delra said: I have a quick question. It's something I remembered from the old days of FM that I am not sure still applies to training in FM24.
Does it cost more CA to go from (for example) Acceleration 19->20 than it costs to go from Acceleration 9->10? And if it does, what's the ratio? If that was true, there'd be a point where it breaks even, when it's no longer good to keep pushing the physical attributes into 20s (when they are most expensive) and is more prudent to switch focus to something else, where you can buy multiple points of attributes for the same CA? So in my example, rather than going Acc 19->20, you could instead go for Dribbling 8->12 and gain a much better, more rounded up player? Expand
Spoiler
1.Extreme way to increase Physical class stats
[Rest]+[Double Intensity]+[Addtional Focus Quickness] or [Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
Explanation: "Recovery" is actually a variant of rest. If there is any other training, "rest" will be invalid, but "recovery" will not, "recovery" is counted as a training, so it can be used as a "weight" to pull the weight of "Pace" and "Acc" and so on to make it closer to the "rest" CA distribution, so the following 7 x recovery are added. It is used to pull the weight of the CA assigned to the Physical stats to pull up (The disadvantage of recovery is that it slightly increases the weight of a less useful physical attribute, such as strength, but this disadvantage can be ignored)
2.Increase moderate levels of CA , as fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Attacking Shadow Play]+[Recovery]x7+[Addtional Focus Quickness] or [Quickness]+[Attacking Direct]+[Recovery]x7+[Addtional Focus Quickness]
3. Add as much CA as possible , while fewer invalid attributes as possible, and as many high value attributes as possible
1. [Rest]+[Double Intensity]+[Addtional Focus Quickness] It sacrifices technical and mental growth and maximizes physical growth
3. [Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness] maximizing CA growth as possbile , avoids the growth of invalid attributes and increases the growth of valid attributes as much as possible
The difference between the two training schedules in the picture is actually that they choose "whether to sacrifice a small amount of physical attributes grow" to allow the technical and mental classes to get weight CA distribution and allow the technical and mental classes to grow
If it yes, it's "3", if it's no, it's "1", The middle solution is 2
Yarema said: Does it? I feel people seriously underestimate the power of CA growth for young players. Your 85 CA 18 year old physical freaks will not get any decent loans, you'll have to keep them at the club costing their further growth because of lack of competitive game time plus you'll want to guide their development. I find that it is unnecessary to cap CA growth, you can reach 20 pace and acceleration on "normal" schedules from these tests. Expand
animatron said: This shows that by far [Rest]+[Double Intensity]+[Addtional Focus Quickness] is the best training for Players under 19 (not including 19). Thank you for sharing this Expand
I think the method used is flexible according to the needs of the actual situation,
1. For example, players in a lower league, let's say that the average young player between the ages of 15 and 18 has only 100 PA, about 70 CA,
Then, the CA they can grow is "100-70=30". Since "30" is low.
They do not have enough "residual CA" to grow "technical and spiritual" at the same time, so training that is more inclined to physical attributes can make them ultimately have higher "combat effectiveness".
2. In contrast to the other situation, suppose that the player in a good club, 15-18 years old young players generally have 170 PA, about 70 CA,
Then, the CA they can grow is "170-70=100". Such a large number of "residual CA" is enough for players to develop in different types of attributes
That said, in this case, you may need to worry about whether the "100" remaining CA will be used up in time before the player age 25, If the 100 "surplus CA" is not used up in time, it is a waste of PA
3. another situation I mentioned on the 1st floor, or you can just look at the linked table:
physical class attributes that have a hard-coded scale/numerical upper limit.
For the same training schedule, at age 20, it increases the CA by 22.5, assuming an increase in Pace of 2-2.5 At age 18, it increases the CA by 30.5, again only increasing the Pace by 2-2.5 At age 15, it increases the CA by 29, again only increasing the Pace by 2-2.5 At the age of 23, it increases the CA by 20, again only increasing the Pace by 2-2.5
The above results are based on the test conditions of 30 matches. If you look at my previous post, there was a part of the test that was "0 matches played", and for that case, it also only increased the Pace by 2-2.5
So what happened to the extra CA? For example, at age 20, it increases CA by 22.5, and at age 18, it increases CA by 30.5. They differ by 30.5 - 22.5 = 8 As you can see in the table, the "8" ,they are mostly assigned to technical and mental attributes
In other words, the golden age of 15-18 years old , Training technique and mental class stats is far more effective, Training the physical attributes is also more effective, but only slightly increases the effect
Being much more effective doesn't mean you have to do it. Like I said, I think it's Adjust according to the actual situation
In my training schedule above, the growth of the physical is more independent of the number of match played, that is, even if there is no match at all, their physical part growth will not be significantly reduced By the same principle, the CA brought about by the increase in the number of matches is more allocated to the "technical and mental" attributes , and More match played than 30 will further increase the value of "30.5" for 17-18-year-olds in this picture
Immediately apologize about my English, all my text will be from a translator.
I came across @harvestgreen22's post. And decided that all the characteristics of players are important for the game, not just speed/agility/acceleration.
Went to see the original source where he got his information from:
by Evidence Based Football Manager.
His chart has stats for all types of drills and what the best ones are, he wrote them out as well: Attacking Defendinng Possession Overall (workouts that give more than 0.5 PA increase).
For goalkeepers: Ground defense Attacking overlap Chance creation
Based on this data, I decided to compose a training that should realize the potential of a player at 100% or close to these values. The results are sterile because there are no 2 matches per week, no 3 matches per week. Players do not leave for national teams.
Below are the variations of the training on which the tests were done: 1 2 2.1 3 4 5 5.1 6 7 8 9 10 11 12 13
As well as which individualized training sessions for the players:
Training intensity:
6 tests were done for ages 15-20.
Test 1-4. Initially each player has professionalism 20 and ambition 20 all other attributes 10. All indicators are also 10 CA 82\PA 200
The results were quite mediocre (in my opinion) at both ages, because I was looking for tactics that should realize the potential of the squad and not a particular player around 85-90%. The only optimal tactics were 2 and 5.
Test 5. Everything is the same as in test 1-4 for the players. I decided to refer to that blogger's table again to see where I went wrong and came up with a new tactic that should utilize all players and pump up their potential, which was tactic 13.
