herne79 Posted May 9, 2023 Share Posted May 9, 2023 4 hours ago, lied90 said: I'm asking for something less anecdotal and more empirical. ok, not really sure what you'd like so I'll follow your OP. The first 4 screenshots below are from my non-league save (Herne Bay). As you can see we're predicted to finish 16th. In the squad list nobody is particularly outstanding for Pace and in comparison to the rest of the league our Pace is below average (actually 13th). Without any edits we finished 3rd. The next 4 are from West Ham season 1. Predicted to finish 10th, only 6 players out of a large 26 man squad have Pace of 15 (3 of which are defenders). Compared to the other Premier League teams our Pace is well below average (17th). Again with no edits we finished runners up. So what does this tell us about Pace? Seeing as I've achieved almost the exact opposite of your experiment, should I draw the conclusion that Pace is in fact useless and other attributes are far more important, or that the game is balanced or unbalanced? No of course not - the fact is that from such a small data set and lack of investigation into other contributing factors, there are no conclusions that can be drawn (other than I'm better than the AI ). And therein lies the issue. Drawing conclusions in this manner is fraught with danger. What this thread and experiment actually is is two things: 1) A fun little experiment which demonstrates how Pace can be used to our advantage. 2) Something which appears to show a potential imbalance in the game. If the thread had stopped there I would never have jumped in. But it didn't. It went on to draw conclusions and to say rather than the game appearing to have an issue, it does have an issue. That's a big difference and one that is fundamentally flawed because not only is it open to interpretation but discounts other possible explanations, such as your own inputs being the root cause. @XaW above goes into much better detail on that front. I'm really not trying to knock your experiment. It could actually be useful for some people and give them food for thought about a new strategy they could try out by focussing more on Pace. It's the conclusions that are the problem. Link to post Share on other sites More sharing options...
Twostepsaway Posted May 10, 2023 Share Posted May 10, 2023 1 hour ago, herne79 said: ok, not really sure what you'd like so I'll follow your OP. The first 4 screenshots below are from my non-league save (Herne Bay). As you can see we're predicted to finish 16th. In the squad list nobody is particularly outstanding for Pace and in comparison to the rest of the league our Pace is below average (actually 13th). Without any edits we finished 3rd. The next 4 are from West Ham season 1. Predicted to finish 10th, only 6 players out of a large 26 man squad have Pace of 15 (3 of which are defenders). Compared to the other Premier League teams our Pace is well below average (17th). Again with no edits we finished runners up. So what does this tell us about Pace? Seeing as I've achieved almost the exact opposite of your experiment, should I draw the conclusion that Pace is in fact useless and other attributes are far more important, or that the game is balanced or unbalanced? No of course not - the fact is that from such a small data set and lack of investigation into other contributing factors, there are no conclusions that can be drawn (other than I'm better than the AI ). And therein lies the issue. Drawing conclusions in this manner is fraught with danger. What this thread and experiment actually is is two things: 1) A fun little experiment which demonstrates how Pace can be used to our advantage. 2) Something which appears to show a potential imbalance in the game. If the thread had stopped there I would never have jumped in. But it didn't. It went on to draw conclusions and to say rather than the game appearing to have an issue, it does have an issue. That's a big difference and one that is fundamentally flawed because not only is it open to interpretation but discounts other possible explanations, such as your own inputs being the root cause. @XaW above goes into much better detail on that front. I'm really not trying to knock your experiment. It could actually be useful for some people and give them food for thought about a new strategy they could try out by focussing more on Pace. It's the conclusions that are the problem. Spoiler You can't really expect to compare 115 CA players to West ham players (Imagine that if you call that overachieve with "73" points), you can try to edit players by following the cosmetic highligheted attributes (which is absurd that a game is suggesting you to go to the wrong direction) and keep the same CA to make those players faster Yet we don't know which tactic did you used, we don't know many stuff, just you trying to point out that "pace" ain't broken. But if you just opened the Arena test about those kind of attributes you would have noticed how impacteful they are. And since it happen that i have my own test league as well and you can check your self : https://bit.ly/DeliciousTest23 on the way to drive "conclusions" i made a "150/160" CA vs Elite teams and one with "140/50" CA vs Elite. Do you know what the differencet between those test leagues are ? One has Beto as AF(150/160) and One has Scamacca as AF(140/150), and overall you can check the over 3600 games the tactics results are totaly different between one table to the other. IF for you over 3600 games the same tactics with faster players make 10 points "overall", not