Transparency matters. If we are going to ask you to trust our model, we need to show you how it is actually performing. Here is the seven-day snapshot from the last eight days of racing, covering 704 races across all tracks.
Our first pick has won at a rate of 24.6% over the period. In a six-runner sport, that means our top selection is winning roughly one and a half times more often than random chance would suggest. The place rate for first picks sits at 45.9%, so nearly half the time our best-rated dog is finishing in the first three. Those are solid, workable numbers if you are backing level stakes or building place accumulators.
The combined figures are where it gets interesting though. When you look at whether any of our top three picks found the winner, the strike rate climbs to 63.4%. That means in nearly two out of every three races, one of our three selections was winning. The place rate across the top three hits 88.9%, which means we had at least one placed runner in almost nine out of ten races. That kind of consistency gives you a real edge when constructing forecasts and tricasts.
The best day in the sample was June 21st, when our first pick landed 39 winners from 120 races for a 32.5% strike rate. That is the kind of day where level-stakes backers would have had a very profitable afternoon. The weakest day was June 23rd, with 14 first-pick winners from 59 races at 23.7%. Even on the quiet days, the model stayed close to its long-term average, which tells you there is genuine consistency underpinning the numbers rather than wild swings between feast and famine.
We will keep publishing these weekly updates so you can track the model's performance over time and make your own judgements about how to use the picks.