There is a number hiding in every bet you place. You never see it on the slip. The bookmaker never mentions it. But it is there, silently compounding, priced into every unit you stake. It is the overround, the margin embedded in the odds. The worked comparison below uses a 7.5% retail-pricing scenario and a 1.8% sharp-pricing scenario. These are modelling scenarios, not fixed platform margins. The gap shows how quickly pricing can preserve or consume an edge.
This article is about that number. What it is. How to calculate it. And what it means for your returns over a season, a career, a lifetime of betting. What follows is the math with worked odds, worked-through formulas, and a full 500-bet calculation. No abstractions. Every number gets calculated.
What the overround actually is
The overround is the gap between the sum of implied probabilities in a market and 100%. A fair market sums to 100%. A real market usually sums to something higher. That excess measures the price cushion in the market; it is not the same as the bookmaker's realised profit or the bettor's exact expected loss.
The formula is straightforward:
Overround = Sum of (1 / decimal odds) for every outcome.
Overround excess % = (Overround - 1) * 100.
Approximate proportional margin drag % = (1 - 1 / Overround) * 100.
Take a two-way Asian handicap market with both sides priced at 1.95:
Implied probability for each side: 1 / 1.95 = 51.28%. Sum them: 102.56%. Overround = 1.0256. The overround excess is 2.56%, while the proportional margin drag is approximately 2.50%.
Under the proportional-margin model, a flat bettor gives up approximately 2,50€ of expected payout for every 100€ wagered on this market. The difference looks small once. Across sustained turnover, it becomes a meaningful cost.
Now here is what that same calculation looks like on a typical Premier League match. The table below compares worked odds for an Arsenal vs Chelsea fixture at two pricing levels. The difference is not subtle.
Arsenal vs Chelsea: same match, two pricing levels
| Outcome | Sharp pricing example | Implied % | Retail pricing example | Implied % |
|---|---|---|---|---|
| Arsenal (Home) | 2.70 | 37.04% | 2.60 | 38.46% |
| Draw | 3.60 | 27.78% | 3.40 | 29.41% |
| Chelsea (Away) | 2.65 | 37.74% | 2.55 | 39.22% |
| Total implied probability | 102.55% | 107.09% | ||
| Overround | 1.0255 | 1.0709 | ||
| Overround excess | 2.55% | 7.09% | ||
| Proportional price drag | 2.49% | 6.62% |
The retail example carries nearly three times the overround excess of the sharp example. What does that mean per bet? If you stake 100€ on the correct outcome:
| Outcome | Sharp example pays | Retail example pays | Payout difference |
|---|---|---|---|
| Home Win | 270,00€ | 260,00€ | 10,00€ |
| Draw | 360,00€ | 340,00€ | 20,00€ |
| Away Win | 265,00€ | 255,00€ | 10,00€ |
A flat bettor does not notice 10€ missing from a winning payout. That is the mechanism. The margin is invisible on any single wager. But it is real, it is mathematical, and it compounds across every bet you place.
500 bets: where the margin stops being invisible
Let us run the numbers for a serious recreational bettor: 500 bets per season, 100€ flat stake, 50.000€ total turnover. We will look at two separate scenarios. The first shows the raw structural cost. The second shows what happens when a bettor with a genuine edge bets into those same prices.
Scenario A: Margin drag for a fair bettor
This is what the pricing structure extracts before any skill enters the equation. The calculation assumes the margin is distributed proportionally across outcomes. The exact margin drag formula is:
Margin Drag = Turnover × (1 − 1 / Overround)
If sharp pricing averages 1.8% overround while a retail alternative prices the same market at 7.5% overround:
Sharp pricing: 50.000€ × (1 − 1/1.018) = 884€ expected pricing drag.
Retail pricing: 50.000€ × (1 − 1/1.075) = 3.488€ expected pricing drag.
Difference: 2.604€ per season. Over ten years: 26.040€. This is the structural gap between the two pricing environments, before a single pick is made. For a quick estimate, the linear approximation Turnover × Margin% gives roughly 900€ and 3.750€ (difference: 2.850€). The exact values are more precise; the approximation is easier to do in your head.