even mentioning the GD. Faster players will make things crazier compared to the slower one. And since you even call out "strategies" you can even see that one with Scamacca prefer lone striker the Beto's one can even build around 3 strikers builds. N.B. You need to understand the difference between players, but yet you can't compare 10 pace players to 20 one. (you can drive the overall difference) but if you on West Ham got more or less 14/5 pace and other not cosmetic attributes you will tap the gap. The Current M.E. and it is since FM19 or even before that pace/acceleration are totaly broken isn't anything new, if you play gegenpress style you want your players to run and move fast. Physical's attributes are just better compared to Techical one. And if you have slower players and i have all Flash one, believe me you will lose the ball and i will score on counter. By broken it means if i have the same tecnique/Mental player but with +2 pace it will produce noticeble difference between the slower one. That could be 10/20/30 points on your "personal games". There are many more impacteful attributes, but there are many many more cosmetic stats. And to draw any conclusion you guys should use the max-output tactics not really something like those preset one. Otherwise you don't know where the tactic power fall down. You might not believe it, or you gonna deny till you want but it's really clear from this test : https://fm-arena.com/thread/5351-should-you-follow-the-highlighted-attributes-of-the-roles/page-1/ and isn't an "experiment" is just how the engine works. I did run one fast simulation with West ham : So, if by simulating i can do that, i won't even imagine if i do actually play with West Ham boys... Lost more games on away then anything : You might be right that the thread was a bit exaggerating about "pace" is the king, because is Pace/Acceleration/Jumping Reach/Agility/Strenght/Stamina (and you don't need them on 20) the dominat's one. Ofc Workrate/Marking/Determination are important too, but there are 37 attributes... (basically 10/12 doing something rest are just for the eyes) Welcome to plug and play tactics. (the most (99.9%) of them are all gegenpress-system if you are asking your self) Cheers 1 Link to post Share on other sites More sharing options...
lied90 Posted May 10, 2023 Author Share Posted May 10, 2023 (edited) 11 hours ago, XaW said: Fine, let's do this: The experiment is flawed because it fails to control enough variables. You have run it once as far as I know. You are not even scratching the surface to know where on the curve that run happened. What is your standard deviation for example? Your entire experiment is flawed as your hypothesis is prime for confirmation bias. You haven't even given much thought in how to disprove your own hypothesis, as far as I can tell. Just a random test where you seemingly randomly changed some attributes. As far as I can tell, you haven't uploaded your save for anyone to look at to confirm your findings. You've used something that looks very much like an exploit tactic (I think I've seen it or a very like one in knap's thread). Once again, VARIABLES, and here is one example of what I mean: How did the goals happen? What if most are scored from corners because you use a corner exploit? Would that still mean Pace is important? You are failing to acknowledge that different attributes might might be more important in different setups .So to "Prove" anything, you'd have to have a representative sample of different tactical setups and run each setup for a number of times for each attribute and thus create a standard deviation and evaluate from there. It would be very time consuming to do it, but if you did that then I would accept your claim, but not before. Until then you have a hypothesis, or a theory at best and not backed up by the same empirical evidence you ask for. And I don't mean to say you are wrong, you could be right for all I know, but I really despise claims like "proof" or "truth" or the like for simple tests that does not do the actual legwork required to prove anything. This is not on you alone, I've seen enough of these types of hyperbolic and exaggerated claims be hyped up around lately to be rather annoyed at it, so take it for what it is. Lots of valid points. I don't disagree with any of it, it's not like I ran a one season simulation and consider it to be proof. I think simulations I've done, I've seen others to, and the larger tests done by FM-arena gives a clear indication of how important pace is compared to other attributes. It's flawed, but it's something and it's more than purely anecdotal. Flawed empirical data can still be valuable if you take the limitations of it into account. I'm not holding anyone to a higher standard than myself, and would hope people on the other side of the argument could produce something that would suggest that me and others are wrong in our assumptions. The reason I only tested for one season is that: 1) Personal time limitation 2) The result was so shocking that even if the total points deviated by 20+ points from an expected average, it would still be insane to me considering the low quality of the players. The team is the worst team in terms of every single attribute except speed, it's just insane to me that it is possible in a game that has been developing for decades. 