Training assignment: Team 1: only 13 tactics
2 team: only 2.1 tactics (it differs from 2 only by the load in the training week, so that the speed training is spread over the whole week and not on 1 day).
3 team: 2, 13 with alternating each week
4 team: 5.1, 2 with alternating each week (5.1 minor change to practice with goalie)
Logically, the best tactic is the 2nd tactic, because it realizes 93%, but in addition to the realization of potential, it was important to me that there was not too much slippage in some players, and others have too much growth. I would also like to point out that the player with the lowest number of matches has the lowest PA realization (AMC played only 2 games in the 4th team and 8 games in the 3rd team). For myself, I highlighted 13 and 2.1 as the best tactics,,but this is all under ideal conditions and with the professionalism of 20 and ambition of 20, I had to check how it would be when the attributes are all 10. And also how it works on players from 20-25 years old with professionalism 20 and ambition 20 and also when the same attributes are 10.
Test 6.
1 team 15-20 (20 prof/amb): 2 years - tactics 2.1, 2 years - tactics 13, 1 year - tactics 2.1.
2 team 15-20 (10 prof/amb): 2 years - tactics 13, 2 years - tactics 2.1, 1 year - tactics 13.
3 team 20-25 (20 prof/amb): alternate each year tactics 13\2.1.
4 team 20-25 (10 prof/amb): alternate every 4 weeks with tactic 13\2.1.
From the tests I have concluded that the realization of potential is affected quite strongly by professionalism and ambition, I can assume even if a player will have CA = 200, but his ambition and professionalism will be below 10, then this you will not be able to fully realize his performance even at least 60-70%.
I think the best tactics from tests 13 and 2.1 they can be varied under those players. which you have, and also under the game week.
I tested 13 and 2.1 , check it in H21 and I21 above The "Training intensity" and [Addtional Focus nn] of this test are based on the way you mentioned You can see the effect of each property directly
If you look through the entire table, you can see very clearly that the training is assigned CA due to their weight, but not "each training" give "a certain number of CA" corresponding of attribute
Among other things:
The H21 uses 12 training programs and provided CA of 34.7 The I21 uses 6 training programs and provides CA of 31.4 while The I20, uses only 3 training programs, also provides CA of 31
This is because the CA is assigned to the body class attribute under the default allocation, and the body class attribute has a "hard-coded upper limit", and this part of the extra allocation is wasted. Some training actually takes this "wasted weight" and pulls it back into the technical and mental class stats.
So, why does 34.7CA appear? Looking at the table H21, you can see that CA are allocated to some attributes such as "Decision", and they "occupy" more CA per 1 attribute, so the phenomenon is that CA increases
In other words, the idea of "Evidence Based Football Manager" is wrong If you like the technical and mental attributes, I can actually use 3-4 training, To achieve the effect of the 12 training of H21,
I maintain my view that there are some technical and mental attributes that take up a lot of CA and don't have much in-game effect, so I try to avoid assigning them growth weights as much as possible
[Rest]+[Double Intensity]+[Addtional Focus Quickness] or [Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
Explanation: "Recovery" is actually a variant of rest. If there is any other training, "rest" will be invalid, but "recovery" will not, "recovery" is counted as a training, so it can be used as a "weight" to pull the weight of "Pace" and "Acc" and so on to make it closer to the "rest" CA distribution, so the following 7 x recovery are added. It is used to pull the weight of the CA assigned to the Physical stats to pull up (The disadvantage of recovery is that it slightly increases the weight of a less useful physical attribute, such as strength, but this disadvantage can be ignored)
2.Increase moderate levels of CA , as fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Attacking Shadow Play]+[Recovery]x7+[Addtional Focus Quickness] or [Quickness]+[Attacking Direct]+[Recovery]x7+[Addtional Focus Quickness]
3. Add as much CA as possible , while fewer invalid attributes as possible, and as many high value attributes as possible
[Rest]+[Double Intensity]+[Addtional Focus Quickness] or [Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
Explanation: "Recovery" is actually a variant of rest. If there is any other training, "rest" will be invalid, but "recovery" will not, "recovery" is counted as a training, so it can be used as a "weight" to pull the weight of "Pace" and "Acc" and so on to make it closer to the "rest" CA distribution, so the following 7 x recovery are added. It is used to pull the weight of the CA assigned to the Physical stats to pull up (The disadvantage of recovery is that it slightly increases the weight of a less useful physical attribute, such as strength, but this disadvantage can be ignored)
2.Increase moderate levels of CA , as fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Attacking Shadow Play]+[Recovery]x7+[Addtional Focus Quickness] or [Quickness]+[Attacking Direct]+[Recovery]x7+[Addtional Focus Quickness]
3. Add as much CA as possible , while fewer invalid attributes as possible, and as many high value attributes as possible
lasko911 said: I did some parallel testing with Ajax, since they have a lot of good youngsters with high PA. In experiment A, I used V7 in first team, U21 and U19. In experiment B, I used assistant manager for training for all teams. At the end of the season, I took note of all CA growth + what attributes rose the most.
In B, when AI controls training, the CA growth is actually bigger. Players grow more rounded and they get more attributes increased. In A, with V7, they get less CA points growth, but ACC/PAC are slightly more increased than in B.
I'm a bit disappointed, to be honest, since I got more growth by simply not touching training. True, ACC/PAC with V7 over a few years are significantly higher, but overall players profit more with AI training. Expand
Maybe I didn't explain myself very well, Because there are strong and weak stats ,The origin purpose of this training was not to develop a balanced CA, or have max growth in CA , The purpose is to avoiding the waste of PA by "invalid attributes" as much as possible, and at the same time allowing "effective attributes" to grow as much as possible Some attributes, will take up a lot of PA, no effect or even negative effects, Such as Decision and Technique
I know people who have done this kind of work, and I just want to see what exact numerical effects it has on specific training programs
The age of 19 is a Key node. It slows down growth, The age of 27 is another Key node. Significantly reduced growth.
