Keeping a betting record: yield, ROI and closing line value

Almost every bettor believes they are roughly break-even. Almost none are, and the gap is not dishonesty — it is how memory works. What memory does to your record: • Wins are more memorable than losses, especially large ones and near misses • Losses get discounted with excuses — a bad refereeing decision, an injury, bad luck. Each excuse is individually plausible and collectively fatal • Small routine losses vanish entirely, and they are the bulk of the damage • The overall total is never actually computed, only estimated, and the estimate is generous What a record gives you that nothing else can: • A real number for whether you are profitable • The ability to see which parts of your betting work. Most bettors who track properly discover their record is positive in one area and heavily negative in another — typically positive on researched pre-match bets and negative on in-play, accumulators and secondary markets • Evidence to act on rather than impressions to argue with The discovery that in-play betting is where all the losses live is one of the most common outcomes of starting a record, and it is invisible without one.

The reason column is the one that changes behaviour. Forcing yourself to write why, in a sentence, before staking, kills a meaningful share of impulsive bets outright. If the honest sentence is I feel like it, you have your answer. The closing odds column requires you to check the price at kick-off and record it. It takes a few seconds and it is the difference between a log and a diagnostic tool.

Yield = profit ÷ total staked. The standard measure of betting performance. • 1% to 3% is genuinely good over a large sample • 5% or more sustained over thousands of bets is exceptional • Anyone claiming 20% or more over a long period is either betting tiny volumes in a niche market or not telling the truth ROI on bankroll = profit ÷ starting bankroll. A different question — how hard your capital worked — and heavily influenced by how aggressively you stake. High ROI with high stakes is not skill, it is leverage. Closing line value (CLV) = whether the price you took was better than the price at kick-off. • You back at 2.10, the market closes at 1.95. You beat the close • Do that consistently and you are, by definition, ahead of the market's final estimate Why CLV is the most useful of the three: it converges far faster than profit does. Yield needs perhaps a thousand bets to become meaningful, because variance dominates before that. CLV becomes informative in a couple of hundred, because it measures your prices rather than your results. The practical implication: if you are losing money but consistently beating the closing line, you are probably unlucky and should continue. If you are winning money but consistently taking worse-than-closing prices, you are probably lucky and it will end.

Sample sizes, honestly: • Under 100 bets: tells you nothing at all about skill. Do not draw conclusions, in either direction • 100 to 500: patterns start to appear. CLV becomes readable. Yield still does not • 500 to 1,000: yield begins to mean something, with wide error bars • Over 1,000: a real signal, assuming your approach did not change halfway through The decisions a record supports: 1. Cut what loses. If one bet type is heavily negative over a few hundred bets, stop it. This is the single most profitable action a record enables 2. Concentrate on what works. If one competition carries your entire profit, that is where your knowledge is 3. Adjust stakes to the bankroll, up or down, from the actual number rather than from a feeling 4. Detect tilt. Sorting by date reveals whether losses cluster after losses. If they do, that is chasing, visible in your own data Review monthly. Daily review is reacting to noise, and reacting to noise produces the exact behaviour the record exists to prevent. One warning: a record only helps if it is complete. A log missing the bets you would rather forget is worse than no log, because it manufactures false confidence with the authority of a spreadsheet.