How Fractional Betting Odds Work With Real Payout Examples
September 7, 2026 3 Views
Fractional odds describe potential profit as a ratio to the stake. For a broader view…
What Is Vig in Sports Betting and Why It Matters
The Cost Behind the Odds 
Load More - Link
A winning record can still produce a losing balance.
Contents show
A bettor makes 100 wagers at -110 and correctly picks 52. Each win earns $100, producing $5,200. But the 48 losses cost $110 apiece, totaling $5,280. Despite being right more often than wrong, the result is an $80 loss.
Top Crypto Offers for September 2026
Get a 250% Up to $2,500 With Code SPWELCOME1
Full terms and conditions apply. 18 + only.
18+ Full terms and conditions apply. Crypto banking - Bitcoin, BitcoinCash, Litecoin, Cardano, BNB, ETH, USDT, USDC
100% Deposit Bonus up to 300 USDT
18+ only. Full terms apply.
That gap comes from the vig—the sportsbook’s built-in charge. In a balanced market, both sides may be listed at -110 even though only one can win. Each price implies a 52.38% probability, so the combined implied probability is 104.76%. The amount above 100% represents the bookmaker’s margin. At -110, a bettor must win about 52.38% of wagers just to break even, not merely half.
Key terms
Vig, overround, and hold
- Vig or juice
The margin built into sportsbook odds rather than charged as a separate fee.
- Implied probability
The win probability represented by a price. At -110, it is 52.38%.
- Overround
The amount by which all implied probabilities in a market total more than 100%.
- Hold
The sportsbook’s actual revenue as a percentage of accepted wagers over a given period.
Inside the odds
Where the sportsbook’s margin appears
The price reveals the cost before the result is known.
Vig becomes visible after converting every offered price into an implied probability. In a two-sided market priced at -110 on both outcomes, each side implies a 52.38% chance. Together they total 104.76%, leaving an overround of 4.76 percentage points above a fair 100% market.
The cash flow makes the same idea concrete. Suppose one bettor risks $110 on Team A and another risks $110 on Team B. The sportsbook collects $220, then returns $210 to the winner: the original $110 stake plus $100 in winnings. With perfectly balanced action, the book keeps $10, equal to 4.55% of the $220 handled.
The 4.76% overround and 4.55% balanced-book margin differ because they use different denominators. Overround measures excess implied probability; the margin calculation compares revenue with total stakes.
Realized hold is not the same as vig. Vig is embedded in the posted prices before wagers settle. Hold is the amount the sportsbook actually retains, and it can rise or fall with uneven betting, winning outcomes, promotions, and line movement.
This pricing cost is central to understanding how betting odds, lines, and value work together. Even when a bettor selects winners half the time in a balanced -110 market, the built-in margin makes the long-run result negative.
Worked example
How the vig appears in a -110 market
- Convert the first side to probability
For negative American odds, divide the odds value by that value plus 100. The implied-probability calculation is 110 ÷ (110 + 100) = 0.5238, or 52.38%.
- Repeat for the other side
If both outcomes are priced at -110, the second side also carries an implied probability of 52.38%.
- Add the two probabilities
Together, the prices imply 52.38% + 52.38% = 104.76%. A fair two-outcome market would total 100%.
- Measure the overround
Subtracting 100% leaves a 4.76-percentage-point overround. That excess is the margin embedded in the posted prices.
“Percentage points” is precise here: the implied probabilities total 4.76 points above 100%.
Overround is not guaranteed revenue
The 4.76-point overround should not be read as the sportsbook’s expected share of every dollar wagered. It describes the pricing of the market, not the final betting results.
Actual hold depends on how much money lands on each side, which side wins, and whether wagers receive different prices. Even with equal stakes at -110, the book takes $220 and pays $210 to the winner, retaining $10—about 4.55% of handle, not 4.76%. An unbalanced market can produce a higher profit, a lower profit, or even a loss.
Removing the vig with proportional normalization
A standard estimate, not a uniquely correct answer
A common way to estimate fair probabilities is proportional normalization. Each side’s implied probability is divided by the total implied probability:
No-vig probability = implied probability ÷ market total
For a balanced market priced at -110 on both sides, each side implies 52.38%. The market total is 104.76%, so normalization gives:
- Side A: 52.38% ÷ 104.76% = 50.00%
- Side B: 52.38% ÷ 104.76% = 50.00%
The calculation removes the overround while preserving the relative weight of each side.
An uneven moneyline
Consider a favorite at -150 and an underdog at +130:
| Side | Implied | Proportional no-vig |
|---|---|---|
| -150 favorite | 60.00% | 57.98% |
| +130 underdog | 43.48% | 42.02% |
The implied probabilities total 103.48%. Dividing each by that total produces fair odds of roughly -138/+138.
These figures are estimates because proportional normalization is only one margin-removal model. If the 3.48 percentage-point overround were instead split equally between the two sides, the estimates would be 58.26% and 41.74%. Proportional normalization assumes the margin inflates both probabilities at the same relative rate; equal subtraction assumes the same percentage-point adjustment. Odds alone do not reveal which assumption best matches the sportsbook’s pricing process.
