Last updated: 22 June 2026
You have had the kind of month that makes you question everything. The selections felt right, the reasoning was sound, and the bank still went backwards. Nothing obvious went wrong, which is exactly the problem. The money did not disappear in one bad afternoon you could point to and learn from. It leaked, slowly, a few pounds at a time, through a hole you could not see.
Every bettor has holes like this. I have them, you have them, and the backer quietly making a profit has them too. The difference is not that the winning bettor found some perfect method with no leaks. It is that they have found and plugged more of their holes than you have. Each one you close makes you instantly more profitable than you were the day before.
What is horse racing record keeping?
Horse racing record keeping is the habit of logging every bet you make, then studying those records to find where you consistently win and where you consistently lose. Split your bets into winners and losers, look for patterns across odds, course, distance and class, and the leaks costing you money start to show themselves.
Why every bettor has holes in their betting
You have holes because nobody bets equally well in every situation. You read some races better than others. Some prices suit your style and some flatter you into mistakes. The conditions where your judgement is sharp quietly fund the conditions where it is not, and the net result looks like a vague, frustrating drift rather than an obvious error.
A perfect betting record with no holes does not exist, and chasing one is a waste of a good evening. The realistic goal is narrower and far more useful. Find the leaks that are doing the most damage, close them one at a time, and let the improvement compound. That turns a frustrating month into a solvable problem, which is the whole point of keeping records in the first place.
Start with a complete record of every bet
You cannot find a leak you have not recorded. The first job is a full history of your bets, every one of them, not just the ones you remember fondly. If you have not been keeping records, you are not starting from zero. Ask your bookmaker for your account history, or log in and download it. If you bet on the exchange, your statement is sitting there ready to export.
What you want against each bet is the detail that lets you slice it later: the date, the course, the price you took, the going, the race type and class, the distance, the field size, and where your horse sat in the market. The more honest the record, the more it can tell you. A record that quietly leaves out the bets you would rather forget will quietly hide the very holes you are looking for.
Split your winners from your losers and hunt for patterns
Once you have the history, separate it into two piles: winning bets and losing bets. Now you go looking for patterns in both, because the holes show up as the gap between where you win and where you lose under the same heading.
Run your eye across every dimension you recorded. The odds your bets win and lose at. The courses. The race types. The going. The class of race. The market rank of your selection, whether you back favourites or outsiders. The distance. The number of runners. Even the sex of the horse. There are more, but that list is enough to find most leaks. What you are hunting for is a large, consistent difference between your winners and your losers for the same heading.
For example: “I am winning ten percent of my bets in races over a mile, but thirty-eight percent of my bets in races under a mile.” That is the kind of gap worth stopping on. It is not proof of anything yet, but it is a thread worth pulling, and you would never have seen it without the record in front of you.
A pattern is not a hole until it makes sense
This is the step almost everyone skips, and skipping it is how careful backers talk themselves into nonsense. A difference in your records is only a real hole once you can explain why it exists. Find a possible pattern, then go back through those selections a second time and look for the reason behind it.
Stay with the distance example. You dig into the longer-trip losers and notice that most of those horses had done nearly all their previous racing over shorter trips. Suddenly the gap makes sense. You are backing horses up in distance into races they are not proven to handle, so of course the longer-trip results are poor. That is a confirmed hole, and now you know what to do about it.
If you cannot find a reason, do not fix it. It is dangerously easy to invent a pattern out of a handful of results, pile on conditions until it fits your past perfectly, and end up describing noise rather than a genuine edge. Keep looking for the explanation if you want, but until you have one, leave those bets alone. A hole you cannot explain is usually just a small sample wearing a disguise.
Plug the leak, then prove it with the numbers
Once a hole is confirmed and it makes sense, plugging it is simple: stop making that bet, or only make it when the condition that caused the leak is absent. Then keep watching, because the honest test of whether you have actually improved is not how the next few results feel. It is whether the numbers move.
The cleanest way to judge that is actual versus expected, measured at Betfair SP. It simply asks whether a group of your bets won more often than their prices implied they should. A figure above one across a real sample means the angle is genuinely paying its way. A figure below it means you are following habit rather than value, and the time to find that out is before it costs you more, not after. This is the same discipline that separates a real betting system from a story you tell yourself about your good days.
Race Advisor’s analysis of more than 1.5 million UK runners between 2013 and 2026 shows why that test matters. Win rate on its own swings wildly depending on where a horse sits. The market favourite wins 33.4% of the time, the second favourite 19.9%, the third 13.4%, and anything fifth in the betting or longer just 3.9%. Yet across every one of those positions the actual versus expected figure sits between 0.99 and 1.01.
The same flat line holds when you cut the data the way your own records do. Split by going, strike rates run from around 8% on yielding ground to 14.5% on firm, and the actual versus expected stays glued to 1.00 throughout. Split by trip, races under a mile, middle distances and staying trips all come in between 10% and 11%, with the actual versus expected barely moving off 1.00 in each band. Race Advisor’s own PR ratings tell the identical story: the top-rated horse in a race wins 32.3% of the time and anything outside the top four wins 4.6%, a vast spread in raw strike rate, and still the actual versus expected holds between 0.99 and 1.01 right across the ranks.
That stability is the whole point. The market prices the obvious factors almost perfectly, so no blanket angle built on rank, going or trip alone is going to pay its way. The edge you are hunting is conditional and personal, the gap between where your judgement is sharp and where it is not, and a baseline this steady is exactly what lets you see it. When a slice of your own records pulls clear of an actual versus expected of 1.00 and you can explain why, that is a real hole. When it does not, you were looking at noise. Keep the record, split the piles, find the gap, confirm the reason, and plug it.
If you have never kept a proper record, that is the place to start, and you do not need a spreadsheet of your own to do it. Race Advisor’s free Bet Tracking tool logs your bets and keeps the history in one place, so the next time the bank drifts you have something to slice instead of a vague feeling. Then pull up today’s racecards and bet the next one knowing it is going into a record that will eventually tell you the truth.
