When I read this thread, I didn't get the sense that anybody was trying to ridicule you. Naturally, it's your choice whether to continue posting, so this may be a waste of my time.
"In an urn there are 100 marbles, 50 of them are white and 50 are black. If you extract 10% * 100 = 10 marbles and paint them black and then put back in the urn how many white marbles will be in the urn now? Between 40 and 50 depending on how many white marbles you extract, but the mean is 45."
I agree that the mean is 45 white marbles. Five of the marbles you took out were black already, so painting them black did not change the colour.
There has been a 10% swing in the the split of white and black marbles, but the number of white balls has changed by only 5% of the total marbles.
Isn't this just a different reading of the same data?
In the example below, White makes an error that gnubg reports as costing 9.26% wins: your chances to win have changed from 43.98% to 34.72%.
Your wins relative to your opponent have swung from 56.02% - 43.98% = 12.04% difference to 65.28 - 34.72% = 30.56% difference. That 9.26% error causes a swing of 30.56 - 12.04 = 18.52 = 2 * 9.26%.
How do you think gnubg should report this? Are you arguing that gnubg should say you have given up 1 - (34.72 / 43.98) = 21.06% of your match winning chances?
Perhaps you could post an example which doesn't start with an even split of marbles in the urn. That may help.
The score (after 0 games) is: Blue 0, White 0 (match to 1 point, post-Crawford play)
Move number 3: White to play 61
| Blue | 4 |
|---|---|
| |
| White | 14 |
Position ID: DwAAQBgAAAAAAA Match ID: cAknAAAAAAAA | |
• White moves 6/off 2/1
| # | Ply | Move | MWC | |
|---|---|---|---|---|
| 1 | 2 | 6/5 6/off | 43.98% | |
| 0.4398 0.0000 0.0000 - 0.5602 0.0000 0.0000 | ||||
| • | 2 | 2 | 6/off 2/1 | 34.72% ( -9.26%) |
| 0.3472 0.0000 0.0000 - 0.6528 0.0000 0.0000 | ||||