15-18 Age is the largest growth period, 17-18 Age peak growth. The graph on the leftis "[Rest]+[Double Intensity]+[Addtional Focus Quickness]" training, which still allows for Pace and Acceleration Grow at age 33 (at the expense of skill and spirit). At age 24, it will inevitably reduce total CA
The graph on the right, [Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness] is displayed. It represents maximizing the growth of CA , while avoiding the waste of PA by "invalid attributes" as much as possible, and at the same time allowing "effective attributes" to grow as much as possible It can't keep CA from declining at about age 35
At the age of 15-18, compared to the age of 20, the increase in CA can not be added to Physical Class Stats,it can only add more technical and mental Class Stats, because the physical class Stats have some kind of hard-coded upper limit per season.[Double Intensity]+[Addtional Focus Quickness] It unlocks this upper limit so that the corresponding stats, such as Pace, can grow by 2-3 times compared to "normal"
Professionalism 10. compared to Professionalism 20 It reduces growth by about 30-35%
If you need more information, you can see in excel
Default test condition It's only been tested once. Randomness introduces errors. All attributes 10, PA200, age 20, injury, dirty 1, Professionalism 20. The team consists of 1 goalkeeper, 2 centre-backs, 2 full-backs, 1 midfielders, 1 front midfielders, 2 wingers and 1 striker. The default distribution in the group, The defensive team has 2 centre-backs, 2 full-backs, 1 centre-back, The attacking team has 1 midfielders, 1 midfielders, 2 wingers and 1 striker. The coaching staff are all 5 stars, the training facilities are 5 stars, and 30 games are played. The number of matches and Professionalism had a great influence. [Double intensity]= training page, Rest page, training intensity scheduling, Set to "No pitch or gym work, No pitch or gym work, No pitch or gym work, double intensity, double intensity"
Orion said: What do you think about Team Cohesion? Asking because I've played a whole season with almost the same players and yet we have like 'neutral' for team cohesion. Would adding a 1 team bonding session to the training schedule mess up with the things? Expand
It won't have a big effect, it's just a slight decrease a very little bit of CA, you can just add it as you need
delra said: All these start with Quickness focus. What would happen if we switched away from Quickness to General->Physical, wouldn't that make players overall slower, but stronger in other attributes, especially give them more stamina and work rate to better survive high intensity tactics and more jumping reach to win more headers? Let's say could we test a version of I8, but change from Quickness to Physical? See how much Pace/Acc we sacrifice to get other physical attributes higher? Expand
check Z19, It's very close to the Quickness version Its biggest benefit is that it adds more Work Rate Because of Randomness,You can't tell the other difference, The difference in the previous tests: the bad side is a slight reduction in Pace and Acceleration
Default test condition It's only been tested once. Randomness introduces errors. All attributes 10, PA200, age 20, injury, dirty 1, Professionalism 20. The team consists of 1 goalkeeper, 2 centre-backs, 2 full-backs, 1 midfielders, 1 front midfielders, 2 wingers and 1 striker. The default distribution in the group, The defensive team has 2 centre-backs, 2 full-backs, 1 centre-back, The attacking team consists of 1 midfielders, 1 midfielders, 2 wingers and 1 striker. The coaching staff are all 5 stars, the training facilities are 5 stars, and 30 games are played. The number of matches and Professionalism had a great influence. [Double intensity]= training page, Rest page, training intensity scheduling, Set to "No pitch or gym work, No pitch or gym work, No pitch or gym work, double intensity, double intensity"
Agility ,Pace, Vision, Aerial Reach, Reflexes, Technique They are the 6 most effective goalkeeping attributes Other attributes, have some effect, but not as powerful as they are
How i set the test: All players have 10 attributes, locking various variables with FMrte, the league has 4 teams, Team A is using God of Chaos v1 (without Tacking - Get Stuck in), Opponents B, C, and D use 433 wing play, 433 control possession, and God of Chaos v1 standard tested ,Record conceded goal. Then change one of the attributes , increase it by 10 , Record the conceded goal after the change.
If after a certain attribute +10, it leads to a significant reduction in the number of goals conceded, that means that this attribute has a great help in reducing the number of goals conceded, it is a very effective attribute I should actually test another set of -10 attribute , but I did't do that to save time
Test data(I didn't translate it to save time)： https://pixeldrain.com/u/RismZb3k Standard error: all-attribute of "goals conceded" 's standard deviation is adjacent. The standard deviation of the "goals conceded" is relatively small, so it can be more accurate with fewer tests. Mathematically, "900 games, 30 seasons" reduces the "standard error" to about "1.1" "600 games, 20 seasons" the "standard error" is about "1.35" "450 games, 15 seasons" the "standard error" is about "1.55" There is still a margin of RNG error, but I think it is acceptable (save time).
It would be more accurate to measure it by goal difference, But the standard deviation of goal difference is larger = it requires a larger sample size After have increased sample size enough (which takes a lot of time and won't result quickly), consider switching to goal difference
babasalat said: I´m training with [Quickness]+[Match Practice]+[Recovery]x7+[Attacking]+[Addtional Focus Quickness]+[Double Intensity] in my u19 with maxed out facilities, good youth trainers, good training ratings and Professionalism between 11 and 16 but my players dont even get 10 CA per season. Some player dont even got 1 Acc or Pace after 2 seasons... Expand
I mentioned it on the first floor (if not, it might have been missed when the translator was translating).
To highlight the differences between different training programs, Before testing, the attributes of all players were set to 10, age 20, potential 200PA, Professionalism 20 And in the top column of the table below, "30" represents 30 games per season. These conditions all increase the amount of CA a player can get "per season". These CAs are then assigned to different training plans.
In other words, if your players do not meet these conditions, the expected effect will be reduced. And if your player's condition is better than that, like Z10 in the table, it changes the condition to 17 years old, so it gets 8 more CA points
This test setting is mainly used to control the variables, and the actual effect you get will vary according to the actual situation
twkmax said: Just looking at the raw results in your most recent post.
By my calculations C12, J11, K10, Q10 are the best options based on how much useful CA they generate.
(@harvestgreen22 I'm not quite sure why you are focussing on E12 and U11?)
Based on my analysis you can pick one of those 4 I listed based on how extreme you want to go with focussing on only developing the 'impactful' attributes.
Most Extreme -> C12 -> J11 -> K10 -> Q10 -> Least Extreme
With C12 and J11 you will lose attributes in other places.