3) This was never intended as an experiment thread, I just rolled with it when the thread was moved and people requested me to do more (not just here, but on DMs and reddit). About disproving my(our?) hypothesis, this was never meant to be an experiment, so anything I've done after the original post have just been requests from people on how to disprove it. The attributes I changed aren't random, the pace was decreased and other attributes highlighted for the role was increased for the CA to remain roughly the same (+ - 1-2). I then tested it with a preset tactic because that was also requested to see how the team would do with a weaker tactic, with default set pieces. About the whole exploit tactics thing. Without derailing the entire thread, I'd say that I think the word exploit is thrown around on this forum very lightly compared to any other game I've played. I think it's much more accurate to call them meta tactics, and I don't think there is anything wrong with playing the meta (in most games it's the norm). I had corners aimed at front/back post and scored 11 goals that way, 3 from indirect free kicks. If you get annoyed by frequent hyperbolic claims that understandable. It's not optimal but ruffling feathers usually brings attention to a topic instead of having it instantly die out. EDIT: I tried to upload the save files but I get an error saying the file sizes are too large. Edited May 10, 2023 by lied90 2 Link to post Share on other sites More sharing options...
XaW Posted May 10, 2023 Share Posted May 10, 2023 43 minutes ago, lied90 said: Lots of valid points. I don't disagree with any of it, it's not like I ran a one season simulation and consider it to be proof. I think simulations I've done, I've seen others to, and the larger tests done by FM-arena gives a clear indication of how important pace is compared to other attributes. It's flawed, but it's something and it's more than purely anecdotal. Flawed empirical data can still be valuable if you take the limitations of it into account. I'm not holding anyone to a higher standard than myself, and would hope people on the other side of the argument could produce something that would suggest that me and others are wrong in our assumptions. The reason I only tested for one season is that: 1) Personal time limitation 2) The result was so shocking that even if the total points deviated by 20+ points from an expected average, it would still be insane to me considering the low quality of the players. The team is the worst team in terms of every single attribute except speed, it's just insane to me that it is possible in a game that has been developing for decades. 3) This was never intended as an experiment thread, I just rolled with it when the thread was moved and people requested me to do more (not just here, but on DMs and reddit). About disproving my(our?) hypothesis, this was never meant to be an experiment, so anything I've done after the original post have just been requests from people on how to disprove it. The attributes I changed aren't random, the pace was decreased and other attributes highlighted for the role was increased for the CA to remain roughly the same (+ - 1-2). I then tested it with a preset tactic because that was also requested to see how the team would do with a weaker tactic, with default set pieces. About the whole exploit tactics thing. Without derailing the entire thread, I'd say that I think the word exploit is thrown around on this forum very lightly compared to any other game I've played. I think it's much more accurate to call them meta tactics, and I don't think there is anything wrong with playing the meta (in most games it's the norm). I had corners aimed at front/back post and scored 11 goals that way, 3 from indirect free kicks. If you get annoyed by frequent hyperbolic claims that understandable. It's not optimal but ruffling feathers usually brings attention to a topic instead of having it instantly die out EDIT: I tried to upload the save files but I get an error saying the file sizes are too large. First of all, don't get me wrong, I do like that people run these types of test to look for way to improve the game, or to point out what works well or not, but recently there has been this kind of "gotcha" focus that seemingly tries to implicate that the game cheats or lies or whatever, and that's what annoys me. I mean, my real life work is software testing, so I do this kind of thing for a living! (I don't work for SI, and the mod role here on the forums is something I do in my spare time to help the community out, just to be clear) What I think the best cause of action for this to get SI attention to these types of tests, is to report them in the bug tracker with the findings and saves uploaded following the guidelines there (it's very easy, trust me!). I know others have done so already and many who continue to do so will be noticed by SI and some might even be asked by SI to do further tests at times. So that's something I think you should do with this if you main goal is to get focus on a potential issue. For what it's worth, I know SI run millions and millions of soak tests to get the statistics and players as close to real life as possible. I don't mean to shut you down here, as I do welcome these types of tests, And I even think you might be on to something from the test you ran. Your conclusion of "Pace is King, and mentals/technicals don't matter" is my issue. As you haven't proved, just hypothesized so far. And that's why I wrote the post outlining what I think would be required to "prove" it. You are free to report it or not, but that's what I would do to get focus on it since it seemingly is something you care about. 2 Link to post Share on other sites More sharing options...
