. A player with high Professionalism with have personalities such as 'Professional' 'Model Citizen' or 'Perfectionist'.
Were you thinking about setting up a Discord channel so people can help running the tests? I think if you taught few willing people how to do them you would have much bigger sample
I thought about it but may be not at the moment. Gathering the sample size is not the main challenge as I use a macro program to repeat the trials. It's coming up with the tests / constructing the editor files / constructing the macro / data analysis that take the bulk of my time, which can't really be delegated to other people.
Hi mate. If we were to send a player to an inactive/view only league with a decent reputation, would we see additional player development speed at clubs with better training facilities?
Hey, it's hard to say. You've mentioned 3 different factors that affect growth: inactive/view-only league, league reputation and the quality of training facilities. It's hard to say which of these factors are the most influential. So I must resort to a generic response: it's case by case, consider all factors involved!
How do you know what a players professionalism is if you don't have the in-game editor? Is there a hierarchy which you could point out to me/us please?
There is no way of knowing the exact value of professionalism / ambition without using the editor. One clue is looking at the players' personality. A player with high professionalism will have personalities such as 'professional' 'model citizen' or 'perfectionist'.
Whats the point of natural fitness attribute than? I'm confused. Also loaning out player should affect player growth if he is getting play time at that club opposed to sitting on a bench in a parent club. Last point is kind of weird to say the least. If anything player should grow more in a loaded league. Weird
- NAT is possibly more to do with recovery after a match. - Yes loaning out a player will be more beneficial for a player if it means more game time for the player. This experiment shows that the state of being a loanee itself does not affect growth. - Yes I also found it weird that an inactive league speeds up growth more than an active league..
You have to use the editor to know the exact value. But the personality gives you some hints eg. Model Professional / Model Citizen indicates a high Professionalism.
@@renyygade Each trials were done for 1 season. Graphs are for separate players aged between 15-30. It's not the record of a single player's growth throughout his career. eg. If a 15yr old grew by 20 CA points and a 16yr old grew by 15 CA points, it does NOT mean that the player's CA became 100 > 120 > 135. It means a 15yr old player's CA became 100 > 120, and in a separate trial, a 16yr player's CA became 100 > 115. Hope that answers, let me know if it's still unclear
Do you have a discord group or something to discuss about it? The CA progresssion was also good for when you don't load too many leagues but still want a competitive CL or other international competitions. Before this update it was pretty easy to win CL after some seasons because players in non active leagues didn't develeop well
I reckon the leagues not being full detail helps as the amount of factors that can go wrong when developing a player, things like sackings, coaches, injuries, morale, ineptitude by the AI, etc. are largely glossed over and CA growth is applied more cleanly. Certainly a thing I'll keep in mind for my saves.
With regards to the League Rep tests, I think it would be interesting to see more relevant spreads of reputation used, as 200 and 100 are worlds apart, essentially comparing the Premier League+ to the Norwegian second tier (4.5 stars vs. 2.5 stars in game) , and the former value probably being impossible to achieve organically? Not to mention the fact that if your wonderkid can get gametime in the Premier League, he's most likely gonna be fine. It would be more interesting to essentially compare different data points equivalent to in game leagues, maybe to see if the effect is linear or eventually peters out.
I really appreciate the work you're doing, but sometimes I think you need to focus less on IF something has an effect, and more on finding the point of deminishing returns that are more relevant to FM players, though to your credit you did find great data on the Mental attributes in this video.
Hey bud, yup I agree 1/200 league reputation is extreme. I did feel the need to use extreme parameters in my tests to make the differences as visible as possible. If I tested using leagues with similar reputation eg. 170/200, 175/200, 180/200, the difference will be too minor and may lead me to incorrectly conclude that league reputation has no effect on player growth.
My thinking is that nonloaded leagues are sort of averaging out all the factors and thus make the progress more smooth and steady potentially leading to slightly bigger gains.
@@Draddar I think so. As I showed in the video, inactive/view only leagues don't have any coaching staff so maybe the game engine just assumes perfect quality coaching.
I like that the benefits of coaches and training facilities seem to be non linear. Going from 1 to 10 is way more important (and easier) than from 10 to 20.
