Monday brought 133 races across the British card and the first-pick win rate came in at 20.3% -- 27 winners from the model's top selections. That sits below the seven-day average of 24.1%, so it was not a vintage day by recent standards, but it was far from a bad one either. You would need to find something genuinely wrong with the form assessment to be critical of a rate that comfortably outpaces chance.
Where June 1 held up better was in the broader numbers. The model's top three picks produced at least one winner in 83 of 133 races -- an any-top-3 win rate of 62.4%. That tells you the right runner was being identified and rated within the model's top selections most of the time, even when a different dog within that group took the race. In greyhound racing, where the margins between runners in the same grade can be razor-thin, finishing with 62.4% of races won by a top-3 pick represents solid form reading.
The place data adds context. First-pick selections placed in 54 of 133 races, meaning the model's number-one choice was first, second, or third in more than 40% of all races. When you combine the any-top-3 place rate (119 places from 133 races), it is clear that the model was in the right area throughout the card -- the variance was coming from within the top group, not from the model missing the field entirely.
133 races in a single day produces a lot of variance. A first-pick win rate of 20.3% on a Monday card is an honest result that accurately reflects how closely matched runners in similar grades can be. The any-top-3 rate of 62.4% is the number worth keeping an eye on as a measure of underlying model health.