How to calculate the percentage of winning using the evaluation bar?

Chess.com has moved on to an expected points model (Classification V2) which uses percentage and Elo to calculate whether the move was a blunder, brilliant, etc.

I want to know how to use the evaluation bar's numerical value and transform that into a percentage of winning without using any of Elo calculation. How can this be done?

For example: If you are white, and the evaluation bar is 0.00, you have 50% chance of winning, while -5.00 means that you have a lesser chance of winning.

Strict evaluation differences don't work since if you are white in a +28.00 position, and you make a move that makes it +24.00, it shouldn't be a blunder.

So, can anyone help?

I was doing research on this exact topic and found out how to calculate expected points from eval.

0,00 -M1

0,01 -M2

0,02 -M4

0,03 -M8

0,04 -M16

0,05 -M32

0,10 -15

0,20 -7

0,30 -3

0,40 -1

0,50 0

0,60 +1

0,70 +3

0,80 +7

0,90 +15

0,95 +M32

0,96 +M16

0,97 +M8

0,98 +M4

0,99 +M2

1,00 +M1

I dont know how to use Elo to calculate expected points, i hope this helps.

I don’t know if I agree that 0 eval is 50% chance of winning, maybe it is a lot less. This study does not consider drawing chances.

Yes it is... The game starts at 0.3 evaluation which is 55% win rate for White... If the evaluation became 0, it only means that Black had equalize the position...