So many myth busters in few minutes! Thanks!
You're making amazing videos and analysis man, keep it up! I cannot wait for you to take all your findings and knowledge of both FM22/23 and creating the ultimate training schedule to put everyone else's to shame... although hopefully balanced to include keepers and defenders into the growth since they were both neglected a bit in 22 I found.
Your videos are amazing, I hope the FM Devs actually watch them to learn how their game actually performs
The last point is really surprising. As I understood the patch notes, CA progression was really significantly worse in non-loaded leagues before and they fixed that. Supposedly to make it equal to playable. Possibly overdone it? I have been noticing quite decent growth from my players loaned to those leagues lately, hard to quantify as there are so many factors in a "normal" save, I'd say anecdotally not worse than playable leagues and maybe even better. So your experiment kind of confirms this, but I'm still shocked.
Yup, in the Training Analysis Part 1 video I noticed how Endrick in a non-active Brazilian league didn't grow as well as I anticipated. This was back in v23.1. So maybe now the devs fixed it but possibly overdone it.
@@ebfm ya according to this data you found they overcooked it a bit
Wasn't so excited to see a notification in a while
Best FM UA-cam channel? No experiment needed to figure that answer out. (I mean yes, obviously)
Interesting results, thanks for the time you put into making these.
I think some of your conclusions need more testing:
- For natural fitness, you're looking at players with 1 injury proneness, playing a roughly equal number of games across 1 season. I've been doing some alternative testing (analysis of a 20 year sim with default database) and it suggest natural fitness does have some effect. I think it's likely that it increases the amount of game time a player is available for over the course of their career, which would allow greater opportunity to develop & retain ability.
- As you touched on, your goalkeeper sample size is small. I think this probably accounts for some of the differences without outfield player development, I'm not convinced GKs are treated differently when it comes to determination, ambition etc.
- A minor one, but determination & ambition effect on development appears to be a curve that levels out rather than a cutoff at a given number. As you found, very high ambition/determination aren't much better for development than average values, but there was still a small correlation with growth in my tests. This could also be down to player attribute distribution, i.e. better players tending to have higher determination from the start.
- You recommend loaning to inactive leagues, but also identify that league rep, facilities & staff quality are fairly important for development. In normal gameplay, inactive leagues are likely to be lower quality ones (depending on db setup). I think this warrants further testing to understand how these factors interact.
You raise valid points.
- It appears that NAT is more to do with recovery rather than growth/decline. Therefore it is possible that players with high NAT will on average be able to play more matches throughout his career which will translate to more growth during teenage years and less decline during 30s. This way, NAT contributes indirectly towards a player's growth/decline.
- I agree there's no reason why the GKs would be treated differently to outfield players by the game engine. In this video there were 2 perceived differences between outfielders and GKs, ambition and training intensity. It is possible that the relatively small sample size of GKs produced statistical noise and the effects of ambition / training intensity on GKs were not represented accurately by the line graphs.
- Yup, the effects of determination & ambition on growth are probably logarithmic rather than a 'cut-off'. eg. 20/20 determination may be a tiny bit better than 10/20 determination.
- Absolutely, there are multiple factors that should be considered when choosing a club to loan out your players, including facilities, staff, league reputation etc. And yes the inactive leagues tend to be lower reputation leagues, so all these factors should be considered when making your decision.
This is the best FM content being created today
Hi Max, great video as always. With the loan test, I think the real test is whether it is better to loan them out to play regular football at a lower level or to stick around not playing but training with good facilities and coaches.
Hey bud, I didn't test this but my guess is that it is better to go out on loan and get playing time, since the benefit of regular match exposure > harms of inadequate facilities & staff.
This is assuming that the loaning club has bad facilities & staff. If a top club like Bayern Munchen wants to loan your player with promises of regular appearances, why not?
Just going through your videos, very interesting & great presentation. I don`t mind going back to school!
Very good content. A lot of it was as I thought but the last point no 10 was very surprising.
