No-Vig Odds Explained: How to Calculate True Probability from Any Odds

What Is the Vig (and Why You Must Remove It)

Every bookmaker embeds a margin (also called vig, juice, or overround) into their odds. This margin is how they guarantee profit regardless of the outcome. In a perfectly fair two-way market, both sides would have 50% probability, and the odds would be 2.00 / 2.00. But a bookmaker might offer 1.91 / 1.91 instead.

Converting those odds to implied probability: 1/1.91 = 52.36% per side. Total: 52.36% + 52.36% = 104.72%. The probabilities sum to more than 100% — the extra 4.72% is the bookmaker's margin. This means the raw odds overstate the probability of every outcome. To find the true probability, you must strip the vig out.

Without de-vigging, you cannot accurately calculate expected value, measure Closing Line Value, or determine whether a bet has genuine edge. De-vigging is the foundation of quantitative sports betting.

The Proportional Method (Most Common)

The proportional method is the most widely used de-vigging technique. It assumes the bookmaker applies their margin equally across all outcomes — each side's implied probability is inflated by the same proportional amount.

Or equivalently:

This is the method used by most odds comparison sites, professional bettors, and academic research. It's fast, transparent, and accurate for markets where the bookmaker's margin is distributed relatively evenly.

Worked Example: 2-Way Market (Pinnacle NBA)

Pinnacle offers the following moneyline odds on an NBA game:

Team A: 1.85 | Team B: 2.05

Step 1: Convert each to implied probability.

Step 2: Sum the implied probabilities to find the overround.

The margin is 2.83% — typical for Pinnacle on a major market.

Step 3: Divide each implied probability by the overround.

Step 4: Convert back to fair odds.

Now you have the no-vig prices. If any other bookmaker offers odds higher than 1.903 on Team A or higher than 2.108 on Team B, that bet has positive expected value relative to Pinnacle's efficient market.

Worked Example: 3-Way Market (Football)

Football (soccer) match-result markets have three outcomes: Home Win, Draw, Away Win. The proportional method works identically — just with three terms instead of two.

Pinnacle odds: Home 2.40 | Draw 3.30 | Away 3.10

Step 1: Implied probabilities.

Step 2: Total overround.

Margin: 4.23% — slightly higher than two-way markets because three-way markets carry more uncertainty.

Step 3: De-vig each outcome.

Step 4: Fair odds.

If a soft bookmaker offers 2.65 on the Home Win while Pinnacle's fair price is 2.501, that's a 5.95% edge — a strong value bet worth sizing with the Kelly Criterion.

The Shin Method

The Shin method (Shin, 1991, 1993) offers a more sophisticated approach to de-vigging. It assumes that the bookmaker's margin is not distributed equally across all outcomes. Instead, it assumes that part of the overround exists to protect the bookmaker against insiders — bettors with private information about the likely outcome.

The key insight: the bookmaker inflates the probability of favourites less and inflates longshots more. This is because insiders are more likely to bet on longshots (where the information asymmetry creates the largest profit opportunity).

The Shin method solves iteratively for a parameter z (the proportion of insider trading) and adjusts each probability accordingly. The math is more involved:

Where q_i is the raw implied probability, n is the number of outcomes, and z is solved numerically. In practice, the Shin method produces results very close to the proportional method for sharp bookmakers like Pinnacle (where margins are small), but differs more significantly for high-margin bookmakers.

The Power Method

The Power method (also called the odds ratio method) finds a single exponent k such that raising each side's implied probability to the power k makes them sum to exactly 100%.

The value k is solved numerically (typically k is slightly less than 1). The de-vigged probabilities are p_A^k and p_B^k. This method shares the Shin method's property of adjusting favourites and longshots asymmetrically — favourites are adjusted less, longshots more.

For most practical purposes with sharp bookmaker odds, the three methods produce very similar results. The proportional method is simplest and most transparent, which is why it's the default in most tools including our No-Vig Calculator.

Why Pinnacle No-Vig Is the Gold Standard

Not all no-vig lines are equal. De-vigging a soft bookmaker's odds gives you that bookmaker's implied probability minus their margin — but their odds may not reflect true probability in the first place because they limit sharp bettors and price based on liability management rather than probability.

Pinnacle is different for three reasons documented in our Pinnacle benchmark analysis:

1. Accepts all bettors. Sharp syndicates, algorithmic traders, and professional bettors all trade freely on Pinnacle. This means their odds reflect the collective intelligence of the sharpest money in the world.

2. Lowest margins. At 1.5-3% on major markets, there's less vig to remove, meaning the de-vigged odds are closer to reality.

3. Academic validation. Pinnacle's closing odds correlate with actual outcomes at r² = 0.997 (Buchdahl, 2003). No other publicly available odds source achieves this level of calibration.

How OdinPicks Uses No-Vig to Find Value

The OdinPicks engine's core process for detecting value bets relies directly on no-vig calculations:

1. Capture Pinnacle's current odds for every market across supported leagues.

2. Apply proportional de-vigging to extract the no-vig implied probability for each outcome.

3. Compare this no-vig probability against the odds available at soft bookmakers (Bet365, Betway, 1xBet, and others).

4. When the soft book's odds imply a probability meaningfully lower than Pinnacle's no-vig estimate, an edge exists.

5. Cross-validate with AI-driven contextual analysis (form, injuries, rest, head-to-head) to confirm the edge is genuine.

6. Publish only bets with +3% EV or higher, sized using fractional Kelly.

This process is transparent and verifiable. Every pick is timestamped with a SHA-256 hash, and CLV tracking confirms whether our no-vig estimates are accurate over time.

When to Use No-Vig Calculations

1. Checking If a Bet Has Value

The most common use case. De-vig Pinnacle's odds, convert the no-vig probability to fair odds, and compare with the price your bookmaker is offering. If your bookmaker's odds exceed the fair odds, you likely have a +EV bet.

2. Measuring Closing Line Value (CLV)

CLV should be measured against the no-vig closing line, not the raw closing odds. Raw closing odds include the vig, which can mask whether you truly beat the market. De-vig the closing line first, then calculate CLV = (Your Odds / No-Vig Closing Odds) - 1.

3. Calculating True Probability for Models

If you're building your own statistical model, Pinnacle no-vig probabilities serve as the best available training labels. When your model's probability estimate diverges from Pinnacle's no-vig by more than 3%, that's worth investigating as a potential edge — or a potential model error.

4. Evaluating Tipster Claims

When a tipster claims +EV picks, de-vig the odds they recommended against Pinnacle's closing no-vig. If the tipster's recommended odds don't consistently exceed Pinnacle's fair price, their “edge” is likely noise, not skill.

Common Mistakes in De-Vigging

Using soft book odds as the benchmark

De-vigging Bet365's odds does not give you true probability — it gives you Bet365's probability estimate minus their margin. Their estimate may be wrong to begin with because their odds are shaped by liability management, not price discovery. Always de-vig the sharpest available source (Pinnacle).

Ignoring the method for high-margin books

The proportional method works well for low-margin books (Pinnacle, Betfair Exchange). For high-margin books (8%+ overround), the Shin or Power method may produce more accurate results, because high-margin books tend to inflate longshot probabilities disproportionately.

Treating no-vig as certainty

No-vig probabilities are the best available estimate, not ground truth. A no-vig probability of 55% doesn't mean the true probability is exactly 55% — it means the market's best guess, after removing the margin, is 55%. There is still uncertainty, which is why risk management through fractional Kelly sizing remains essential.

Frequently Asked Questions