Horse Racing Betting Records: How to Track P&L, ROI and Strike Rate Like a Professional
Available in US
I went through three full years of betting before I bothered to track my P&L properly. Three years of telling myself I was probably “about even”, and three years of avoiding the spreadsheet that would have told me the truth. When I finally sat down and entered every bet I could reconstruct from bank statements and bookmaker history, the number was minus £4,200. I’d lost the rough equivalent of a holiday every year and never noticed because I’d never measured.
That afternoon changed how I bet on UK racing more than any analytical breakthrough ever did. The spreadsheet wasn’t an accounting exercise; it was the truth-telling mechanism that exposed which selections, days, race types and stake sizes were quietly costing me money. Within twelve months of disciplined tracking my running loss had turned into a small but consistent profit, with no change in the form work I was doing. The difference was visibility.
This article walks through what to record for every bet, how to calculate ROI, yield and strike rate correctly, how to set up a working spreadsheet from scratch, and how to use the monthly review to spot leaks and confirm where your genuine edge sits.
Essential Fields: What to Record for Every Bet
The temptation when setting up a betting log is to record everything you can imagine. Resist it. A spreadsheet you don’t fill in every day is worth less than a simple one you complete consistently. Start with the minimum fields, add complexity only when you’ve proved you’ll maintain it.
The non-negotiable columns are date, meeting, race time, horse name, bet type (win, each-way, place, lay), stake, odds taken, bookmaker or exchange used, and result. Those nine fields are enough to calculate every metric that matters. Anything beyond that is editorial — useful, but not foundational.
The second tier of fields, once the basic habit is established, includes pre-race odds versus odds taken (to track whether you’re shopping prices effectively), recommended stake versus actual stake (to catch yourself sizing up or down emotionally), source of selection (your own analysis, a tipster, a hunch), and a brief note on the analytical case. The note doesn’t need to be a paragraph — three or four words is enough. “Best closer in field” or “stable form, going suits” gives you something to review later when you’re trying to understand why you backed a horse.
The mistake to avoid is recording bets after results are known. The brain is exceptionally good at retroactively rationalising bets that won and forgetting bets that lost. Bets logged before the off, with the analytical case written down in advance, produce a database that reflects your actual decision-making. Bets logged after the result reflect your story about your decision-making, which is usually flattering and usually wrong.
One field I now consider essential: emotional state. A single letter — C for calm, F for frustrated, T for tilt, H for hot streak. Across hundreds of bets, the pattern emerges. My T bets lose 15 percentage points more than my C bets in level-stakes ROI terms. That kind of insight is invisible without the data, and it’s the most actionable improvement most punters can make.
Calculating ROI, Yield and Strike Rate Correctly
Three numbers tell you almost everything about your betting performance, but they have to be calculated correctly. The casual punter who says they have a “20% strike rate” rarely knows whether that means anything, because they don’t have the other two figures to give it context.
Strike rate is the simplest. Winners divided by total bets, expressed as a percentage. If you’ve placed 200 bets and 60 won, your strike rate is 30%. That number on its own tells you very little. UK favourites win 30 to 35% of races overall, with the rate dropping to about 25.7% in handicaps and rising to roughly 33% in novice races. A 30% strike rate on selections that were mostly favourites is unremarkable. The same strike rate on selections averaging 8/1 in price is exceptional. Strike rate is only meaningful alongside average odds.
Return on investment is the more honest measure. ROI is total profit divided by total stakes, expressed as a percentage. Bet £1,000 across the year, end up with £1,080 in returns, your ROI is plus 8%. Bet £1,000 and end with £930, your ROI is minus 7%. Backing every UK favourite at level stakes returns roughly 93% of stakes — an ROI of minus 7%. Second favourites return around 88%, third favourites about 85%. Those baseline numbers anchor what good selection actually looks like. A consistent positive ROI above zero across 500+ bets is genuine evidence of edge.
Yield is closely related to ROI but expresses profit per pound staked rather than per pound returned. The maths is similar — profit divided by turnover. Where yield differs in practice is in how each-way bets and progressive staking are accounted for. For level-stakes win-only betting, yield and ROI converge to essentially the same figure.
The variance numbers matter as much as the headline ROI. The top 1% of UK racing bettors generate around 52% of total betting revenue, which means high-volume punters absorb variance that smaller bettors can’t. A punter placing 30 bets a year experiences enormous variance — the same edge that produces 8% ROI across 1,000 bets can produce minus 20% or plus 40% across 30 bets purely from luck. Knowing how many bets your sample contains tells you how much to trust the ROI figure.
The trap to avoid is comparing your strike rate or ROI to inflated benchmarks. Tipsters claiming sustained 20% ROI are almost always overpromising. Genuine edge sustained across years sits in the 5 to 12% range for the best amateur punters. If your own tracked ROI sits at break-even after 500 bets, you’re already substantially ahead of the casual punter average. If it’s positive at all you’re in the minority of UK punters across any meaningful sample.
Setting Up a Betting Spreadsheet from Scratch
The technical setup is straightforward. Open a spreadsheet, label the columns from the essential fields list, set up a few formulas, and you’re done. Most punters overcomplicate this stage and then abandon the spreadsheet because data entry becomes a chore. Simpler is better.