For the staff attributes experiment, wouldn't it be a good idea to see how the star rating of the assignments affect player growth, not just the accumulative attributes of the staff. I understand you partially did test this, as the better stats the higher the stars, but with enough coaches it should be possible for every coach to have five stars, and then test this against a team with four stars, three, etc. Maybe even link it up with the attributes improved, as explained when you hover over the assignment in the edit coach assignment screen.
This also brings to question the focus the Staff page puts on accumulative staff attributes, when it compares you to the rest of the league. Would a team with only ten coaches, each coaching one assignment to 2.5 stars (maybe with one or two extra 1ca coaches assigned to everything just to lower the workload) have the same result as ten coaches with perfect stats, assigned to all ten categories, where the star rating is watered down to 2.5 stars (or whereever it plateaus).
Hey bud, valid points. It's not just the coaching attributes but the stars of the individual coaching assignments that count.
Assessing the impact of the coaching assignment rating will be a bigger job however since it'll involve looking at the growth of different attributes and not just CA. For now I just wanted to keep it simple and look at the staff attributes. In the future I'll make a video dedicated solely on staff so I can delve deeper into the coaching assignments.
@@ebfm I do have a test idea that can probably be quickly assessed at CA level. Namely if we add a bad coach to reduce coach workload does it have an impact. I always do this to bring the workload to average, and the game tells us A. that the quality isn't impacted (stars stay the same) and B. that lower workload (or at least not too high) for the coaches gets better results. Both of which are I think highly questionable.
The B part might require a bit more work, but for A just taking your 20/20 coaches test and adding an additional 1/20 coach and assign him to help with everything and you can easily compare it to 20/20 baseline. Perhaps even adding multiple of 1/20 coaches and assigning all of them to everything to first of all actually reduce workload visibly in-game and also potentially "dilute" the work that the good coach does.
I know you are swimming in propositions for various kinds of tests so no pressure 😉 I just love this kind of content.
@@Draddar Interesting suggestion mate, kind of similar to what Mads Larsen has suggested (the impacts of coaching assignment ratings).
For now I'm thinking more about topics like: what happens when you place O23 players in U23 team / O18 players in U18 team / U18 players in senior team. But after that I can certainly take a look into coaching assignments.
I was not surprised by the loaned out test as it was run. However, the point of a loan is to get the player playing time. If I am considering loaning a player then odds are he is not going to get much playing time at my squad. You showed in your last video (or perhaps the one before that) that playing time is much better than just training sessions, and full game playing time is better than coming in as a sub, for player growth. Loans are used to get the player the playing he would not get a home. The better test would be to test how a loaned player, playing full time at loaned club, compares in development to a player staying at his hime club and just training, playing as a sub, or playing in the youth team.
While I agree with the above it's still worth to check in this purely isolated form and maybe build from there.
Hey mate, absolutely the main purpose of a loan is to get your players more playing time. In this video I wanted to find out whether the act of loaning out a player itself had any effect on the rate of growth.
@@ebfm i think a good test to simulate if it would be good to loan youngsters is to do test 1 where the player would not play matches and do an optimal training schedule, such as yours or ZaZ, and test 2 where the player is loaned out to a team where he will play every match but have suboptimal default training schedules
@@ebfm If test 1 would prove to be better than test 2 we could have more grown at club players and influence their training rating(if it actually matters) and help their hidden attributes, this would also give us more control on what attributes the player could develop, and their familiarity with different positions
I found your contents are very deep and useful, thanks
Hello mate, thanks for time you've spent on this research. Amazing work overall but i wanted to add a bit of criticism:
1) Since FM16 i am watching how people try to decide which hidden attribute affects player growth the most. And honestly for me it is a waste of time. Because in game you are not working with one or another attribute - you see player's personality as a whole. And personality is some kind of a summ of these attributes. You will never find a player with 1 Det, 1 Amb and 20 Pro, and you will never find a player with good personality (i mean with good pro attribute) with low numbers in det and amb. So for me even a Balanced guy with, let's say 12 determination, and Reserved/Evasive media handling style worth keeping in club, cause he already have these hidden attributes which will allow him to progress further.
