Statistics

Betting Yield Explained: ROI, Hit Rate and Worked Examples

Betting Yield Explained: ROI, Hit Rate and Worked Examples — ABV Tips
On this page
  1. What is yield in betting?
  2. Betting yield vs ROI vs bankroll growth
  3. Yield vs hit rate: why more winners can still lose money
  4. How to calculate yield with voids and partial settlements
  5. Gross yield, fees and the result a reader could reproduce
  6. Comparing yield across tipsters fairly
  7. A spreadsheet structure for an honest betting record
  8. FAQ
  9. Sources and calculation notes

Short answer: Betting yield is net betting profit divided by total stakes, expressed as a percentage. A profit of 12 units from 200 units staked produces a 6% yield. Hit rate measures how often selections win; it does not measure profit. In betting reports, “ROI” often means the same as yield, but it can also mean return on starting bankroll. Always check the denominator, settlement rules and costs before comparing records.

This guide explains betting yield with hypothetical examples. It separates three questions: how much was won or lost, how much was staked, and how much capital was available. None of the examples describes an expected return from ABV Tips or any other service.

What is yield in betting?

The standard turnover-based calculation is:

Yield (%) = net betting profit / total stakes × 100

Total stakes is turnover: the sum of the amounts placed on the selections included in the report. It is not the initial balance and it is not the number of predictions. Money can be staked, returned and staked again, so turnover may substantially exceed the starting bankroll.

For an ordinary winning bet with decimal odds, gross return includes the original stake. Net profit excludes it:

Winning bet profit = stake × (decimal odds − 1)

A losing bet has profit equal to minus its stake. A fully refunded bet has zero betting profit before any separate costs. The odds guide explains the underlying price formats.

A complete worked example

Assume four independent entries in a hypothetical ledger, with no commission or tax:

SelectionStakeDecimal oddsResultGross returnNet profit
A2 units2.00Win4 units+2 units
B1 unit3.00Loss0−1 unit
C2 units1.80Win3.6 units+1.6 units
D1 unit2.20Loss0−1 unit
Total6 units—2 wins, 2 losses7.6 units+1.6 units

Yield is 1.6 / 6 × 100 = 26.67%. Hit rate is 2 / 4 × 100 = 50%. The high yield is simply the result of this tiny illustrative sample, not a sensible performance expectation.

Betting yield vs ROI vs bankroll growth

The term betting ROI calculation is ambiguous unless the report defines its capital base. Two reports can display different percentages while using the same ledger correctly.

MetricCalculationWhat it describes
Betting yield / turnover ROIProfit ÷ stakesProfit per unit staked
Starting-bankroll returnProfit ÷ starting bankrollProfit relative to initial capital
Hit rateWins ÷ decided selectionsFrequency of winning selections
Profit in unitsReturns minus stakesAbsolute result under the stated unit convention

Suppose a starting bankroll is 100 units, total stakes are 200 units and net profit is 12 units. Yield is 6%; return on the starting bankroll is 12%. With no deposits or withdrawals, the ending bankroll is 112 units. The 12 units have not been compounded into a 6% return on the bank.

Deposits, withdrawals and changing stake sizes complicate bankroll comparisons. A higher end balance can come from added money rather than profitable selections. Compare betting performance separately from cash movements.

Yield vs hit rate: why more winners can still lose money

For equal stakes at one fixed decimal price, the break-even win probability before separate costs is 1 / odds. At 1.50 it is 66.67%; at 2.50 it is 40%. With varying prices, there is no single meaningful break-even percentage based on hit rate alone.

Two equal-stake records

Record A contains 100 one-unit bets, all at 1.50. There are 65 wins and 35 losses. Returns are 65 × 1.50 = 97.5 units; profit is −2.5 units; yield is −2.5%. The 65% hit rate is insufficient at these prices.

Record B contains 100 one-unit bets, all at 2.50. There are 45 wins and 55 losses. Returns are 112.5 units; profit is +12.5 units; yield is +12.5%. It has fewer winners and a better observed financial result. Neither sample proves a repeatable advantage.

For changing odds, calculate each entry's profit and add the results. Multiplying win rate by the average odds of all bets can be wrong because the odds of winners and losers may differ. This matters when assessing a tipster's track record.

How to calculate yield with voids and partial settlements

Voided bets and pushes

Reports differ on whether fully voided stakes enter turnover. Either convention can be described transparently, but comparisons require the same one. If profit is 10 units, decided stakes total 100 and voided stakes total 20, yield is 10% when voids are excluded and 8.33% when they are included. Profit has not changed.