The price sets the required win rate
A small difference in odds becomes meaningful across a large betting sample.
A bettor can pick winners more often than losers and still lose money. The reason is simple: at negative odds, each loss costs more than each win returns. The break-even rate comes from the price, not from a universal 50% threshold.
For American odds of -A, the required win rate is A ÷ (A + 100). That produces three distinct benchmarks:
| Odds | Break-even win rate |
|---|---|
| -110 | 52.38% |
| -105 | 51.22% |
| +100 | 50.00% |
Consider 100 bets with 53 wins and 47 losses, each sized to win $100. The selections and results remain identical; only the purchase price changes.
| Price | Amount risked per bet | Net result |
|---|---|---|
| -110 | $110 | +$130 |
| -105 | $105 | +$365 |
| +100 | $100 | +$600 |
At -110, the 53 winners earn $5,300 while the 47 losses cost $5,170. Reducing the price to -105 saves $5 on every losing wager, adding $235 to the same record. At even money, wins and losses carry equal dollar weight.
This is why line shopping matters. Five cents of price improvement cannot turn a losing pick into a winner, but repeated savings can materially change long-run returns.
Price changes value, not the result
A better number does not improve the chance that one wager wins. It reduces the win rate needed for the same betting strategy to become profitable over time.
Standard juice is not universal
Market conditions can move the margin far beyond the familiar -110 price.
The familiar -110 on both sides is a convention, not a rule. It is common in major pregame point-spread and totals markets, where competition and heavy betting volume help keep prices relatively tight.
Margins can differ by sportsbook and sport. This helps explain why odds vary between sportsbooks: each operator has its own customer flow, risk tolerance, trading models, and desired return. A book facing lopsided liability may shorten one price to discourage further action, while a competitor may offer better odds to attract it.
Wider margins often appear where prices are harder to estimate or bets are less likely to balance:
- Niche sports and low-liquidity markets may have fewer reliable data points and less competition.
- Futures tie up money for longer and expose the book to changing circumstances.
- Live bets must account for fast-moving information and execution delays.
- Player props can carry limits, injury uncertainty, and specialized modeling risk.
- Same-game parlays require the sportsbook to price correlations between selections.
Timing matters too. Early lines may include extra caution; mature markets can tighten as information and money arrive. House rules—including push, void, dead-heat, and overtime treatment—also affect a wager’s real value, even when the displayed odds look similar.
Reduce the cost
Price shop before placing the bet
The same selection can produce a better return at another sportsbook.
Price shopping means checking an identical wager at several sportsbooks before placing it. This is one of the simplest forms of saving money by comparing sportsbook prices, because it improves the potential return without requiring a different prediction.
| Same selection | Price A | Price B | Better price |
|---|---|---|---|
| Team to win | -110 | -105 | -105 |
| Player over 2.5 shots | +110 | +120 | +120 |
With negative odds, the number closer to zero is better. A $100 stake at -105 produces about $95.24 profit, compared with $90.91 at -110. With positive odds, the higher number is better: a $100 bet at +120 earns $120 profit rather than $110 at +110.
The comparison must be exact. Team -3 is not the same wager as team -3.5, and an overtime-inclusive moneyline may differ from a regulation-only market. Settlement rules, listed pitchers, player participation requirements, and maximum stakes can also affect whether two prices are truly equivalent.
A practical check takes only a few steps:
- Match the event, market, selection, and line.
- Confirm the sportsbook’s grading rules.
- Choose the highest positive odds or least-negative price.
Odds boosts and promotions can improve a price as well, but the headline number is not enough. Eligibility, opt-in requirements, stake caps, expiration times, excluded markets, and withdrawal conditions should be checked. A boost from +110 to +120 is useful when it applies cleanly; a tiny limit or restrictive terms may make the ordinary market price more valuable.
Betting checklist
Is the estimated edge worth the price?
- Set a fair probability
Estimate the outcome’s chance before considering the sportsbook’s odds. This separates the prediction from the price being offered.
- Find the break-even rate
Convert the odds into the win rate required to avoid losing money over time. At -110, that hurdle is 52.38%.
- Measure the cushion
Subtract the break-even rate from the estimated probability. A 54% estimate at -110 leaves only a 1.62-point edge—possibly too narrow to withstand modeling error.
- Search for a better price
Compare the exact market across sportsbooks. Prediction is uncertain, but choosing -105 instead of -110 when both are available is a controllable improvement.
- Pass when the edge disappears
Recheck limits, rules, and promotions. If those costs erase the estimated advantage, declining the wager is the rational choice.
Conclusion
- Small estimated edges deserve extra skepticism because minor forecasting errors can reverse expected value.
- Bet frequency matters less than consistently refusing unfavorable prices.
Vig is best treated as a cost of entry. A prediction has betting value only when its estimated edge clears that cost with enough room for uncertainty.
The outcome cannot be controlled; the accepted price can. Consistently taking the best available number lowers the required win rate and gives any genuine forecasting edge more room to matter.
