Let's assume that the true score of some tactic is 60 pts.
You test it for 5,760 matches and get 61 pts.
Then you re-rest it for 5,760 matches again and get 59 pts.
So you can conclude the RNG swing of a 5,760 matches test can be as high as 1 point ( -/+ 1 point away from the true score).
Btw, our RNG measure was for FM22 and it seems for FM23 the RNG has slight increased so it could be 1.5 points for a 5,760 matches test. Expand
Have you thought about using something like a Monte Carlo sampling on the tests? Basically, take smaller samples of the 5760 matches repeated times, then make statistics based on the statistic of those samples (for example, taking the medians of those subsamples, and calculating the median of them).
It's not hard to implement, and it can increase the precision of tests without increasing the computational effort used. There are also other types of subsampling, including bootstrapping and other techniques.
CBP87 said: Sorry @Zippo slightly confused, so are you saying that even playing 5760 matches that RNG swing is greater than initially thought (+1 either way) Expand
One can go high, the other low, then that makes two points of difference.
Trying to find a better balance between attack and defense. Defensive midfielders are set to participate more in attacking actions, and tactic is also set to play shorter passing, as advised here, and I removed counter-press to avoid counter-attacks.
Trying to find a better balance between attack and defense. Wingers are set to participate more in defensive play, while defensive midfielder is set to participate more in attacking actions, but without leaving much his position. The tactic is also set to play shorter passing, as advised here, and I removed counter-press to avoid counter-attacks.
Trying to find a better balance between attack and defense. Wingers are set to participate more in defensive play, while defensive midfielder is set to participate more in attacking actions, but without leaving much his position.
Trying to find a better balance between attack and defense. Wingers are set to participate more in defensive play, while defensive midfielders are set to participate more in attacking actions. This is just like 6.2, but more direct and counter-pressing, to fit more what I like to watch.
Trying to find a better balance between attack and defense. Wingers are set to participate more in defensive play, while defensive midfielders are set to participate more in attacking actions.
Trying to increase focus on defense by using Much Lower Defensive Line and support duty winger. Since IF is not attacking, I set them narrower to open space for wingbacks, and set cross low for the wingbacks.
Poacher said: Nahh, I don't think it's about the consistency roll.
I've seen many times how 2-3 players had 5.8-6.0 ratings in the 1st half and then 7.5-8.0 ratings in the 2nd half.
It's just players might have a very bad RNG for their actions in the 1st half but in the 2nd half they might a very good RNG for their actions.
We all know even with a very good tactic sometime you can hit a bad run ( few loses/draws in a row ) but we don't change our tactic after that because we know it's good and it was just bad luck that happens. The same goes for the players in a match, they might have periods when they have bad RNG for their action but it won't last forever and as I said if a player did poorly in 1st half then he might do great in the 2nd half and subbing him you just take away his opportunity for a good RNG.
That's my philosophy. Expand
It might be a good way to end a run of bad results. Hard to know what is more efficient during a season.
Poor performance is a good indicator that players got a bad roll for consistency. They can obviously recover with a goal or some good moves, but I would rather switch then for someone with a possible better roll than keep then at the risk of having them in a bad day.
Usually I sub any player performing badly (6.3 or less). Then, if anyone is tired or has a yellow card, I give it some thought about changes. I try to make 3-4 changes at half time, and save at least one just in case of injuries.
I would like to see for prevent short GK. I mean, if playing from the back is not good, then removing that should make people play more from the back, which should help you. However, it seems worse without it in my tests.
pixar said: Do you use FM RTE* in your personal tests? (*for freeze morale) Expand
No. The league is just edited so all players in the league have maximum consistency and minimum injury proneness. It also removes all leagues other than English ones.
Got a bit tired of the same shape, so I'm trying to shake things a bit. Using CM instead of Vol, even though they haven't work very well in this patch yet.
Sane said: @ZaZ Hey, are you planning to release an update on this tactic?
There are no good results with this 4-3-3 on the fm arena yet... Expand
I have those two versions here to be tested, but I didn't have time to test them thoroughly yet. I usually take at least a couple of days testing before uploading anything here, and I only post if it is up to some standard.
I'm running some experiments in that direction, but it takes some times before I have some results to post here.
Let's assume that the true score of some tactic is 60 pts.
You test it for 5,760 matches and get 61 pts.
Then you re-rest it for 5,760 matches again and get 59 pts.
So you can conclude the RNG swing of a 5,760 matches test can be as high as 1 point ( -/+ 1 point away from the true score).
Btw, our RNG measure was for FM22 and it seems for FM23 the RNG has slight increased so it could be 1.5 points for a 5,760 matches test.
Have you thought about using something like a Monte Carlo sampling on the tests? Basically, take smaller samples of the 5760 matches repeated times, then make statistics based on the statistic of those samples (for example, taking the medians of those subsamples, and calculating the median of them).
It's not hard to implement, and it can increase the precision of tests without increasing the computational effort used. There are also other types of subsampling, including bootstrapping and other techniques.
One can go high, the other low, then that makes two points of difference.
Then you could probably just test with Manchester City or Liverpool.
I've seen many times how 2-3 players had 5.8-6.0 ratings in the 1st half and then 7.5-8.0 ratings in the 2nd half.
It's just players might have a very bad RNG for their actions in the 1st half but in the 2nd half they might a very good RNG for their actions.
We all know even with a very good tactic sometime you can hit a bad run ( few loses/draws in a row ) but we don't change our tactic after that because we know it's good and it was just bad luck that happens. The same goes for the players in a match, they might have periods when they have bad RNG for their action but it won't last forever and as I said if a player did poorly in 1st half then he might do great in the 2nd half and subbing him you just take away his opportunity for a good RNG.
That's my philosophy.
It might be a good way to end a run of bad results. Hard to know what is more efficient during a season.
I've seen many times how a poor performer in the 1st time becomes the man of the match in the 2nd half.
Also, how ratings works in the game is a joke. Looking at the rating you really can't say for sure whether the player is doing well or not.
That's why I always prefer subbing players exclusively on their Conditions because the Conditions level makes a huge difference - https://fm-arena.com/table/20-condition-morale-testing/
Poor performance is a good indicator that players got a bad roll for consistency. They can obviously recover with a goal or some good moves, but I would rather switch then for someone with a possible better roll than keep then at the risk of having them in a bad day.
ZaZ - Water 5.5 IF
( Reliable Set Pieces )
https://fm-arena.com/thread/3100-reliable-set-pieces-fm23/
Thanks for testing! I feel stupid for not using the reliable set pieces.
No. The league is just edited so all players in the league have maximum consistency and minimum injury proneness. It also removes all leagues other than English ones.
There are no good results with this 4-3-3 on the fm arena yet...
I have those two versions here to be tested, but I didn't have time to test them thoroughly yet. I usually take at least a couple of days testing before uploading anything here, and I only post if it is up to some standard.