herne79 Posted May 10, 2023 Share Posted May 10, 2023 7 hours ago, Twostepsaway said: just you trying to point out that "pace" ain't broken Nope. Just pointing out how such a basic test as in the OP (and all the following anecdotes) is fundamentally flawed, open to interpretation, easily contradicted and in no way sufficient to draw conclusions: 10 hours ago, herne79 said: So what does this tell us about Pace? Seeing as I've achieved almost the exact opposite of your experiment, should I draw the conclusion that Pace is in fact useless and other attributes are far more important, or that the game is balanced or unbalanced? No of course not - the fact is that from such a small data set and lack of investigation into other contributing factors, there are no conclusions that can be drawn Anyway, hopefully you’ve provided all of your test data and results to SI for their review 👍. Link to post Share on other sites More sharing options...
DMaster2 Posted May 10, 2023 Share Posted May 10, 2023 Il 3/5/2023 in 12:12 , lied90 ha scritto: Unrealistic yes, but not a bug. If you field 11 Usain Bolt on the field you will get relegated irl, because running is nowhere near enough to play football you also need mental and technical qualities. So yes this is indeed a bug in my opinion. One thing is overachieving a bit, another is getting consistent CL football with a Championship quality (at best) team. 3 Link to post Share on other sites More sharing options...
azymin Posted May 10, 2023 Share Posted May 10, 2023 17 hours ago, herne79 said: I know. And what you and everyone else who is agreeing with the OP and this experiment don’t understand is that it’s the experiment which is causing the imbalance. In this context the game isn’t unbalanced, the inputs are unbalanced. You are choosing to use these inputs which is causing the disruption (the imbalance). If you choose not to use such inputs and play the game as normal as Forest (to continue the example) you can still overachieve (or not) using different strategies. The whole point of this thread is to try to demonstrate that Pace is king/overpowered/unbalanced/whatever you want to call it. It isn’t. It is if you choose to use it as such but that’s what causes the imbalance. If the game was so unbalanced that Pace were indeed “king” the paciest sides would always win. They don’t. They can if we mess around with them but that’s our inputs, not the game. Corrected that for you. I’m confused by this. Just play the game without messing around with it and you’ll see the paciest sides don’t always win. Do you not see that when you play the game without altering it? Would you like to see screen shots of my West Ham save where I won the league first season without altering my player’s Pace? Or any just read any of the guides I keep mentioning in the Tactics forum which don’t discuss Pace? Pace is important for sure. All attributes are important, some perhaps more so than others. There have been other experiments such as this (for example with altering Determination or Jumping Reach and height) where the conclusions were erroneously drawn that those attributes don’t matter. They were quickly and easily debunked. So for sure Pace is an important attribute but your inputs here are causing the issue, not the game. You are playing the game in an unintended manner which unsurprisingly then results in the outputs you are seeing. But you and others here won’t believe that, you’re convinced there’s an issue with the game. So, for the last time, give your data to SI and let them check to see if the game is as unbalanced as you believe. Anyway, I’ll leave this now as it’s just going round in circles. Give your data to SI 👍. I don't see how you can test balance of a game if inputs are balanced. That's just a test at ideal conditions that doesn't stress test the system at all. I do agree that we're going in circles now. We'll agree to disagree on this one. 1 Link to post Share on other sites More sharing options...