I would say the most 'balanced' option is definitely K10.
(also @harvestgreen22 why did you at some point you switch from 'attacking wings' to 'attacking shadow play'?) Expand
You can choose "Most extreme increase in body Attributes", "Intermediate", "Most extreme Increase in CA" You are free to choose this, according to your needs, your opinion is Good
And then the change, I forget, was someone in our community commented and asked for more of a certain attribute, He convinced me, so I compared the two of them and chose the one he asked for with more attributes You can compare the figures in the table and choose the one you need
flob said: Hey all, been reading all comments and would like to ask if it's okay, in my situation, to change to one of these trainings during a season. Or would you say it's better to wait till the start of a season?
I would also like to ask, to make sure I understand, that with individual training I don't select the player to it's position in my tactic anymore, but leave it at default? Did I understand that correctly? Expand
you can change it
with individual training I don't select the player to it's position in my tactic anymore, but leave it at default? Yes
This will maximize the blue highlighting Stats. If you accidentally choose one, don't worry too much, just lose some of the blue highlights and add them to the green
Yarema said: Would it be possible to test some other additional focuses? Like strength for jumping reach or ball control for dribbling, maybe others. For example with V7 schedule or whatever people most commonly use except with a different additional focus, and you can probably use same metrics as you already do.
I ask because at a certain point one might prefer other attributes than pure speed. For example center backs don't seem to develop much jumping reach and strength using these schedules, or maybe they do and I've just been unlucky. I'd rather have a 14 pace 14 acc 14 jumping reach center back than 16, 16 and 10. 10 vs 14 is probably too big of a gap, but just to illustrate the point. Expand
Yes, you can set the properties you need. It works the same way, transferring a certain percentage of growth weigh to it
I can do "another focus" test next time , You can give an example of which focus you need to test (For example with V7 schedule + xxx focus)
vdowning36 said: @harvestgreen22 Thank you for all the work you're doing, it's much appreciated. If you have time, would you be able to test: Quickness + Physical + Additional training quickness + double intensity? Expand
mmigueis said: It seems pretty conclusive, IMO! (Note that in comparing the "technique = 20" team to the "technique = 1" teams, you could average across all the "technique = 1" teams, as they are set up exactly the same).
Have you ever done similar tests for other attributes? It would be very interesting to see the results. Expand
bigloser said: In the control group did you lower their technique? By default nearly all the Man City players have over 10 technique so you wouldn't be able to raise it by 10.
In the Machine Learning weights technique was only moderately weighted high on DM(RPM), AMC, ST. One theory is it's bad to have defenders with high technique/flair. The attribute table from 2022 is basically the same as today.
Or another is technique (decisions/vision/passing?) is related to the success of certain traits like tries killer passes/curls ball/round the keeper/shoots with power/places shots/etc. And you would need said traits to get any benefit. But that doesn't explain why it's negative as you would expect it to be neutral. In the Man City example Haaland only has "tries first time shots" related to technique. This is the only thing I can think of.
The most likely scenario is technique was good at some point in previous editions but other changes made in the engine over the years has unintentionally made it bad. Expand
As before, I did 10 seasons (this means 300 matches). The configuration is the same as the other comment.
Databattle 1 has ALL attributes on 20. The other teams have ALL attributes on 20, EXCEPT Technique which is on 1.
Hypothesis: IF Technique doesn't matter, what we expect is to that results of the league will be random or close to random. IF Technique is NEGATIVE, what we expect is that Databattle 1 will be WORSE than the other teams overall.
Results: Databattle 1 overperformed all other teams in most seasons, having better GD, Position, winrate and more points most of the time. It won the league 7 times (7/10), and has higher mean points, GD, wins, and position. See images for more info (On the correlation matrix, the "Technique 20" data point is simply 1 if the team is Databattle 1, and 0 if it isn't. Also, remember that POSITION is "inversed", so having lower position is actually better). Even in the seasons it did not win, it came second by AT MOST 4 points.
Conclusion: When everything else is at 20, is seems to be better to have technique at 20 than it is to have it at 1. Eventually I'll add more data to this analysis, as I feel 10 tournaments/300 matches is a small amount, but for now I'll test other attributes in the same manner. Expand
The latest test results , with 20,000+ match tested
1. The results contradicted my initial judgment, In some cases, Technique and Flair will be positive. It is not known which stage setting/instruction produced this effect. in the 11 tactic have been measured, Technique and Flair are negative in more cases ,and positive in few.
2. With fewer sample, the standard deviation of Preset 532 Catenaccio is decrease accordingly to Flair increase, which may be caused by insufficient test samples
3. Based on the average standard deviation of all tactic ,estimate, it is not convergent to speculate on the results of one or two seasons It take about 20-25 seasons , 600-750 matchs each situation to have a Reasonable Confidence Intervals
4. To reduce distractions, Three of the identical tested tactic were randomly replaced with different opponent tactic, The result is the same . Therefore, judging "positive or negative" does not depend on the opponent's formation
You can, They differ very little (example, https://pixeldrain.com/u/tytFZ98Y , I17 and J17) and don't lose too many attributes
28/12/2024 update (colorize)
Does it cost more CA to go from (for example) Acceleration 19->20 than it costs to go from Acceleration 9->10? And if it does, what's the ratio?
If that was true, there'd be a point where it breaks even, when it's no longer good to keep pushing the physical attributes into 20s (when they are most expensive) and is more prudent to switch focus to something else, where you can buy multiple points of attributes for the same CA? So in my example, rather than going Acc 19->20, you could instead go for Dribbling 8->12 and gain a much better, more rounded up player?
https://pixeldrain.com/u/5RA6S88y
i tested it , check J21,K21,L21 above
Yes, its growth will decline, and we have the flexibility to adjust according to the actual situation and needed
[Rest]+[Double Intensity]+[Addtional Focus Quickness]
or
[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
Explanation:
"Recovery" is actually a variant of rest. If there is any other training, "rest" will be invalid, but "recovery" will not, "recovery" is counted as a training, so it can be used as a "weight" to pull the weight of "Pace" and "Acc" and so on to make it closer to the "rest" CA distribution,
so the following 7 x recovery are added. It is used to pull the weight of the CA assigned to the Physical stats to pull up
(The disadvantage of recovery is that it slightly increases the weight of a less useful physical attribute, such as strength, but this disadvantage can be ignored)
2.Increase moderate levels of CA , as fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Attacking Shadow Play]+[Recovery]x7+[Addtional Focus Quickness]
or
[Quickness]+[Attacking Direct]+[Recovery]x7+[Addtional Focus Quickness]
3. Add as much CA as possible , while fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
or
[Physical]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
1.
