Camillo wrote: What math did you use??

Yeah, that's a good question. I had to think about that for a bit, but I think I remember.
I made two assumptions:
- the stamina bug is a wrap-around effect
- I made an assumption about rounding during the calculation
Both assumptions can be completely wrong of course
I also have to assume something about the consistency of the player. Maximum performance change due to consistency is + or - 20%, so I assumed the average: 10%
If a player with this consistency has stamina 20, then I assumed that due to consistency the "modified stamina" can vary between 18 and 22. I assumed rounding at this point, so for the "modified stamina" 5 values are possible: 18, 19, 20, 21, 22. Then I assumed there to be a wrap-around bug, which turns these "modified stamina" values into: 18, 19, 20, 1, 2. The average of which is 12. So with all these assumptions a player with stamina 20 and average consistency would on average play like player with stamina 12.

Oh and before anybody says "I don't think the calculation works like that" I just want to point out that I don't believe the calculation works like that. There are 100 ways to code something like this and it would be totally amazing if I guessed right. But the calculation does produce sensible results which do not seem to conflict with observations.

I've recently acquired a 20 Stamina player so am interested to see how this plays out, as there are some reports on these forums claiming it's no longer a bug, whereas the linked game above from just a year ago seems pretty conclusive.

Am I right in understanding that stamina only affects a player's end of game rating? And that the starting rating will be unaffected by this bug, comprising a combination of skills/BPOS/consistency etc? As a lower league team I don't think it's likely to have a very significant impact - most of my players have relatively low stamina anyway so I regularly see a 0.5-1* drop-off between start and end ratings anyway. But can understand at the higher levels where stamina is high and margins are small it can make a bigger difference.

Do we know if consistency plays a part in this bug at all? ie. are high consistency players any more or less likely to be impacted by it? My guy is CON 18, which is why I ask.

From my experience with stamina 20 bug. It is a wrap-around effect. They 'fixed' the bug by showing it as going from 5* to 4*, but it used to show going from 5* to .5*.

It does only appear in the end game rating.

If consistency works like it is suppose to, a CON 20 player with stamina 20 should not be affected because a CON 20 player is suppose to play at is skill ratings at all times. The further from 20 the CON, the more variation from initial skill ratings are possible, up or down.

It's been a while since I've had a stamina 20 or a CON 1 player and I don't remember for certain which did it, but one of those had the bad effect about every third game they played.