A Probabilistic Framework for Pairwise Comparisons
Go to CalculatorThe Bradley-Terry model is a statistical framework for modelling pairwise comparison data, first proposed by Bradley and Terry (1952). It converts a series of binary comparisons into a ranking. Pairwise comparisons reduce cognitive load compared with direct ranking of multiple items[1][4].
Given items \( \{1, \dots, n\} \) with positive parameters \( \pi_i \), the probability that item \( i \) is preferred over \( j \) is:
\( \pi_i \) represents the "strength" or "utility" of item \( i \).
Bradley & Terry (1952), Zermelo (1929)
Parameters are usually constrained (e.g., \( \sum \pi_i = 1 \)) for identifiability.
Bradley & Terry (1952)
This shows the model is a special case of logistic regression.
Bradley & Terry (1952), Springall (1973)
If \( P(i > j) > 1/2 \) and \( P(j > k) > 1/2 \), then \( P(i > k) > 1/2 \).
Davidson (1970)
Maximum likelihood estimates can be computed iteratively:
\( w_i \) is the total wins for item \( i \); \( n_{ij} = y_{ij} + y_{ji} \).
Bradley & Terry (1952), Zermelo (1929)
This shows equivalence to Elo ratings with scale factor 400.
Bradley & Terry (1952)
| Extension | Description | Reference |
|---|---|---|
| Plackett-Luce | Rank multiple items | Plackett (1975), Luce (1959) |
| Davidson Model | Accommodates ties | Davidson (1970) |
| Covariate BT | Includes item features | Springall (1973) |
| BT Regression Trunk | Tree-based modeling of preferences | D'Ambrosio et al. (2023) |