[Rest]+[Double Intensity]+[Addtional Focus Quickness]
It sacrifices technical and mental growth and maximizes physical growth
3.
[Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
maximizing CA growth as possbile , avoids the growth of invalid attributes and increases the growth of valid attributes as much as possible
The difference between the two training schedules in the picture is actually that they choose
"whether to sacrifice a small amount of physical attributes grow" to allow the technical and mental classes to get weight CA distribution and allow the technical and mental classes to grow
If it yes, it's "3", if it's no, it's "1",
The middle solution is 2
animatron said: This shows that by far [Rest]+[Double Intensity]+[Addtional Focus Quickness] is the best training for Players under 19 (not including 19). Thank you for sharing this
Excel Training English 2
https://pixeldrain.com/u/R3imL2wX
I think the method used is flexible according to the needs of the actual situation,
1.
For example, players in a lower league, let's say that the average young player between the ages of 15 and 18 has only 100 PA, about 70 CA,
Then, the CA they can grow is "100-70=30".
Since "30" is low.
They do not have enough "residual CA" to grow "technical and spiritual" at the same time, so training that is more inclined to physical attributes can make them ultimately have higher "combat effectiveness".
2.
In contrast to the other situation, suppose that the player in a good club, 15-18 years old young players generally have 170 PA, about 70 CA,
Then, the CA they can grow is "170-70=100".
Such a large number of "residual CA" is enough for players to develop in different types of attributes
That said, in this case,
you may need to worry about whether the "100" remaining CA will be used up in time before the player age 25,
If the 100 "surplus CA" is not used up in time, it is a waste of PA
3.
another situation I mentioned on the 1st floor, or you can just look at the linked table:
physical class attributes that have a hard-coded scale/numerical upper limit.
For the same training schedule, at age 20, it increases the CA by 22.5, assuming an increase in Pace of 2-2.5
At age 18, it increases the CA by 30.5, again only increasing the Pace by 2-2.5
At age 15, it increases the CA by 29, again only increasing the Pace by 2-2.5
At the age of 23, it increases the CA by 20, again only increasing the Pace by 2-2.5
The above results are based on the test conditions of 30 matches.
If you look at my previous post, there was a part of the test that was "0 matches played", and for that case, it also only increased the Pace by 2-2.5
So what happened to the extra CA?
For example, at age 20, it increases CA by 22.5, and at age 18, it increases CA by 30.5.
They differ by 30.5 - 22.5 = 8
As you can see in the table, the "8" ,they are mostly assigned to technical and mental attributes
In other words, the golden age of 15-18 years old ,
Training technique and mental class stats is far more effective,
Training the physical attributes is also more effective, but only slightly increases the effect
Being much more effective doesn't mean you have to do it. Like I said, I think it's Adjust according to the actual situation
In my training schedule above, the growth of the physical is more independent of the number of match played, that is, even if there is no match at all, their physical part growth will not be significantly reduced
By the same principle, the CA brought about by the increase in the number of matches is more allocated to the "technical and mental" attributes ,
and More match played than 30 will further increase the value of "30.5" for 17-18-year-olds in this picture
Immediately apologize about my English, all my text will be from a translator.
I came across @harvestgreen22's post.
And decided that all the characteristics of players are important for the game, not just speed/agility/acceleration.
Went to see the original source where he got his information from:
by Evidence Based Football Manager.
His chart has stats for all types of drills and what the best ones are, he wrote them out as well:
Attacking
Defendinng
Possession
Overall
(workouts that give more than 0.5 PA increase).
For goalkeepers:
Ground defense
Attacking overlap
Chance creation
Based on this data, I decided to compose a training that should realize the potential of a player at 100% or close to these values.
The results are sterile because there are no 2 matches per week, no 3 matches per week. Players do not leave for national teams.
Below are the variations of the training on which the tests were done:
1
2
2.1
3
4
5
5.1
6
7
8
9
10
11
12
13
As well as which individualized training sessions for the players:
Training intensity:
6 tests were done for ages 15-20.
Test 1-4. Initially each player has professionalism 20 and ambition 20 all other attributes 10. All indicators are also 10
CA 82\PA 200
The results were quite mediocre (in my opinion) at both ages, because I was looking for tactics that should realize the potential of the squad and not a particular player around 85-90%.
The only optimal tactics were 2 and 5.
Test 5. Everything is the same as in test 1-4 for the players. I decided to refer to that blogger's table again to see where I went wrong and came up with a new tactic that should utilize all players and pump up their potential, which was tactic 13.
Training assignment:
Team 1: only 13 tactics
2 team: only 2.1 tactics (it differs from 2 only by the load in the training week, so that the speed training is spread over the whole week and not on 1 day).
3 team: 2, 13 with alternating each week
4 team: 5.1, 2 with alternating each week (5.1 minor change to practice with goalie)
1 team:
GK - 195
DCR - 194
DCL - 193
WBR - 175
WBL - 175
DM - 175
MC - 175
WL - 189
WR - 198
ACM - 156
ST - 192
Realization of potential = 91.5%
2 team:
GK - 175
DCR - 182
DCL - 191
WBR - 188
WBL - 185
DM - 191
MC - 195
WL - 175
WR - 175
AMC - 196
ST - 198
Realization of potential = 93%
3 team:
GK - 175
DCR - 189
DCL - 186
WBR - 193
WBL - 195
DM - 190
MC - 189
WL - 189
WR - 175
AMC - 164 (8 games)
ST - 175
Realization of potential = 91.5%
4 team:
GK - 185
DCR - 191
DCL - 187
WBR - 187
WBL - 195
DM - 199
MC - 199
WL - 176
WR - 192
AMC - 156 (2 games)
ST - 175
Realization of potential = 92.5%
Logically, the best tactic is the 2nd tactic, because it realizes 93%, but in addition to the realization of potential, it was important to me that there was not too much slippage in some players, and others have too much growth. I would also like to point out that the player with the lowest number of matches has the lowest PA realization (AMC played only 2 games in the 4th team and 8 games in the 3rd team).