2) Natural fitness allow player to recover faster after match. It means that he will return at high training intensity faster, what at the same time lead to better growth. I am now playing in Iceland league where whole tournament (22 games+cup+eul) being played in 4 months. From july to september team have 3 matches per week and i definetly see how nat fitness works. Guys with low attribute don't train at all and are a subject to rotation :D
3) During FM20 times my computer melted and i started playing on crap laptop with little number of leagues. And i had the same question - what if i send youngsters on loan to inactive league, will they grow somehow? And i saw an official answer on SI forum (can't find it right now sorry), where one dev confirmed that there is no actual difference between active or inactive league at this case. Because what matters - type of contract (part time/full time), match ratings and league reputation. And i wonder how you interpret the result of your experiment. And what exactly you did. Because if you sent players on loan from worst playable league with part time contracts to a test league with full time contract and better reputation with condition as regular player, even if test league is inactive, i have no question why they progress better then in wales. And i dont speak of adap attribute. You see there are more factors here. A mind breaking question for me here appears when i have to decide wether i send a player on loan to 4rth division but at a team wich will won competition OR i send him to 3rd division at a team wich will fight against relegation (regular starter in both cases ofc). So i am choosing between lower rep but higher ratings AND higher rep and lower ratings. I hope you get what i mean.
Sorry for my english, hope you understood what i tried to say. :3
Hey mate, appreciate the feedback.
1) Of course there won't be any real players with 1 Det 1 Amb 1 Pro. But in my tests I had to use these extreme values in order to isolate the effects of individual attributes and evaluate their effects.
2) NAT may allow players to recover faster after a match, which will allow players to return faster to training which will accelerate growth. This is something that I did NOT test in this video. It is possible that in my tests, there are 30 matches/season which is not too busy (about 1 match/week), so all players trained about the same amount regardless of their NAT, resulting in similar amount of growth regardless of NAT.
3) Sure, there will be more than 1 factor that affects the rate of growth. My test results suggest that players develop faster in non-active leagues, all other factors being equal. Not sure what the devs meant back in FM20. I do know that in the FM23 v23.2 patch the devs made some adjustments to player growth in non-active leagues, which may have changed things.
@@ebfm actually ,speaking of personalities, i was pretty wrong. Just for fun had a look at newgens in my save (2031 year), and found few guys with professional/fairly professional personalities with extremely low determination 1-7. For me this is nonsence because i was sure that even fairly professional has 10-14 det attribute. But all in all pattern for newgens seems to be the same - high professionalism goes hand in hand with at least good det attribute. It is what i see since FM16.
just came across your channel, awesome content!! cant wait what else your research and experiments proves.
"31 and BYE!" - what is common a playboy boss and an fm player :D
The more I've fiddles about with training the more I think it's intrinsically tied to game time, which no one's really tested in depth yet. Just like when you do a match practice session and they use it to practice their individual training, actual matches seem to work exactly the same way. I'm increasingly getting the impression that whatever you have set in individual training is boosted massively by match time. For players that are actually regular starters, for me, I'm seeing that general training doesn't seem to matter much at all, more so the individual training. I'm managing a team that's now in 2044 and wins everything so my players don't really get much time training due to rest time, but their improvements very much follow what is set in individual training.
Hey mate, I haven't tested the effects of individual training yet so can't say for sure. It is a frequently requested topic though - hopefully I can get to that topic sometime in the future.
A small criticism : players of high age are the best in my opinion ! I'm already mourning the fact that I won't be able to get Zlatan in my team next year :( but FM isn't just for tryharding, it's also a bit of fun and an old legend like Buffon, Neuer, Rammos etc. is always great in the team.
Sound is great on this one, thanks!
This channel does god's work
Great video..thanks for your efforts
Also really interested to see an experiment on the impact that injury proneness has on player growth / decline please.
Yes we need to test this attribute
타고난 체력이 선수 어빌 감소에 별 영향이 없다는 사실
활성화된 리그 보다 비활성화된 리그에서 선수가 더 잘 성장한다는 사실
이 2가지는 놀랍네요.
저도 놀랐습니다..