Scenario B: What happens with a genuine edge
Now take a bettor with a +3% edge over fair odds. For this worked example, assume fair odds of 2.00, a true win probability of 51.5%, and that the margin is applied uniformly to implied probabilities. At each venue, the offered odds and expected returns look like this:
Sharp pricing (1.8% overround): The fair 2.00 becomes 2.00 / 1.018 ≈ 1.9646 on screen. Expected ROI: 0.515 × 1.9646 − 1 ≈ +1.18%. Over 500 bets of 100€ each: approximately +589€ per year.
Retail pricing (7.5% overround): The fair 2.00 becomes 2.00 / 1.075 ≈ 1.8605 on screen. Expected ROI: 0.515 × 1.8605 − 1 ≈ −4.19%. Over 500 bets: approximately −2.093€ per year.
Difference: roughly 2.682€ per season. The bettor's picks are identical. The strike rate is identical. The venue choice alone determines whether the edge compounds or is consumed by the pricing structure.
| Venue | Scenario Overround | Offered Odds (fair 2.00) | Expected Annual Result |
|---|---|---|---|
| Retail bookmaker | 7.5% | ~1.8605 | −2.093€ |
| Sharp-pricing scenario | 1.8% | ~1.9646 | +589€ |
Pricing structure is one of the largest controllable variables in a bettor's P&L. A bettor with a genuine edge can see that edge consumed by wider pricing, while the same worked picks at sharper prices produce a positive expected return. Margin drag is built into the price rather than charged as a separate fee.
What it takes to break even
The margin does not just affect returns when a bet loses. It changes the odds on the screen, and therefore the win rate you need to break even.
In a perfectly efficient market with fair odds of 2.0 (a true 50/50 proposition), each bookmaker's margin pushes the offered odds lower. If we assume the margin is distributed uniformly across all outcomes (an approximation, but a standard one for comparison), the odds you actually see on screen are:
Offered odds ≈ Fair odds / Overround
In the sharp-pricing scenario (overround 1.018), the fair 2.0 becomes roughly 1.9646 on screen. The break-even win rate at those offered odds: 1 / 1.9646 ≈ 50.9%.
In the retail-pricing scenario (overround 1.075), the same fair 2.0 becomes roughly 1.8605 on screen. The break-even win rate at those offered odds: 1 / 1.8605 ≈ 53.75%.
| Scenario | Scenario Overround | Fair Odds | Offered Odds | Break-Even Win Rate |
|---|---|---|---|---|
| Sharp-pricing scenario | 1.8% | 2.00 | ~1.9646 | 50.9% |
| Retail-pricing scenario | 7.5% | 2.00 | ~1.8605 | 53.75% |
That 2.85 percentage point gap means you need roughly 2.85 more winning bets per hundred at retail pricing just to break even, compared to sharp pricing. Same underlying event. Same fair probability. The margin turns a fair 50/50 into a bet you need to win nearly 54% of the time just to get your money back.
Why sharp bookmakers run tighter margins
The business model is different. Many traditional sportsbooks combine wider margins and promotional offers, then reduce limits on accounts they consider unprofitable. Pinnacle is built around tighter prices and higher volume, using incoming market action to refine its lines.
The mechanism starts with an early line and lower opening limits. Informed action moves the price and adds information. The line sharpens. Limits can rise as the market matures, giving Pinnacle the volume needed to operate tightly priced markets at scale.
Former Pinnacle trader Marco Blume described the model as "aggregating, in a smart way, the world's models, opinions, and information" to arrive at the most accurate price. That is not marketing. That is the mechanism.
Winners are not the enemy
Pinnacle's former trading lead Olavi Kuosmanen has described the model as volume-led and open to professional customers, including bettors who win over time. That explains why informed action is useful rather than automatically unwanted.
Sharp bettors contribute to price discovery. Their action helps the trading team test an early line and respond before limits rise. The product is designed to learn from that volume instead of treating every successful account as a problem.
Closing line value: the metric that separates process from results
CLV is the difference between the odds you bet and the odds the market closes at. A sharp closing line is a useful benchmark because it incorporates late information, mature liquidity, and the market's final price discovery.