For myself, I highlighted 13 and 2.1 as the best tactics,,but this is all under ideal conditions and with the professionalism of 20 and ambition of 20, I had to check how it would be
when the attributes are all 10.
And also how it works on players from 20-25 years old with professionalism 20 and ambition 20 and also when the same attributes are 10.
Test 6.
1 team 15-20 (20 prof/amb): 2 years - tactics 2.1, 2 years - tactics 13, 1 year - tactics 2.1.
2 team 15-20 (10 prof/amb): 2 years - tactics 13, 2 years - tactics 2.1, 1 year - tactics 13.
3 team 20-25 (20 prof/amb): alternate each year tactics 13\2.1.
4 team 20-25 (10 prof/amb): alternate every 4 weeks with tactic 13\2.1.
1 team:
GK - 184
DCR - 190
DCL - 187
WBR - 175
WBL - 175
DM - 175
MC - 175
WL - 189
WR - 199
AMC - 189
ST - 192
CA realization = 92%
2 team:
GK - 144
DCR - 164
DCL - 151
WBR - 164
WBL - 150
DM - 177
MC - 160
WL - 169
WR - 157
AMC - 166
ST - 166
CA realization = 80%
3 team:
GK - 171
DCR - 169
DCL - 170
WBR - 177
WRL - 174
DM - 180
MC - 183
WL - 178
WR - 176
AMC - 179
ST - 179
CA realization = 88%
4 team:
GK - 145
DCR - 155
DCL - 150
WBR - 149
WBL - 161
DM - 160
MC - 150
WL - 162
WR - 156
AMC - 161
ST - 164
CA realization = 77.5%
From the tests I have concluded that the realization of potential is affected quite strongly by professionalism and ambition, I can assume even if a player will have CA = 200, but his ambition and professionalism will be below 10, then this you will not be able to fully realize his performance even at least 60-70%.
I think the best tactics from tests 13 and 2.1 they can be varied under those players. which you have, and also under the game week.
I throw a folder with files here changes in the players (not all players) on the tests of training 1-4 and 5, as well as saving (1-6) of these tests FM24, where you can personally see how the players have progressed in each tactic:
https://drive.google.com/drive/folders/1mwjIa36Jg9CCrZnr0c5-58aHt_0H7CfY?usp=drive_link
https://pixeldrain.com/u/2vaLuuU2
I tested 13 and 2.1 , check it in H21 and I21 above
The "Training intensity" and [Addtional Focus nn] of this test are based on the way you mentioned
You can see the effect of each property directly
If you look through the entire table, you can see very clearly that
the training is assigned CA due to their weight,
but not "each training" give "a certain number of CA" corresponding of attribute
Among other things:
The H21 uses 12 training programs and provided CA of 34.7
The I21 uses 6 training programs and provides CA of 31.4
while
The I20, uses only 3 training programs, also provides CA of 31
This is because the CA is assigned to the body class attribute under the default allocation, and the body class attribute has a "hard-coded upper limit", and this part of the extra allocation is wasted.
Some training actually takes this "wasted weight" and pulls it back into the technical and mental class stats.
So, why does 34.7CA appear?
Looking at the table H21, you can see that CA are allocated to some attributes such as "Decision", and they "occupy" more CA per 1 attribute, so the phenomenon is that CA increases
In other words, the idea of "Evidence Based Football Manager" is wrong
If you like the technical and mental attributes, I can actually use 3-4 training, To achieve the effect of the 12 training of H21,
I maintain my view that there are some technical and mental attributes that take up a lot of CA and don't have much in-game effect, so I try to avoid assigning them growth weights as much as possible
1.Extreme way to increase Physical class stats
[Rest]+[Double Intensity]+[Addtional Focus Quickness]
or
[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
Explanation:
"Recovery" is actually a variant of rest. If there is any other training, "rest" will be invalid, but "recovery" will not, "recovery" is counted as a training, so it can be used as a "weight" to pull the weight of "Pace" and "Acc" and so on to make it closer to the "rest" CA distribution,
so the following 7 x recovery are added. It is used to pull the weight of the CA assigned to the Physical stats to pull up
(The disadvantage of recovery is that it slightly increases the weight of a less useful physical attribute, such as strength, but this disadvantage can be ignored)
2.Increase moderate levels of CA , as fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Attacking Shadow Play]+[Recovery]x7+[Addtional Focus Quickness]
or
[Quickness]+[Attacking Direct]+[Recovery]x7+[Addtional Focus Quickness]
3. Add as much CA as possible , while fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
or
[Physical]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
1.Extreme way to increase Physical class stats
[Rest]+[Double Intensity]+[Addtional Focus Quickness]
or
[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
Explanation:
"Recovery" is actually a variant of rest. If there is any other training, "rest" will be invalid, but "recovery" will not, "recovery" is counted as a training, so it can be used as a "weight" to pull the weight of "Pace" and "Acc" and so on to make it closer to the "rest" CA distribution,
so the following 7 x recovery are added. It is used to pull the weight of the CA assigned to the Physical stats to pull up
(The disadvantage of recovery is that it slightly increases the weight of a less useful physical attribute, such as strength, but this disadvantage can be ignored)
2.Increase moderate levels of CA , as fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Attacking Shadow Play]+[Recovery]x7+[Addtional Focus Quickness]
or
[Quickness]+[Attacking Direct]+[Recovery]x7+[Addtional Focus Quickness]
3. Add as much CA as possible , while fewer invalid attributes as possible, and as many high value attributes as possible
[Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
or
[Physical]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness]
Excel Training English 2
edit 27/12/2024 :
https://pixeldrain.com/u/R3imL2wX
In experiment A, I used V7 in first team, U21 and U19.
In experiment B, I used assistant manager for training for all teams.
At the end of the season, I took note of all CA growth + what attributes rose the most.
