Being up over a period never proves that a method works. You need to cross-check profit, ROI, expected value, variance, sample size, drawdown and risk, plus the quality of the prices you got. This page points to each measure without going into detail.
The short answer
Being up over a period, even a long one, doesn't prove that a betting method is profitable. The final result is only a starting point. You read it alongside the number of bets, the average odds, the variance, the sample size, the dips in capital along the way and the risk taken.
I opened the page with that "+30% after 10 bets" because it shows the problem well: a clean number, with everything you'd need to read it missing.
Three things to keep in mind. A profit can come from variance. One measure is never enough. And a result says nothing about the quality of the decision: a good decision can lose, a bad one can win.
"I'm up after 20 bets, so my strategy works."
A small sample gives results very far from the real performance of a method. Twenty bets are almost never enough to tell a solid method from a favourable run that's down to chance.
What does "being profitable" mean?
Being profitable doesn't mean you made money this week. It means a method produces, on average and over time, a positive result that the variance of a sample that's too short can't explain on its own.
The nuance matters because the margin built in by operators makes the expected value negative for a bettor with no independent estimate. An isolated positive result is still possible. Demonstrated profitability needs far more evidence than a balance that's going up.
Profit is an amount, positive or negative, over a given period. ROI relates that amount to total staked. Bankroll growth tracks how the capital evolves, which also depends on stake sizes and how they change over time.
Three different numbers, often mixed up.
Profit, ROI and yield
ROI is the first measure a record gets reduced to: net profit relative to total staked. It lets you compare records with different volumes, which profit alone can't do.
It has limits. A high ROI on a few bets is worth less than a modest ROI on a large number of decisions.
The ROI calculation, its formula and its limits →
Expected result and actual result
ROI looks backward: it tells you what happened. Expected value looks forward: it describes what should happen on average, if the estimate used is accurate. The two answer each other.
What expected value measures, and its limits →
Why do results fluctuate?
Two identical decisions, taken under the same conditions, can produce opposite results. That's variance: the spread of results around an expected value. It exists even when you haven't made a single mistake.
Why a good decision can still lose →
How many bets before you can conclude anything?
There's no universal number of bets beyond which a result becomes reliable. What the answer depends on is known, though: average odds, assumed edge, variance and correlation between bets.
What decides how much data you need →
Can a positive performance still be very risky?
Two strategies that end on the same result may have exposed their capital to very different levels of risk. The final result says nothing about the path taken to get there.
Measuring the dips in a bankroll from its peak →
How do you measure risk?
You can control your exposure. You can't control the outcome of an event. Risk management works on the first, never on the second.
What you control, what you don't →
Can you lose your entire bankroll?
The bigger the share of capital exposed at each decision, the more an unfavourable run costs. That's what risk of ruin describes: a real risk that no favourable estimate removes.
How a bankroll can be exhausted →
Can this be compared to trading or the stock market?
Sports betting shares methods with trading and investing: estimating, comparing to a price, managing exposure. The economic nature differs. The comparison helps you borrow measurement tools. It turns misleading the moment it treats a bet as a financial product.
What the comparison allows, and what it doesn't →
How to track your data properly
None of these measures exist without a properly kept record. Date, odds taken, stake, result: without these columns, nothing above can be calculated.
The data to record for every bet →
The role of the bankroll
The bankroll organises financial exposure. It makes no strategy profitable. It sets the speed and size of capital swings, never their direction.
What a bankroll is and how it differs from a budget →
Why the price matters too
Two identical bets taken at different prices don't carry the same information. A value bet assumes an estimate independent of the odds. Closing line value assesses a price without waiting for the result. Odds movements show how that price forms before kickoff.
Understanding the result isn't enough. Understanding the price it was obtained at matters just as much.
Where OddScore fits in
Looking only at the final result hides part of the picture. OddScore compares the odds of several bookmakers, removes the built-in margin and tracks how they move up to kickoff. The AI reading sums up what the market says before the match. It isn't a bet to follow, and OddScore gives no staking advice and makes no promise of performance.
