You sit down with a racecard, open a second tab for the form, a third for the going report, and a fourth because you remembered the trainer had a sick yard in February. Forty minutes later you have ten factors per horse and no idea which one to trust. So you back the favourite anyway, lose, and tell yourself next time you will be more thorough.
That instinct, that more factors must mean a better answer, is an expensive habit in form study. The more useful question is how few you need.
How many factors should you use to analyse a horse race?
Four is a sensible number: one from each of four categories, which are speed, suitability to today’s conditions, fitness and connections. Extra factors often repeat what you already know, and the market may have priced that in too. Treat four as a practical starting point to test, not a proven edge.
Why more factors can leave you no better off
Adding factors makes you feel more confident. It does not necessarily make you more right. One reason is overlap: many of the things you check are different ways of saying the same thing.
Our own ratings show it. Across GB and Irish racing from 1 July 2013 to 14 October 2026, our PR Rank 1 horse (the horse our PR rating puts top) is also the Betfair favourite in 81% of races. Our Race Predictor’s top horse (the model’s number one) is the favourite in 41%, and VDW Rank 1 (top of our VDW ratings) in 31%.
So when two or three signals point at the same horse, you are often looking at the favourite wearing different hats. The market has already shortened the price to reflect it. That is what we can see for these signals. We have not shown that every extra factor you could add behaves the same way.
Here is how the simple signals have fared. All figures are at Betfair SP (the Betfair starting price), level stakes, after 2% commission, with fallers and pulled-up horses counted as losers. “Return on stakes” is the profit or loss as a percentage of everything staked, so -1.4% means about 1.4p lost per £1 staked.
| Signal | Races | Win rate | A/E | Return on stakes |
|---|---|---|---|---|
| Betfair SP favourite | 167,257 | 34.4% | 1.00 | -1.4% |
| VDW Rank 1 | 153,920 | 21.6% | 0.99 | -3.8% |
| Race Predictor top horse | 167,314 | 25.5% | 1.01 | 0.0% |
A/E means actual wins divided by the wins the odds imply, and 1.00 means the price was about right. If the term is new, our A/E ratio guide explains it in plain English.
Every row lands within a few percent of break-even. That is a market doing its job. These are past results, so read them as the shape of the thing rather than a promise about next week.
The four categories
Every race can be read through four lenses. Each answers a different question, so one factor per lens covers the ground. A second factor inside the same lens usually re-asks a question you have already answered.
Speed. Can this horse run quickly enough for this race? Pick one speed measure or pace position, such as a speed rating, and stop there. Our guide to speed ratings covers how to read one.
Suitability to conditions. Does this horse handle today’s going, distance and class? A horse that is rated 110 in 12-furlong soft-ground handicaps is not the same horse over 7 furlongs on the all-weather. This is the lens people forget to ask at all.
Fitness. Has the horse been racing recently, and was its last effort genuine? Days since its last run and a glance at the recent form figures usually settle it.
Connections. Is the trainer in form, and does the jockey suit this horse and this track? Trainer and jockey form over the last 14 days gives you both readings from the racecard.
Four lenses, four answers. With practice you can score a race in a few minutes. Which factor you choose inside each lens matters less than choosing deliberately and using the same ones every time.
Where combining does help
If you want several signals working together, it is cleaner to let a model do the combining than to stack ratings by hand. Our Race Predictor is itself built from many factors, so the combining has been done for you.
Even so, in our data the Race Predictor’s top horse sits at an A/E of 1.01, roughly in line with the market. A combined model is a better starting point than a pile of overlapping ratings. It is not a licence to expect more than the figures show. Our guide to combining ratings goes through what has and has not been tested.
One more distinction is worth holding. Agreement between signals tells you how confident to feel. Whether the price is generous is a separate question, and only the price can answer that. Most of us blur the two.
A test you can run
The four-lens method is untested as a way to make money, so test it before you trust it. This version tests two things: whether your ranking finds winners, and whether it ever disagrees usefully with the market.
- Pick one race a day, the first of the day’s card.
- Before looking at any prices, write down your four lenses and the one factor you will use for each.
- Score every runner out of ten on each lens, using the same factors every time.
- Add the scores and write down your top three.
- Do not bet. After the race, note the winner, where your ranking put it, and the Betfair SP of your top-ranked horse.
Repeat for 20 days. That is a small sample, so it can only be suggestive. Aim to keep going to 100 races or more before drawing a conclusion. Early results swing about a lot, and a good or bad fortnight tells you very little.
Two things come out of the record. First, which lens did the most work in finding winners for you. Second, and more useful, the prices. If your top-ranked horse is usually the favourite, you have rebuilt the market and found nothing new. The races worth studying are the ones where your top pick was not the favourite, because that is where your judgement differs from the price. Compare how those picks performed against what their odds implied.
What to take from this
You are not trying to find the horse most likely to win. You are trying to find horses where the price is wrong, and extra factors help with that less often than it feels, because the market sees much of what you see. Four lenses, one factor each, chosen on purpose and checked against your own record, is enough to start. It is also a calmer way to spend a racing morning.
On a Race Advisor card our ratings sit alongside trainer and jockey form, so picking one from each lens takes seconds. Open today’s race cards and try it on the first race that catches your eye.