enigmatic Posted May 10, 2023 Share Posted May 10, 2023 (edited) I actually think as non-scientific tests bending game logic go, this one's pretty good. Unlike the "11 Jamie Vardy's" or the ingeniously constructed "1CA" players and formation, this one is a bunch of pretty average quick-but-not-ridiculously-so real world players someone could reasonably hope to sign, and they've run the experiment to show that yes, the formation overachieves a bit with normal players too. Pace merchants are the sort of thing someone can incorporate into their setup pretty easily and doesn't feel like a total cheat. Most of us have known for years that pace is very important in most roles, average pace merchants can do a lot of damage in attack and bail you out in a high defensive line but I'm still surprised by just how well these guys have done in the Premier League (where 16 pace is not standout fast) considering that most of them are awful. as for whether it's a bad thing... well it is and it isn't. If you actually watch Traore play as a striker in game he's as frustrating as the real thing (actually preferred Pedro Neto as an advanced forward last time I managed Wolves) but yeah, of course he'll get enough chances to score for Man City and he's a lot easier than most frustratingly limited players to build a tactic around and get good results from. But yeah, Awoudja and Tebo should make too many mistakes as Premier League defenders regardless of what you do tactically and that midfield should get easily passed around It's also inevitable that pace (A player with 17 pace will almost always run faster to get the ball than a player with 15 pace) will have an outsized impact on the ME compared with some of the more abstract variables (A player with 17 positioning will be in what the ME coders think is generally a better defensive position slightly more often than a player with 15 pace... but maybe not if they're not concentrating or are too slow to get there and maybe the defensive position isn't that great anyway) and those other attributes also get modded down by mentality and consistency. I don't think there's any avoiding that. And I'm not sure it's possible or desirable for the game to not have certain types of players that are more effective than others despite them being judged to have lower ability and attacking tactics that tend to overachieve with small clubs. I do think there are specific situations like tackles where physical attributes make players too effective though (most notably terrible tacklers with good speed and agility will actually win the ball back plenty) Edited May 10, 2023 by enigmatic 3 Link to post Share on other sites More sharing options...
Sanassy Posted October 1, 2023 Share Posted October 1, 2023 Le 05/05/2023 à 10:21, shirajzl a dit : Pace, Acceleration, Balance and Agility have been the king in FM for years. And I mean as far back as CM/FM split. Sure, you can be successful with slow players if you know what you're doing, but a sure fire way for massively overachieving is getting fast players, regardless of league, club, tactics, morale management etc. Do you think these are the key attributes of the current meta? and how much minimum must a player have? Link to post Share on other sites More sharing options...
phnompenhandy Posted October 1, 2023 Share Posted October 1, 2023 25 minutes ago, Sanassy said: Do you think these are the key attributes of the current meta? and how much minimum must a player have? In a word. "depends". Do you mean English PL, Belgian league, Scotland, Vanarama South, Hong Kong? I just found and read through this thread. Seems to me one key word is missing, that is causing much confusion of language. What the advocates of this approach are looking for is"fun", and a simplistic route to winning all the time. There is a word for this - arcadey. Unfortunately the advocates choose to use a contradictory and misleading term - "realistic". If you want the arcade experience with game-breaking downloadable tactics and edited teams/players, by all means go ahead - it's all there for you. But if you're going to start claiming that approach is "realistic", then you're wasting everyone's time. Link to post Share on other sites More sharing options...
shirajzl Posted October 18, 2023 Share Posted October 18, 2023 It's certainly not realistic to pack the team with super fast players who are otherwise awful and overachieve astonishingly, but it's simply how it works in the game and has been working for years. Whether you want to use that fact or not, it's everyone's choice. 2 Link to post Share on other sites More sharing options...
MikaelS Posted October 19, 2023 Share Posted October 19, 2023 (edited) As long as I can remember, pace has always been my number 1 attribute to go for. This goes for most versions of CM/FM. When I adventured into lower league saves, pace was just the single thing you needed. Everything else didn't matter as much. Edited October 19, 2023 by MikaelS 2 Link to post Share on other sites More sharing options...
lied90 Posted April 3 Author Share Posted April 3 Hello again I did another one season test with the same premise as the opening post, I also changed the title to be less clickbaity by removing the part about mental/technical not being important. Removed all the players from Nottingham Forest. Replaced them with fast players with around 110-115 CA (lowest being 95, highest 116). There are ofc faster players, but I want to do it with low CA players because most teams can buy them easily and cheap. Slotted them in a gegenpress tactic and went on holiday. I returned a few times to remove unhappiness spreading because of players never being rotated. I also returned mid season to swap Jones from AML to ST, and Thomas-Asante the other way, just to see how Jones would do there. He did much worse than Asante. We had the lowest average in almost all attributes except pace. Highest performer was this guy playing AM, with 7.26 average rating. 15 goals and 10 assists across 42 matches. Some notable matches: 3 Link to post Share on other sites More sharing options...