In B, when AI controls training, the CA growth is actually bigger. Players grow more rounded and they get more attributes increased.
In A, with V7, they get less CA points growth, but ACC/PAC are slightly more increased than in B.
I'm a bit disappointed, to be honest, since I got more growth by simply not touching training. True, ACC/PAC with V7 over a few years are significantly higher, but overall players profit more with AI training.
Maybe I didn't explain myself very well, Because there are strong and weak stats ,The origin purpose of this training was not to develop a balanced CA, or have max growth in CA ,
The purpose is to avoiding the waste of PA by "invalid attributes" as much as possible, and at the same time allowing "effective attributes" to grow as much as possible
Some attributes, will take up a lot of PA, no effect or even negative effects, Such as Decision and Technique
I know people who have done this kind of work, and I just want to see what exact numerical effects it has on specific training programs
The age of 19 is a Key node. It slows down growth,
The age of 27 is another Key node. Significantly reduced growth.
15-18 Age is the largest growth period, 17-18 Age peak growth.
The graph on the leftis "[Rest]+[Double Intensity]+[Addtional Focus Quickness]" training, which still allows for Pace and Acceleration Grow at age 33 (at the expense of skill and spirit).
At age 24, it will inevitably reduce total CA
The graph on the right, [Quickness]+[Match Practice]+[Attacking]+[Recovery]x7+[Double Intensity]+[Addtional Focus Quickness] is displayed.
It represents maximizing the growth of CA , while avoiding the waste of PA by "invalid attributes" as much as possible, and at the same time allowing "effective attributes" to grow as much as possible
It can't keep CA from declining at about age 35
At the age of 15-18, compared to the age of 20, the increase in CA can not be added to Physical Class Stats,it can only add more technical and mental Class Stats,
because the physical class Stats have some kind of hard-coded upper limit per season.[Double Intensity]+[Addtional Focus Quickness] It unlocks this upper limit so that the corresponding stats, such as Pace, can grow by 2-3 times compared to "normal"
Professionalism 10.
compared to
Professionalism 20
It reduces growth by about 30-35%
If you need more information, you can see in excel
Excel (data)
https://pixeldrain.com/u/L9e3LDBe
Testing File
https://pixeldrain.com/u/ok5kUjLN
Default test condition
It's only been tested once. Randomness introduces errors.
All attributes 10, PA200, age 20, injury, dirty 1, Professionalism 20.
The team consists of 1 goalkeeper, 2 centre-backs, 2 full-backs, 1 midfielders, 1 front midfielders, 2 wingers and 1 striker.
The default distribution in the group,
The defensive team has 2 centre-backs, 2 full-backs, 1 centre-back,
The attacking team has 1 midfielders, 1 midfielders, 2 wingers and 1 striker.
The coaching staff are all 5 stars, the training facilities are 5 stars, and 30 games are played.
The number of matches and Professionalism had a great influence.
[Double intensity]= training page, Rest page, training intensity scheduling, Set to "No pitch or gym work, No pitch or gym work, No pitch or gym work, double intensity, double intensity"
edit 27/12/2024↓↓↓↓↓↓↓↓
Excel Training English 2
https://pixeldrain.com/u/R3imL2wX
edit 27/12/2024↑↑↑↑↑↑↑↑
Asking because I've played a whole season with almost the same players and yet we have like 'neutral' for team cohesion. Would adding a 1 team bonding session to the training schedule mess up with the things?
It won't have a big effect, it's just a slight decrease a very little bit of CA, you can just add it as you need
Excel
https://pixeldrain.com/u/L9e3LDBe
check Z19,
It's very close to the Quickness version
Its biggest benefit is that it adds more Work Rate
Because of Randomness,You can't tell the other difference,
The difference in the previous tests: the bad side is a slight reduction in Pace and Acceleration
Default test condition
It's only been tested once. Randomness introduces errors.
All attributes 10, PA200, age 20, injury, dirty 1, Professionalism 20.
The team consists of 1 goalkeeper, 2 centre-backs, 2 full-backs, 1 midfielders, 1 front midfielders, 2 wingers and 1 striker.
The default distribution in the group,
The defensive team has 2 centre-backs, 2 full-backs, 1 centre-back,
The attacking team consists of 1 midfielders, 1 midfielders, 2 wingers and 1 striker.
The coaching staff are all 5 stars, the training facilities are 5 stars, and 30 games are played.
The number of matches and Professionalism had a great influence.
[Double intensity]= training page, Rest page, training intensity scheduling, Set to "No pitch or gym work, No pitch or gym work, No pitch or gym work, double intensity, double intensity"
Agility ,Pace, Vision, Aerial Reach, Reflexes, Technique
They are the 6 most effective goalkeeping attributes
Other attributes, have some effect, but not as powerful as they are
How i set the test:
All players have 10 attributes, locking various variables with FMrte, the league has 4 teams,
Team A is using God of Chaos v1 (without Tacking - Get Stuck in), Opponents B, C, and D use 433 wing play, 433 control possession, and God of Chaos v1
standard tested ,Record conceded goal.
Then change one of the attributes , increase it by 10 , Record the conceded goal after the change.
If after a certain attribute +10, it leads to a significant reduction in the number of goals conceded, that means that this attribute has a great help in reducing the number of goals conceded, it is a very effective attribute
I should actually test another set of -10 attribute , but I did't do that to save time
Test data(I didn't translate it to save time)：
https://pixeldrain.com/u/RismZb3k
Standard error:
all-attribute of "goals conceded" 's standard deviation is adjacent.
The standard deviation of the "goals conceded" is relatively small, so it can be more accurate with fewer tests.
Mathematically, "900 games, 30 seasons" reduces the "standard error" to about "1.1"
"600 games, 20 seasons" the "standard error" is about "1.35"
"450 games, 15 seasons" the "standard error" is about "1.55"
There is still a margin of RNG error, but I think it is acceptable (save time).
It would be more accurate to measure it by goal difference,
But the standard deviation of goal difference is larger = it requires a larger sample size
After have increased sample size enough (which takes a lot of time and won't result quickly),
consider switching to goal difference
What i mean with player traits is , example : Run with ball through centre , Play one-two ETC
I don't know,
have not do any test about the effects of traits .
