Butler by cross-IMPs
For every board, each table’s result is expressed as a single NS-signed score. A pair’s per-board Butler is the mean of the IMP differences between their score and every other table’s that played the board — sum(IMP(mine − theirs)) / (N−1), rounded to 0.1. It answers ‘how did this result compare with the rest of the field?’, and the field’s Butler values sum to ~0 by construction.
Cumulative Butler
A player’s cumulative figure is the sum of their per-board Butler over every board they played, divided by the number of boards — their average cross-IMP per board, to 0.01. Only played boards count; not-played, passed-out and director-annulled boards are excluded.
Team VP & the scales
Team match VPs come from the source (EBL/WBF). The site also checks each published VP against the standard IMP→VP scale for that IMP margin and board count — the continuous 20–0 scale from 2013 on, the older discrete 25–5 scale before it. Any gap between the published VP and the scale is a director adjustment / penalty and is shown as such. Withdrawn boards (annulled at both tables) reduce the board count the scale is keyed on.
Double-dummy analysis
Each deal’s makeable-tricks table and par (minimax) are computed double-dummy (Bo Haglund’s DDS via endplay). A deal popup also shows, for the viewed match, the tricks the declarer could make in the actual contract after the actual opening lead. When the recorded lead card isn’t in the leader’s hand (a data slip), the nearest card of that suit is used and the lead is flagged.
Live rounds & data
A round still in progress is marked provisional — its running scores feed the standings but may change. Data is scraped from the official EBL / WBF result microsites; where a round published only team results (no line-ups), the team VPs still show but per-player Butler isn’t available.