The basic structure has one row per bet, with the columns running left to right in the order you’d naturally encounter the information — date first, then meeting and race, then horse, then bet type, stake and odds, then bookmaker, then the result column filled in after the race. A returns column that calculates payout automatically from stake, odds and result removes a step from the daily logging. A running P&L column that sums profit and loss to date gives you instant visibility on cumulative performance.
The summary section, on a separate sheet or block of cells, calculates the headline metrics from the bet log automatically. Total bets, total stakes, total returns, total profit, ROI, strike rate, average odds, average stake. Add monthly breakdowns once you’ve got a few months of data. Add category breakdowns — by code (flat versus National Hunt), by meeting, by bet type, by selection source — as patterns emerge that you want to investigate further.
The formula for ROI in cell terms is straightforward: total profit divided by total stakes, multiplied by 100 to express as a percentage. For strike rate, count winning bets divided by total bets. Every spreadsheet application handles these calculations natively, and there’s no need for anything more complicated.
The discipline that matters more than the technical setup is daily logging. The bets go into the spreadsheet within 24 hours of being placed, not at the end of the month when you’re trying to remember what you did. Late logging produces incomplete records and tempts the rationalisation problem where forgotten losing bets disappear from the database entirely.
The other discipline is honest accounting of free bets and bonus money. If a bookmaker gives you a £20 free bet that returns £80, the £80 winnings count but the original £20 wasn’t your money — recording it as a £20 stake distorts the ROI calculation. The cleanest approach is to track promotional bets separately and exclude them from the core ROI figures, then add a separate column or section for cumulative bonus value.
Monthly Reviews: Spotting Leaks and Confirming Edge
The spreadsheet generates the data. The monthly review is what turns the data into improvement. Without the review, you have a record of bets; with it, you have a feedback loop that progressively shapes future decisions.
The review process I follow takes about 45 minutes once a month, usually on the first Sunday after the month closes. Step one is the headline numbers — total stakes, returns, ROI, strike rate, change from previous months. Step two is the breakdown by category — which selection sources, race types, days of the week, bet sizes are producing the bulk of profit or loss.
The leaks reveal themselves in the breakdowns. The category that’s consistently negative across multiple months is a problem. The bet type with persistently poor ROI is one you should be doing less of, or analysing more rigorously. The meeting or trainer you keep backing despite a losing pattern is the emotional bias that needs intervention.
The wins reveal themselves too. The category with consistently positive ROI across multiple months is where your genuine edge sits. The bet type that’s quietly producing 8% returns is one you should be doing more of, with larger stakes if the variance profile supports it. The selection source — your own analysis on a specific race type, or a tipster on specific tracks — that’s adding genuine value deserves more of your attention.
The pattern that repeats across most punters’ monthly reviews is concentration. Profitable betting tends to come from a narrow set of conditions where your work is most effective, and loss-making betting tends to come from venturing outside those conditions because you wanted action rather than edge. The monthly review identifies the boundary between the two and informs the next month’s selection discipline.
The honest review also identifies the limits of your sample. A single month of 30 bets tells you very little about underlying edge. Three months of data starts to show patterns. Six months produces conclusions you can act on. Twelve months gives you a sample large enough to make structural changes to how you bet. Resist the temptation to act on small-sample conclusions in either direction.
For deeper context on how tracking interacts with stake sizing decisions, my full guide to horse racing bankroll management walks through the Kelly Criterion, percentage staking and drawdown control that the spreadsheet data feeds into.
From Tracking to Genuine Edge
The mechanical work of maintaining a betting spreadsheet is small. The cumulative effect over a year of disciplined tracking is significant. Every serious UK racing punter I know maintains some version of this record. Every casual punter I know either doesn’t, or maintains one but doesn’t review it. The gap between those two behaviours, more than any analytical skill difference, explains why some bettors stay marginal year after year while others develop genuine sustainability.
Start simple. Nine columns, one row per bet, daily logging, monthly review. The habit compounds across months and years into a database that knows your betting better than you do. Three years from now you’ll have hundreds or thousands of rows that tell you exactly where your money goes, exactly which strategies work, and exactly which patterns you keep falling into. That knowledge is the closest thing to a foundation for sustainable UK racing betting that exists.
How many bets should I log before drawing conclusions about my ROI?
At least 300 bets for early indications, 500 to 1,000 for moderate confidence, and 2,000+ before treating your ROI as a stable estimate of underlying edge. Variance at smaller samples is enormous — the same true edge can produce dramatically different observed ROI across short windows. For category breakdowns (specific race types, trainers, etc), the sample sizes need to be similarly substantial within each category before the figures mean anything.
What is the difference between ROI and yield in betting?
They measure essentially the same thing for level-stakes win-only betting. Both express profit relative to stakes. ROI typically refers to profit divided by total turnover. Yield is used similarly but is sometimes calculated differently when each-way bets and progressive staking are involved. For practical purposes, treat the two terms as interchangeable when you see them quoted across UK racing tipping services and analytical sources.
Written by the editors at Horse Racing bet Strategy.
