Last updated: 19 June 2026
Every so often I get a run of emails that all circle the same problem, and this one comes round more than most. A backer is using a betting system that takes its selections from the Racing Post betting forecast, follows the rules to the letter, and then compares notes with someone running the exact same system. The selections do not match. Same rules, same source, different horses. It is enough to make you think one of you has misread the method.
Neither of you has. There is one simple reason it happens, and once you understand it you can stop worrying about it and start using it.
The Racing Post betting forecast is a prediction of how the public will bet, not a tip for the likely winner, and it is revised between the night before and raceday morning. That is why two backers running the same forecast-based system can end up with different selections; they read the forecast at different times. You should not trust it as a verdict, but you can use it as a guide. Note the early forecast, compare it again on the morning of the race, and treat the horses whose prices have moved as the ones worth a closer look.
What the Racing Post betting forecast actually is
The betting forecast is the little column of prices printed next to each race, putting an estimated odds figure against every runner before a single bet has been struck on course or with the exchanges. It looks like a verdict on the race, and most backers quietly treat it as one. It is not. It is the Racing Post’s estimate of how the betting market will form, which is to say where the public’s money is likely to go.
That distinction matters more than it sounds. A horse can be the shortest price in the forecast not because the compiler thinks it is the best handicapped or the most likely winner, but because they expect it to be the most heavily backed. Reputation, a fashionable trainer, a name from a big yard, a horse that was well found last time; all of these pull public money, and the forecast is trying to anticipate that pull. It is a forecast of sentiment as much as of merit. Treat it as a read on the crowd, not a read on the race.
It is worth knowing this is only one of several Racing Post numbers that get leaned on far harder than they should be. The RTF figure is another Racing Post number that backers mistake for a verdict when it is really just a piece of context. The forecast belongs in the same drawer: useful, worth understanding, but not something to follow blindly.
Why the forecast changes (and why your selections do too)
Here is the part most people never notice. The betting forecast is not fixed. If you look at it the night before racing you will often find a different set of prices to the ones showing late the next morning. It is revised as the picture sharpens, and that revision is exactly why two backers running the same system can finish with different bets. One of them read it the night before. The other read it on the day. The system did its job both times; it just had different inputs to work with.
The reason it shifts is that the early forecast and the later one are estimating two slightly different things. The night-before version is compiled before the market has really formed, so it carries a bit more of the compiler’s own opinion about which horses are best in the race. By the morning, with more information in and the market starting to take shape, the prices drift towards where the public actually looks set to bet. So the forecast does not just get more accurate over time. It changes character, from a partly opinion-led estimate to a closer guess at the crowd.
Once you know that, the changing forecast stops being a nuisance and becomes information. The gap between the early price and the later one tells you how the expected market is moving on each horse, and that movement is something you can read.
How to read the early forecast
Start by looking at the forecast for the races you are interested in as soon as it becomes available, which is usually the night before or first thing in the morning. At this stage you are looking at the compiler’s view before the market has formed, so it carries a little of their thinking about which horses are genuinely the strongest in the race, not only which ones the public will pile onto.
That early lean is a small edge if you use it sensibly. The front few in the early forecast are the runners the compiler rates before sentiment takes over, so they are worth a closer look. I would not bet off that alone, but it is a reasonable shortlist of horses to study properly. Note the prices down. You are creating a reference point, a snapshot of where each horse stood before the crowd had its say.
This is also where it pays to have your own view rather than borrowing the Post’s. If you build your own tissue, even a rough one, you can compare the forecast against it instead of taking it on trust. Working out how to build your own odds line is the single biggest upgrade you can make here, because it turns the forecast from something you follow into something you measure against.
Compare it again on raceday morning
Now come back late morning on the day of racing and go through the same races again. Have any of the prices changed? Which horses have shortened, and which have drifted? Set the morning forecast next to the snapshot you took the night before, and the differences jump out.
