How the value picks work

What A Value Pick Actually Is

A value pick is not a prediction that a horse is going to win. It is a statement about price. Our rating works out a chance for every runner in a race, and when that rated chance is noticeably higher than what the current odds imply, the horse gets flagged as a value pick. The bet isn't "this will win" - it's "this price is bigger than it should be, given everything we can measure about the horse and the race".

That distinction matters because most punters, and most tipping services, focus on picking winners. Value work focuses on picking prices that don't match the true chance. Over enough bets, backing horses whose odds understate their real chance should return more than backing horses at fair or short prices, even though plenty of those value picks will still lose on the day. Racing is a game of small edges repeated many times, not certainties.

Rated Chance vs Implied Probability

Every fractional price can be turned into an implied probability. A horse at 4/1 is being priced by the market at roughly a 20% chance of winning (1 divided by 5). A horse at 9/2 is being priced at closer to 18%. Bookmaker books also carry an overround - the total implied probability across all runners adds up to more than 100% - which is how the layer builds in a margin, so market prices are never a pure read of true chance to begin with.

Our rating ignores the overround and just asks: based on the 15 measurable signals we score every runner on - form, class, pace shape, going, trainer and jockey patterns, and so on - what is this horse's realistic chance of winning this race? When that rated chance sits meaningfully above the implied probability from the current odds, there's a gap. The size of that gap is the whole ballgame - a small gap isn't worth flagging, a wide one is.

Why A Value Pick Can Lose Most Of The Time - And Still Be The Right Call

This is the part that trips people up. If a value pick is priced at, say, 8/1 when our rating thinks it's more like a 5/1 shot, that horse is still an underdog. It will lose more often than it wins. That's simply how prices at those odds work - a 5/1 true chance still loses roughly four times out of five. Flagging it as value doesn't change the horse's actual chance of winning the race, it just says the reward on offer no longer matches that chance fairly.

This is exactly why we publish the full record rather than cherry-pick results. Across 15,107 settled picks over 418 days, the public record shows 25% winners and 50% placed. Most individual picks, including plenty flagged as value, don't come in - that's expected and consistent with racing generally, where favourites themselves only win a minority of races. What matters for judging a value approach isn't any single result, it's whether the gap between rated chance and market price holds up as a genuine edge across a large enough sample, which is precisely why every pick sits in an auditable, dated record rather than a highlights reel.

Because the record is public and settled bet by bet, you can check the shape of it for yourself rather than take our word for it. That transparency is the point - a value angle only means something if you can see the losing picks sitting alongside the winning ones, not just the ones that came in.

The Risk Bands: Careful, Balanced And Bold

Not every value gap is treated the same way, because a small edge and a large edge carry very different risk profiles. We group flagged picks into three bands based on how wide the gap is between rated chance and implied probability: Careful covers gaps of roughly 5-10%, Balanced covers 10-15%, and Bold covers gaps of 15% or more.

A Careful pick might be a shorter-priced runner where our rating sees a modest edge over the market - the kind of gap that needs a large sample to prove itself, because it's easily lost in the noise of a single race. A Bold pick usually sits at a bigger price, where our rating disagrees more sharply with the market's view of the horse's chance. Bold picks will lose more often in absolute terms, simply because bigger-priced horses lose more often full stop - but when they land, the return reflects that bigger gap.

None of the bands are a guarantee, and none of them change the fact that this is still a horse race with all the usual uncertainty - injury, trouble in running, a rider's decision at a crucial moment. The bands exist so you know, before a race off, roughly how wide a disagreement our rating has with the market, and can judge the pick in that context.

How This Gets Flagged In Practice

Every runner is scored across the same 15 measurable signals before a race, producing a rated chance that's independent of what the bookmakers are showing. That rated chance is then compared against the current market price to generate the implied gap, and if the gap clears the threshold for one of the three bands, the horse appears as a value pick with its band shown alongside it.

Nothing here is a promise about the next winner. It's a consistent method for spotting when the price on offer looks generous relative to a measured chance, applied the same way to every race, with every outcome logged in the same public record - 15,107 picks and counting over 418 days, 25% winners, 50% placed - so the method can be judged on its actual, ongoing results rather than on any single tip.