And what happens when train with trait
I mentioned it on the first floor (if not, it might have been missed when the translator was translating).
To highlight the differences between different training programs,
Before testing, the attributes of all players were set to 10, age 20, potential 200PA, Professionalism 20
And in the top column of the table below, "30" represents 30 games per season.
These conditions all increase the amount of CA a player can get "per season". These CAs are then assigned to different training plans.
In other words, if your players do not meet these conditions, the expected effect will be reduced.
And if your player's condition is better than that, like Z10 in the table, it changes the condition to 17 years old, so it gets 8 more CA points
This test setting is mainly used to control the variables, and the actual effect you get will vary according to the actual situation
By my calculations C12, J11, K10, Q10 are the best options based on how much useful CA they generate.
(@harvestgreen22 I'm not quite sure why you are focussing on E12 and U11?)
Based on my analysis you can pick one of those 4 I listed based on how extreme you want to go with focussing on only developing the 'impactful' attributes.
Most Extreme -> C12 -> J11 -> K10 -> Q10 -> Least Extreme
With C12 and J11 you will lose attributes in other places.
I would say the most 'balanced' option is definitely K10.
(also @harvestgreen22 why did you at some point you switch from 'attacking wings' to 'attacking shadow play'?)
You can choose "Most extreme increase in body Attributes", "Intermediate", "Most extreme Increase in CA"
You are free to choose this, according to your needs,
your opinion is Good
And then the change, I forget, was someone in our community commented and asked for more of a certain attribute,
He convinced me, so I compared the two of them and chose the one he asked for with more attributes
You can compare the figures in the table and choose the one you need
I am busy recently, so i can't reply in time
I would also like to ask, to make sure I understand, that with individual training I don't select the player to it's position in my tactic anymore, but leave it at default? Did I understand that correctly?
you can change it
with individual training I don't select the player to it's position in my tactic anymore, but leave it at default?
Yes
This will maximize the blue highlighting Stats.
If you accidentally choose one, don't worry too much, just lose some of the blue highlights and add them to the green
I ask because at a certain point one might prefer other attributes than pure speed. For example center backs don't seem to develop much jumping reach and strength using these schedules, or maybe they do and I've just been unlucky. I'd rather have a 14 pace 14 acc 14 jumping reach center back than 16, 16 and 10. 10 vs 14 is probably too big of a gap, but just to illustrate the point.
Yes, you can set the properties you need. It works the same way, transferring a certain percentage of growth weigh to it
I can do "another focus" test next time ,
You can give an example of which focus you need to test (For example with V7 schedule + xxx focus)
https://pixeldrain.com/u/G1aBqzR3
It's in G12
You can compare it to I11, J11
Roughly :
J11:[Quickness] has slightly more Pace(2.24) and Acceleration, lower Work rate (0.64)
I11:[Physical] has slightly less Pace(2.03) and Acceleration, higher Work rate (0.94)
G12:[Quickness]+[Physical] is intermediate, Pace(2.06), Work rate (0.91)
for Physical class Stats , There is an upper limit of proportion or number
Have you ever done similar tests for other attributes? It would be very interesting to see the results.
bigloser said: In the control group did you lower their technique? By default nearly all the Man City players have over 10 technique so you wouldn't be able to raise it by 10.
In the Machine Learning weights technique was only moderately weighted high on DM(RPM), AMC, ST. One theory is it's bad to have defenders with high technique/flair. The attribute table from 2022 is basically the same as today.
Or another is technique (decisions/vision/passing?) is related to the success of certain traits like tries killer passes/curls ball/round the keeper/shoots with power/places shots/etc. And you would need said traits to get any benefit. But that doesn't explain why it's negative as you would expect it to be neutral. In the Man City example Haaland only has "tries first time shots" related to technique. This is the only thing I can think of.
The most likely scenario is technique was good at some point in previous editions but other changes made in the engine over the years has unintentionally made it bad.
tolec said: isn't this similar to the player attribution table? https://fm-arena.com/table/26-player-attributes-testing/
Slightly fewer matches but Techniques is slightly negative while flair is slightly positive
tolec said: isn't this similar to the player attribution table? https://fm-arena.com/table/26-player-attributes-testing/
Slightly fewer matches but Techniques is slightly negative while flair is slightly positive
caffeiner said: Just a small update on my findings:
As before, I did 10 seasons (this means 300 matches). The configuration is the same as the other comment.
Databattle 1 has ALL attributes on 20. The other teams have ALL attributes on 20, EXCEPT Technique which is on 1.
Hypothesis: IF Technique doesn't matter, what we expect is to that results of the league will be random or close to random. IF Technique is NEGATIVE, what we expect is that Databattle 1 will be WORSE than the other teams overall.
Results: Databattle 1 overperformed all other teams in most seasons, having better GD, Position, winrate and more points most of the time. It won the league 7 times (7/10), and has higher mean points, GD, wins, and position. See images for more info (On the correlation matrix, the "Technique 20" data point is simply 1 if the team is Databattle 1, and 0 if it isn't. Also, remember that POSITION is "inversed", so having lower position is actually better). Even in the seasons it did not win, it came second by AT MOST 4 points.
Conclusion: When everything else is at 20, is seems to be better to have technique at 20 than it is to have it at 1. Eventually I'll add more data to this analysis, as I feel 10 tournaments/300 matches is a small amount, but for now I'll test other attributes in the same manner.
The latest test results , with 20,000+ match tested
1.
The results contradicted my initial judgment,
In some cases, Technique and Flair will be positive.
It is not known which stage setting/instruction produced this effect.
in the 11 tactic have been measured, Technique and Flair are negative in more cases ,and positive in few.
2.
With fewer sample, the standard deviation of Preset 532 Catenaccio is decrease accordingly to Flair increase, which may be caused by insufficient test samples
3.
Based on the average standard deviation of all tactic ,estimate, it is not convergent to speculate on the results of one or two seasons
It take about 20-25 seasons , 600-750 matchs each situation to have a Reasonable Confidence Intervals
4.
To reduce distractions,
Three of the identical tested tactic were randomly replaced with different opponent tactic, The result is the same . Therefore, judging "positive or negative" does not depend on the opponent's formation