Goalkeepers not being affected by training intensity makes sense when you remember they arent the ones doing dribbling or technique drills with the rest of the squad, theyre just stopping shots in team drills, or off doing their own drills, and I doubt goalkeeping specific drills are all that intense
I think the Error in "Playable vs View Only vs Inactive" is caused by the Devs overtuning a fix, as it would make sense that players in those leagues missout on some aspect of training, be it their games played, their training, or their coaches, and so SI had to introduce some buff to those leagues in order to not have players from say Japan, in FM 23, left behind in the world stage. They likely had to set some constant bias to make up for it that they simply set too high
thanks for the info 🙏🙏as always some curious results haha
On the loaned out part, I think a better test would be to see if the player stayed at his own club but is not playing, against going to another club with lower reputation but regular playing time
Check the last video "FM23 Training Analysis 2 - Match appearance [...]" you will find some answers !
Hey bud, for the scenario you proposed, the factors in play are
1. Playing at a lower reputation league = lower growth
2. More playing time = more growth
Factor 2 is more decisive than factor 1. So going on loan to get more playing time will definitely be more beneficial than staying at your own club and not getting any game time.
Great vid. Love it.
It's probably not the best place to write this but quick thought for you: what if age doesn't impact that much the growth of the players, but the attribute score does? The higher the score, the slower the potential growth possible. In reality it makes sense and it's exponential, you usually reach a plateau from which it's really hard to improve. This would mean that a 25 years old player with the same attributes scores as a 15 years player would grow more or less the same given the same circumstances. Have you tested this before? And also to consider, is there a sort of plateau in attributes scores? Like after a certain value, the growth it's so slow that it's better to focus training on other aspects?
Hey mate, the answer is that players growth slows down due to 2 mechanisms.
1. I showed this in the Part 1 video (ua-cam.com/video/zZU8zu_2Xx4/v-deo.html) - there are 2 obvious 'steps' in the rate of growth - the first at 19 and the second at 27.
2. This is exactly what you've suggested - the closer a player's CA gets to his PA, the rate of growth slows down. This was also shown in the Part 1 video.
@ebfm thanks! I'll go watch it again, guess I was distracted 🙃
Hey Max, for the league reputation test, were the players playing matches or just training?
@@baiisun Hey mate, yes there were playing matches.
@@ebfm So is it possible that without matches, with everything else identical between 1st team and U20 (coaches, structure and training), the league reputation doesn't have any impact? Which would mean, training in the 1st team isn't really useful unless players play matches? (well, it's always useful in a real situation because of mentoring and probably better coaches, but sometimes the heavier schedule reduce the amount of training, so it would be interesting to compare that real case situation...)
First of all, thanks for your effort, your content is really helpful to understand some dynamics.
I wonder if you are planning to investigate Technical based training’s effects on team’s performance.
As I see, in FM 23 team performance is not only based on player attributes but also players’ familiarity to the gameplay. For example, when I’m playing with a gegenpress mentality tactic, team performs much more better if I train them Transition - Press session.
hey mate, I am planning to go through the individual training sessions one by one. Not straightaway though, it will probably be covered in Part 6 or 7.
Fantastic video thanks.
I'm guessing Natural fitness has more of an effect on match sharpness rather than how long attributes are held based on your results. I can't picture another effect of the attribute other than that if it's not a bug
It's biggest impact as shown before in one of Max's videos is the rate of condition recovery after a match
Yup, NAT may have other benefits such as speeding up a players' recovery after a match.
Comparing not playing for home team, playing for home team’s reserve team, or playing full time for another team would be informative.
* Natural Fitness affects how fast players recover their condition after playing a game *
Amazing
amazing informations keep it up
I think the loaned out player conclusion may be a bit of a red herring because you would loan out a player to get first team football, so the better comparison would be would loaning a player out be better than them staying at the parent club and not receiving first team football, and secondly would a player playing regular at a team with worse facilities grow better than a player not playing at the parent club with better facilities.
Hey! Thank you very much for all the testing!!
If the players develop more in inactive leagues this means they will develop more if we loan them out to a inactive league than playing in our first team... Not realistic at all... I think is a bug!!
Do you know if they try to fix it on the most recent update? (I think there was a update after you done this video).
Keep up the excellent work, thank you!!!!
Thanks mate, cant say whether they've fixed it in the recent patches..
Thanks!!!