Woo-Cash Posted April 18 Share Posted April 18 (edited) So far you've showed that gegenpress is king. I tried similar tests. I took the worst Polish league team (not much worse than other teams, just the worst). I changed all first team players acc/pace to 10, 12, 14, 16, 18 and 20. I've checked team strength with Genie Scout. With speed 10-12 the strength was the lowest and the team finished mostly last. With speed 14-16 the strength was medium and the team finished between 6 and 15. With speed 18-20 the strength was #6 in the league and the team finished 4-6th. It means that weights for acc/pace that I used in Genie was quite proper, the team finished very similar to its strength with these weights. And I used weights 35 for most important attributes, 25 for less important and 10 for the rest (as showed in roles in FM). So acc/pace had weights depending on the player's position. If your thesis was right, increasing acc/pace would give me much higher positions than predicted by Genie Scout. But it was on default/assman's tactics. Then I downloaded good gegenpress tactics. And guess what. The team almost won with acc/pace 10, when they should be last (third place with 2 points behind winner). And with acc/pace 12 (somewhere around average for the league) they won by 13 points, when they still should struggle for not being relegated. So, yes, acc/pace are important, they make players better, but not much more than other attributes. The most important thing is tactics. The AI doesn't use good tactics. Choose a good tactics and you will win with bad players. Choose an AI-bad tactics and acc/pace is not more important than other role-important attributes. If you want to prove something, don't use tactics that make you win without any effort. Edited April 18 by Woo-Cash Link to post Share on other sites More sharing options...
lied90 Posted April 21 Author Share Posted April 21 On 18/04/2024 at 16:07, Woo-Cash said: So far you've showed that gegenpress is king. I tried similar tests. I took the worst Polish league team (not much worse than other teams, just the worst). I changed all first team players acc/pace to 10, 12, 14, 16, 18 and 20. I've checked team strength with Genie Scout. With speed 10-12 the strength was the lowest and the team finished mostly last. With speed 14-16 the strength was medium and the team finished between 6 and 15. With speed 18-20 the strength was #6 in the league and the team finished 4-6th. It means that weights for acc/pace that I used in Genie was quite proper, the team finished very similar to its strength with these weights. And I used weights 35 for most important attributes, 25 for less important and 10 for the rest (as showed in roles in FM). So acc/pace had weights depending on the player's position. If your thesis was right, increasing acc/pace would give me much higher positions than predicted by Genie Scout. But it was on default/assman's tactics. Then I downloaded good gegenpress tactics. And guess what. The team almost won with acc/pace 10, when they should be last (third place with 2 points behind winner). And with acc/pace 12 (somewhere around average for the league) they won by 13 points, when they still should struggle for not being relegated. So, yes, acc/pace are important, they make players better, but not much more than other attributes. The most important thing is tactics. The AI doesn't use good tactics. Choose a good tactics and you will win with bad players. Choose an AI-bad tactics and acc/pace is not more important than other role-important attributes. If you want to prove something, don't use tactics that make you win without any effort. Using something other than gegenpress tactics for this wouldn't make any sense. Gegenpress requires the team to press (fast), and in general move fast, so it's suited for fast players. No point in having fast players who stand off and mark their zones instead of pressing, or dwell on the ball and play tiki taka instead of running with it. The initial post also wasn't meant to be absolute proof of anything other than how easy the game can be. My experience is that it's very easy to win the league in nations like Poland, Sweden, Norway etc, even with a team predicted to finish low. The reason is that the skill difference between teams is rarely that big, unlike the PL that has teams like Luton facing Man City. This is also why I tested in the PL, because the contrast between the overall skill level of my players and the other teams is so big. Remember that the players I used have 105-115 CA. Feel free to try using a gegenpress tactic in the PL with slow players that have 105-115 CA, that would be more comparable and interesting than what you did imo. Link to post Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now