Keep voids visible in the ledger, give them zero profit, and state whether their stakes are in the denominator. Pending bets should remain outside a settled-results yield; display their stake exposure separately. Never count an unsettled selection as a win because its live position is favourable.

Half wins, half losses and cash-outs

At a quarter-line handicap, a two-unit stake may split into two one-unit parts. One part can win and the other push. At odds 2.00, that settlement returns 3 units in total and produces +1 unit of profit, not a full two-unit win. Record the actual settlement, with the full two-unit stake counted once. See Asian handicaps for line mechanics.

A cash-out also needs the actual amount received. If a two-unit position is closed for 1.4 units, the realised result is −0.6 units. Its original possible payout is irrelevant to the settled result. Document whether the record tracks published selections at normal settlement or a separate cash-out strategy.

Gross yield, fees and the result a reader could reproduce

Commission, subscriptions and other costs can change the practical result. Keep their treatment explicit rather than mixing several definitions under one percentage.

Suppose stakes total 500 units and selection profit is 20 units. Gross selection yield is 4%. If commission is 3 units and the subscription cost allocated to this period is 10 units, the after-cost result is 7 units, or 1.4% of turnover. Tax treatment depends on the relevant jurisdiction and is not assumed here.

Subscription fees are paid regardless of whether a user places every published pick. Someone following fewer selections may have a different cost burden. Our free vs premium tips guide develops that comparison without assuming that payment improves results.

Achievable odds matter too. A published one-unit win at 2.20 produces 1.20 units of profit; the same selection at 2.00 produces 1 unit. Calculate a personal ledger using the actual price obtained. Do not copy a service's headline yield into a personal result.

Comparing yield across tipsters fairly

A useful comparison holds the following reporting choices as consistent as possible:

  • The same date window, with losing periods included.
  • All selections in the defined category, published before kickoff.
  • The same settlement and void policy, with corrected entries explained.
  • Flat-stake results alongside any variable-stake results.
  • Achievable timestamped prices, rather than a best price selected after the event.
  • Counts, turnover, average odds and maximum drawdown alongside profit and yield.
  • Separate labels for live records, simulations and backtests.

An overall yield is turnover-weighted. If one month makes 10 units from 100 staked and another loses 5 from 900, combined yield is 5 / 1,000 = 0.5%. Averaging their monthly percentages gives 4.72%, which does not describe the combined record.

Why there is no universal “good yield”

The same observed yield can have very different uncertainty depending on sample size, odds, stakes and dependence between selections. Ten bets and ten thousand bets do not provide the same information. Several markets on one match also share risk; treating every row as independent can exaggerate confidence.

As a deliberately simplified example, with 100 independent one-unit bets at odds 2.00 and a true win probability of 50%, the standard deviation of total profit is 10 units. The corresponding yield standard deviation is 10 percentage points. A short record can therefore move far from break-even through chance alone. This calculation does not apply unchanged to varied odds or correlated picks.

Repeatedly searching for the best league, month or score threshold after seeing results also favours lucky subsets. Record the rule in advance and assess it on later data. Research on backtest overfitting explains this general selection problem in financial strategy testing; it is not evidence for a particular football tipster.

A spreadsheet structure for an honest betting record

Use one row per selection and keep these columns: publication time, event, market and line, odds, stake, settlement, return, profit, costs and record type. Store a stable pick ID so edits or duplicates can be reconciled.

For a simple ledger with stake in column C, return in D and settled profit in E, enter =D2-C2 for each settled row. For a range containing only settled entries, yield is =SUM(E2:E101)/SUM(C2:C101), formatted as a percentage. Keep pending rows outside that range or filter explicitly by settlement status. A zero turnover denominator is undefined, not a 0% result.

Keep a second summary for costs and cash movements. This makes it easier to separate service performance from personal staking and bankroll management.

FAQ

Is betting yield the same as ROI?

Often, when ROI means net profit divided by stakes. It is different when ROI means profit divided by starting capital. Check the formula beside the figure.

Does a high hit rate guarantee a positive yield?

No. The odds and stake distribution determine whether the winning returns cover losing stakes and costs.

Should voided bets count in yield?

State the convention. Keep voids in the record, and show whether their stakes enter turnover. Compare records only after aligning that choice.

Can past yield predict future profit?

Past yield is an observation. Its usefulness depends on record integrity, uncertainty, changing prices and whether the selection method survives a fresh evaluation. It is never a guarantee.

Sources and calculation notes

The formulas and numerical ledgers above are educational calculations, not reported betting results. ABV Tips publishes prediction content and has a commercial interest in the subject.

18+: Betting involves financial risk. Historical records do not guarantee future outcomes. If gambling is causing harm, support is available through Gambling Therapy.