Great video as always. Can you go into more depth with youth coaching? Some would argue that coach's personality, preferred formations & working with youngsters are more important than actual coaching attributes at youth level.
Interesting, not sure how the coach's preferred formation will affect youth development but I can certainly try to do some tests.
Excellent point, Ahmed. Last week, Dalijt uploaded a video about this topic [ua-cam.com/video/ledh6TAolXM/v-deo.html], "COACHES DON'T MATTER when..". It will be very interesting to see data results about youth teams and how to get better results on developing players.
Could it be possible to do a video about youth intakes? Things like head of youth impact (reputation and attributes wise), facilities, club reputation...
Hey mate, it's a commonly requested topic but I'm working on some other topics at the moment, related to training.
@@ebfm It'll have to wait then. Excited to see your next video, I really enjoy the content. Keep it up!
hope next be youth intake and all the myths and facts that effect it. only small subject to study not take time at all lol.
Coming up soon, although I've been saying that forever 😅
@@ebfm if you do the youth intake study some factors I recommend you check is city attraction,club reputation, state of development,game importance and finally the facilities
@@ebfm Have in mind, that increasing - decreasing parameters for Youth Intake; need some years to have an affect and it's not going to be "visible" in first year.
One thing I noticed scumloading is that even though in December I had a golden generation foreseen, in March the chance to actually get it is below 1% (with exceptional rectruitment and excellent youth facilities in a 5star rep team and league and one of the best HOYD)
@@afasia2341 Interesting. This is also something I've been wanting to test for ages.
I totally approve of your scumloading though. It's the basis for scientific observations such as the one you described.
Very interesting
Even though I enjoy your videos and I am waiting for the "best training schedule" I believe you should run your tests in a real world situation and not in a custom database. Also your findings should be discussed with SI but if you are using custom databases SI won't listen.
There are too many unintended variables in a non-custom database. He is trying to isolate independent factors so he has to make sure everything except that factor stays the same.
The findings with training intensity probably align with why more than 30 games didn't have as much benefit. Fatigued and worn down players not getting the same quality of training.
When it comes to the inactive vs active leagues, it makes me curious how it determines matches in terms of number of matches played, minutes, and general fatigue. If it's assuming they get quality playing time and not actually fatiguing them in the process, then they'd get maximum benefit of both experience and training.
Loaning out players to inactive leagues, there might be another factor in that. Adaptability might cause players to grow worse in other countries/ continents. Would be interesting to see if that makes a difference also good to look at when getting wonderkids from south america if they have bad adaptability maybe it effects their player growth
That's actually a great point. I've been neglecting the role of Adaptability, but it may play a part in player growth, especially when you bring in South American wonderkids to Europe. I'll do a test on it in the future.
@@ebfm I think we can all relate finding a great signing only to see lack of adapt / Consistency / injury prone / Perhipral
I am really curious to see how these effect performance & growth.
Also consistency might not directly effect growth but when on loan to another club they might play em less cus worse performance and thus reducing growth indirectly.
Hey Max. I was just thing about hows players love/hate big games? How do they know it is a big game? This got me thinking about whether reputation affects the outcome of the game? Is something you'd be interested in testing?
Hey mate, it's an interesting question. It probably has to do with the reputation of the competition. The 'Important Match' attribute of the players should also play a part in deciding the player's performance in competitions with high reputation.
I'm focusing on the training aspect of the game at the moment, but maybe in the future I can run some tests regarding competition reputation.
@@ebfm Yeah, sounds good. Hope the training stuff is coming along. Been very good.
I had a GK who was 43 in Japan before thank goodness his natural fitness wasn't 1
Do another test with staff same value but high and low reputation, 1 star and 5 star. do same with players.
Regarding Determination/Ambition/Professionalism, did you somehow freeze them, or did they also progress throughout the season?
Cheers
Didn't freeze them. But D/A/P stayed the same 99% of the time. Which means growth of D/A/P must be due to factors other than training / matches. Some people have commented that Determination can go up when you punish them after a bad performance.
Something to explore is if moneyball style player picking based on statistical performance is still possible in FM23. Since we have seen the meta tactics this year are not as effective as previous years. I wonder if CA/PA is king or if it's still possible (or ever was) to pick a winning team based on xG/match engine performance separate from attributes and CA ?