ROI is the first measure a record gets reduced to. It's also the one people misread most when they look at it alone.
Responsible gambling: no measurement method guarantees a positive result. Sports betting carries a risk of financial loss. Set yourself a budget, don't chase your losses, and use the deposit limits and self-exclusion tools available. Learn more about responsible gambling.
Put it into practice
Apply this guide to real odds.
See a market reading
A real match with its odds, movements and 0-100 signal.
Open the demoMargin calculator
Measure a bookmaker's margin on any market.
Calculate the marginDig into the market
Odds movements are only part of the story. Here are the next topics to read.
ROI and yield: calculation and interpretation
The first measure, and its limits.
Understand ROIExpected value: the theoretical edge of a bet
What the average expected outcome means, and what it doesn't.
Understand EVVariance: why results swing
A good decision can lose. A bad one can win.
Understand varianceHow many bets to judge a performance?
What decides how much data you need.
See the methodDrawdown: measuring the dips in a bankroll
Two strategies with the same final result can have gone through very different dips.
Understand drawdownRisk management: what can you actually control?
Exposure you can control. The outcome of an event you can't.
Understand riskStaking plans: comparing the methods
Each plan is judged on risk of ruin and drawdown, never on a promised return.
Compare staking plansKelly criterion: what the formula sizes
Geometric growth rather than expected gain, and the f* ≤ 0 case.
Understand the Kelly criterionRisk of ruin: exhausting a bankroll
How a badly sized exposure can wipe out a capital.
Understand risk of ruinSports betting, stocks and crypto
Methods that overlap, economic products that differ.
Compare the methodsSports betting winnings tax
The player's regime, jurisdiction by jurisdiction, sourced and dated.
See the comparisonFrequently asked questions
Can you be profitable at sports betting?
On a single decision it's mathematically possible, if your estimate is independent and accurate. But the margin built in by operators makes the expected value negative on average, and no staking method changes that.
How do you calculate profitability?
Start with ROI: net profit divided by total staked. On its own it doesn't tell you how solid the result is. Read it alongside the number of bets, the average odds and the variance you've seen.
What ROI counts as good?
No threshold means anything without volume and odds. +20% on 10 bets and +3% on 2,000 bets tell very different stories, and the second is usually more solid than the first.
How many bets before you can judge your results?
There's no universal number. It depends on the odds you play, the edge you assume and the variance it produces. A small edge at short odds needs far more bets than a clear edge at long odds.
Does good bankroll management guarantee a positive performance?
No. Good management limits exposure and the size of capital swings. It creates no positive expected value and fixes no wrong estimate.
What's the difference between ROI and profit?
Net profit is an amount. ROI relates that amount to total staked, so you can compare records with different volumes. A big profit can come with a small ROI, and the other way round.
Is sports betting the same as trading or the stock market?
Both share methods: you estimate, compare to a price and manage exposure. But a bet isn't an asset you hold, and it guarantees no return.
Sources & methodology
This page points to the measures that make up a performance (ROI, expected value, variance, sample size, drawdown, risk). It draws on the academic literature on expected loss rates and the efficiency of betting markets, on OddScore's odds-analysis methodology, and on the UK Gambling Commission and BeGambleAware for responsible gambling.
- Each measure is treated as a partial signal, never enough on its own to judge a performance.
- Each notion (ROI, expected value, variance, sample size, drawdown, risk) points to its own page instead of being developed here.
- The margin built into operators' odds makes expected value negative on average, whatever the management method applied.
- A single positive result is never presented as proof that a method works.
- Whelan, K. (2025). Estimating expected loss rates in betting markets: theory and evidence. Applied Economics.
- Hegarty, T. & Whelan, K. (2026). Market structure and prices in online betting markets. Oxford Economic Papers, 78(1), 90-113.
- UK Gambling Commission, advice for players
- BeGambleAware, support and advice