How Efficient Are Betting Markets?

A market is efficient with respect to some information if prices already reflect it. The practical version for sports markets is narrower: can a systematic approach, applied over many events, beat the closing price often enough to overcome the margin built into prices.

Three facts shape the answer.

Closing prices aggregate a lot. By the time a market closes, it has absorbed news, line-up confirmations, weather, model updates from many operators, and money from participants with strong records. Empirical work across sports repeatedly finds closing prices to be among the best available predictors, which is why they are used as an evaluation benchmark rather than as something to beat casually.

Margin sets the bar. Prices include a margin, so an approach must be better than the market by more than that margin to profit. A model that is genuinely slightly better than the closing price can still lose money.

Liquidity varies enormously. A main market in a major league attracts large volume and sharp participation. A player prop in a minor competition may be priced largely by model with little corrective money. Efficiency tracks that difference closely.

Opening versus closing

Favourite longshot bias. Across many sports and decades of data, longshots have tended to be overpriced relative to their true chances and heavy favourites underpriced. It is one of the most consistently reported anomalies in the literature, though its size varies by sport and era and it does not automatically survive margin.

Thin markets. Derivatives, player props, lower divisions, and niche sports carry wider margins and less corrective money, which leaves more room for mispricing and also means lower limits and higher variance.

Slow reaction to certain information. Markets can be slower to price information that is public but hard to process, such as detailed personnel or tactical changes, compared with headline injury news.

Behavioural patterns. Popular teams, recent results, and narrative-heavy matchups can attract money that moves prices away from model estimates, particularly at recreational-focused operators.

Structural constraints. Even a real edge is limited by what you can actually stake. Accounts that win consistently are frequently limited or closed, which is a constraint on strategies rather than evidence about prices.

The honest summary is that markets are efficient enough to make consistent profit hard and inefficient enough that measurable deviations exist, mostly where liquidity is lowest.

In-play markets

Use closing prices as the benchmark. Compare your own estimates against the closing price for the same market, not against opening prices or against outcomes.

Measure in price terms first. Whether you beat the closing price is a continuous measurement that stabilises far faster than profit, which is dominated by outcome variance.

Account for margin explicitly. Remove it consistently when converting prices to probabilities, and state which method you used, since methods differ on heavy favourites.

Segment by liquidity. Run the analysis separately for main markets, derivatives, and props. Aggregating them hides the only places where an effect is likely.

Use point-in-time data. Evaluate with the prices and information available when the estimate was made. Using final line-ups or closing prices as model inputs for a historical test invalidates the result.

Check sample size honestly. Small edges need large samples to distinguish from noise, and most published claims rest on far less data than the effect size requires.

Record what you could actually have staked. An edge at a price you could not have obtained, or at limits far below your assumed stake, is not an edge in practice.

This page describes data and method and is not betting advice.