Keep in mind what the forecast is tracking while you do this. The prices are moving towards how the public is expected to bet, not necessarily towards the horse the compiler privately thinks will win. So a price that has shortened overnight is usually telling you that money, or anticipated money, is gathering on that runner. A price that has lengthened is telling you the expected support has cooled. Neither move is a verdict on the horse’s actual chance. It is a read on where the crowd is heading.
The forecast is, in the end, an estimate of the market, and the market is broadly efficient. Race Advisor’s analysis of over 164,000 UK races between 2013 and 2026 shows the market favourite wins 34.4% of the time, with an A/E of 1.00. A/E simply measures actual wins against the wins the price implied, so a figure of 1.00 means the shortest price is paying out almost exactly what it should, no more and no less.
Split that by going, by class and by race type and the figure barely moves, sitting between 0.98 and 1.01 in nearly every slice. That is precisely why the favourite is a fair guide to who is likely to win and a poor source of value on its own. If you want to understand why the crowd’s price is so hard to beat, it helps to know how the market forms in the first place. The forecast is trying to predict that market, so it inherits both its accuracy and its lack of obvious value.
What a shortening or lengthening price is telling you
So you have found a horse whose forecast price has moved between your two looks. What do you do with it? I always prefer to study the race myself first, so in the morning I go through the runners at the front of the forecast in detail. That way, even if nothing has moved, I can still take a bet if one of them looks strong enough on its merits. But where a price has shifted on a horse I already rate, the direction of the move points me towards two very different conclusions.
- The forecast has shortened on a horse you rate: the runner is likely a strong contender, but the shortening warns that it may not be offering much value any more. The crowd has found it too. You can look for a rival to oppose it with, or use the short one as a saver while you place your main bet on a bigger price.
- The forecast has lengthened on a horse you rate: this is the more interesting one. The compiler thought enough of the horse to price it up before the market formed, the expected support has not arrived, and you happen to agree the horse is strong. That combination can be a genuine value bet.
The logic underneath both lines is the same. You are comparing your own assessment of the horse against where the expected market is moving, and betting only when the price is bigger than your honest read of the chance. That is the whole of value in a sentence. The forecast movement does not create the value; it just flags the horses where your view and the crowd’s view have come apart, which is exactly where value tends to live.
Two things are worth holding in your head while you do this. A printed forecast is a bookmaker-style tissue, and it carries a bookmaker’s margin. Across these same races the bookmaker starting-price book works out to an overround of roughly 17%, while the exchange book is effectively fair at around 100%. The forecast is showing you the shape of the market, not a price you could ever actually back to.
The favourite is not equally trustworthy in every race, either. On soft-to-heavy going its A/E slips to 0.91 across more than two thousand races, the one condition where the shortest price throws up clearly more upsets than it implies. When the ground is bottomless, lean on the forecast a little less and your own read of the race a little more.
Trust your own read, use the forecast as a guide
So, should you trust the Racing Post betting forecast? Not as a verdict. It is a prediction of how the public will bet, it carries a dose of sentiment, and it changes between the night before and the morning. Anyone betting off it blind, especially without noticing that it moves, is handing their decisions to a crowd-estimate and wondering why the results feel random.
Used properly, though, it earns its place. Take the early forecast as a shortlist of horses worth studying, note the prices, come back on the morning, and let the movement point you at the runners where the market has shifted. Then do the work the forecast cannot do for you. It tells you where the money is going; it does not tell you whether a horse is well handicapped, suited by today’s trip and ground, or quietly improving. That is form work, and there is no shortcut for it. If you want to get genuinely good at the part the forecast leaves out, learning to look into the race yourself is what separates a backer who follows prices from one who reads races.
That is also where having your own model helps most, because it gives you something to hold the forecast up against. The Race Predictor builds a view of every runner in the race from dozens of factors at once, so instead of comparing the forecast to nothing you are comparing it to your own honest assessment. When you would rather start from your own read of the race and use the betting forecast only to spot where the crowd disagrees with you, open today’s racecards and let the Race Predictor do the heavy lifting.
