TrueSkill

A Bayesian skill-rating algorithm developed at Microsoft Research (Herbrich et al., 2007) that models each player or option as a normal distribution with a mean skill μ and uncertainty σ, updating both after each pairwise match. Originally designed for matchmaking in competitive video games, TrueSkill is now widely used in HCI and accessibility research to rank design options from pairwise or best-worst comparison data because it converges faster than ELO, handles multi-player and partial rankings, and provides explicit confidence intervals. Ranked options whose lower-bound μ−2σ sits above a threshold are typically selected as winners.