Estimate head-to-head outcome by winrates Let's say Fighter 1 winrate is $60\%$ and Fighter 2 winrate is $30\%$. How can we roughly estimate outcome probabilities of their fight just by that info? In other words, how can we rescale those winrates to add up to $100\%$? Something like $x' = \frac{x} {x + y}$ gives us $33.3\%$ for the second fighter, while intuitively, his win probability should go down since he's facing a stronger than average opponent. Is there some known approach for situations like this?

I've found this Log5 approach which seems to be good enough for me.

It estimates the probability of A defeating B via the following formula:

$p_{A,B} = \frac{p_A-p_A\times p_B}{p_A+p_B-2\times p_A\times p_B}$

$60\%$ and $30\%$ become $77.(7)\%$ and $22.(2)\%$ which fits my expectations quite nicely.