The formula:
CLV % = (Your Odds / Pinnacle Closing Odds - 1) * 100
If you take Arsenal at 2.70 and the reference market closes at 2.62, your CLV is (2.70 / 2.62 - 1) * 100 = +3.05%. You secured the better price. If you take Arsenal at 2.60 and the reference market closes at 2.70, your CLV is -3.70%. The result may still win, but the entry price was below the closing benchmark.
Read the direction of your CLV across a broad sample:
| CLV direction | Process signal |
|---|---|
| Consistently positive | You are regularly securing a better price than the close |
| Near zero | Your entries broadly match the final market consensus |
| Consistently negative | Your selection or timing process needs work |
CLV usually becomes informative before profit and loss does, but there is no universal sample size that proves an edge. Market type, odds range, timing, and bet correlation all matter. Use CLV as a leading process indicator and profit as the realised outcome, then review both over a genuinely broad sample.
Calculate your own margin in 30 seconds
You do not need to trust me. You can do this on any market you are about to bet. Here is the method:
Step 1: Convert each outcome's odds to implied probability: 1 / decimal odds.
Step 2: Sum all the implied probabilities. That is your overround.
Step 3: (Overround - 1) * 100 = margin percentage.
Worked example, 1X2 market:
| Outcome | Odds | Calculation | Implied % |
|---|---|---|---|
| Home | 2.50 | 1 / 2.50 | 40.00% |
| Draw | 3.50 | 1 / 3.50 | 28.57% |
| Away | 2.80 | 1 / 2.80 | 35.71% |
| Total | 104.29% |
Margin = 4.29%. Expected payout per 1€ wagered = 1 / 1.0429 = 0,959€. A flat bettor faces 4.1 cents of margin drag for every euro staked.
Quick reference for common overround values:
| Overround | Approx. price drag | Worked pricing context |
|---|---|---|
| 1.018 | 1.77% | Low-margin reference scenario |
| 1.025 | 2.44% | Sharp 1X2 scenario |
| 1.045 | 4.31% | Competitive two-way pricing |
| 1.052 | 4.94% | Mainstream 1X2 example |
| 1.070 | 6.54% | Wider retail example |
| 1.100 | 9.09% | Niche or longshot-heavy market |
The market matters as much as the margin
Not all bet types carried the same average cost in the Hegarty and Whelan sample. The study-level figures below describe that dataset, not current quotes from a named sportsbook:
| Market type | Study-reported average | Illustrative retail range |
|---|---|---|
| Whole-goal Asian handicaps (0, +1, etc.) | ~2.93% | ~4-5% |
| Quarter-Goal AH (+0.25, etc.) | ~3.96% | ~5-6% |
| Half-Goal AH (+0.5, etc.) | ~4.73% | ~6-7% |
| 1X2 (Three-Way) | ~3.0% | ~5-7% |
Whole-goal Asian handicaps had the lowest average figure in that sample. Push settlement is one structural difference, but the observed gap also reflects the actual prices, leagues and selections in the data. Check the live overround on the exact market rather than assuming one line type is always cheaper.
The reframe
Most bettors ask one question after a bet: "Did I win?" Price-sensitive bettors add another: "Did I beat the closing line?"
You cannot control whether Arsenal score in the 87th minute. You can control which bookmaker you use, which market you bet into, and what price you accept. The three things that determine your long-term results, in order of importance:
- The margin on your bets. Wider margins raise the break-even hurdle on every position.
- The price you get relative to the market. Positive CLV is a useful process signal across a broad sample.
- The picks you make. Better pricing improves the starting point, but it cannot rescue poor selection. You still need a genuine edge and disciplined execution.
In the worked 1.8% versus 7.5% comparison, the pricing gap is a powerful variable in the bettor's P&L. For a fair bettor, the difference is 2.604€ per season. For a bettor with a +3% edge, it is 2.682€. It is also a variable you can improve immediately by choosing where you bet, not only what you bet.
Sources
PS3838 product information: AsianConnect PS3838 page. Margin formula and market examples: Pinnacle: How to Calculate Betting Margins. The 1.8% and 7.5% overround figures are illustrative scenario assumptions representing the sharp and retail ends of the pricing spectrum. The line-type averages are study-level observations and are not presented as current PS3838 or Pinnacle quotes. Actual overround varies by sport, market and timing. Break-even calculations assume uniform margin distribution across outcomes; actual margin distribution varies by market.