I think CA/PA is a king this FM
CA/PA are pretty reliable indicators or talent. It is said that Physical attributes are more important than Mental / Technical attributes in FM, so prioritise players with high Physical attributes as well.
How do you know Professionalism on a player? Since it's a hidden attribute.
There's no way of knowing the exact value without using the in-game editor. But the personality of the player can give some hints. A player with high Professionalism with have personalities such as 'Professional' 'Model Citizen' or 'Perfectionist'.
Were you thinking about setting up a Discord channel so people can help running the tests? I think if you taught few willing people how to do them you would have much bigger sample
I thought about it but may be not at the moment. Gathering the sample size is not the main challenge as I use a macro program to repeat the trials.
It's coming up with the tests / constructing the editor files / constructing the macro / data analysis that take the bulk of my time, which can't really be delegated to other people.
Hi mate. If we were to send a player to an inactive/view only league with a decent reputation, would we see additional player development speed at clubs with better training facilities?
Hey, it's hard to say. You've mentioned 3 different factors that affect growth: inactive/view-only league, league reputation and the quality of training facilities. It's hard to say which of these factors are the most influential.
So I must resort to a generic response: it's case by case, consider all factors involved!
Can you have your conclusions at the end of the videos like some of your others?
Hey mate, I did include a summary at the end of this video, you might have missed it 😅
How would you check professionalism, is the best way their personality?
Personalities like Model Citizen, Professional, Model Professional, Resolute, Perfectionist have high Professionalism hidden attribute.
Yup, personality is the only way to estimate a player's ambition / professionalism without using the editor
How do you know what a players professionalism is if you don't have the in-game editor?
Is there a hierarchy which you could point out to me/us please?
There is no way of knowing the exact value of professionalism / ambition without using the editor.
One clue is looking at the players' personality. A player with high professionalism will have personalities such as 'professional' 'model citizen' or 'perfectionist'.
@@ebfm thank you, I love your work btw, I love the analytics - it's what is needed!
Whats the point of natural fitness attribute than? I'm confused. Also loaning out player should affect player growth if he is getting play time at that club opposed to sitting on a bench in a parent club. Last point is kind of weird to say the least. If anything player should grow more in a loaded league. Weird
- NAT is possibly more to do with recovery after a match.
- Yes loaning out a player will be more beneficial for a player if it means more game time for the player. This experiment shows that the state of being a loanee itself does not affect growth.
- Yes I also found it weird that an inactive league speeds up growth more than an active league..
High Natural Fitness will keep his physical attributes high despite being 30+ y/o.
Have you thought of testing work rate as well?
Not for now. I believe Work Rate is more to do with how hard the players work on the field instead of having an influence on player growth.
Is their a way to know a players professionalism without using the editor?
You have to use the editor to know the exact value. But the personality gives you some hints eg. Model Professional / Model Citizen indicates a high Professionalism.
@@ebfm thank you, sir.
How does the CA growth per season add up to more than 200 CA?
The graphs show the growth in CA per person per player. So a y-axis value of 20 means 1 player grew by 20 CA points in 1 season.
@@ebfm Yes, and the Y values on multiple occasions add up to more than 200 CA points in growth over the course of their career. How is that possible?
@@renyygade Each trials were done for 1 season. Graphs are for separate players aged between 15-30. It's not the record of a single player's growth throughout his career.
eg. If a 15yr old grew by 20 CA points and a 16yr old grew by 15 CA points, it does NOT mean that the player's CA became 100 > 120 > 135.
It means a 15yr old player's CA became 100 > 120, and in a separate trial, a 16yr player's CA became 100 > 115.
Hope that answers, let me know if it's still unclear
Hey, are you on Twitter?
I’m waiting your new videos like waiting a new episode series in Netflix 🥹🥹
Do you have a discord group or something to discuss about it? The CA progresssion was also good for when you don't load too many leagues but still want a competitive CL or other international competitions. Before this update it was pretty easy to win CL after some seasons because players in non active leagues didn't develeop well
I do have a discord id Max EBFM#6245
feel free to message me if you have any questions