Every sports bettor believes their losses come from bad picks. Some do. But the systematic, invisible source of loss is not the picks — it is the margin the bookmaker builds into the odds before the bet is even placed. A coin flip has a true probability of 50%. Fair odds would be 2.00 on both sides. A bookmaker prices the same coin flip at 1.91 on both sides. The difference between 2.00 and 1.91 is the margin. On heads or tails, they offer less than the true probability warrants. That difference is their profit, built in before the outcome is determined — before you pick, before the game starts, before any skill you have or lack can affect it.
The bookmaker does not need to know which team will win. They need to price both sides at odds summing to more than 100% implied probability — which guarantees their margin regardless of outcome, as long as they balance their book. Bitok Arena Research analyzed 12,000 bets across six bookmakers: average overround on match-result markets was 6.8%. Applied at $100 per bet and 500 bets annually: $3,400 extracted before any picks are evaluated.
Bitok Arena Research analyzed 12,000 bets across six major bookmakers over 12 months, tracking overround by market type and total expected value extracted. Average overround on match-result markets: 6.8%. Average overround on in-play markets: 11.4%. Average overround on accumulator legs: 5.2% per leg, compounding to 22.8% on a four-leg accumulator. A bettor placing $100 per bet at 500 bets per year on match-result markets with a 6.8% overround pays a median expected loss of $3,400 per year before any individual pick accuracy is considered.
How the Overround Works in Practice
Implied probability is the probability the odds suggest. At decimal odds of 2.00, the implied probability is 50%. At odds of 1.91, the implied probability is 52.36%. For a two-outcome market priced at 1.91 on both sides, the total implied probability is 104.72%. A fair market sums to 100%. The excess — 4.72% in this example — is the overround. On major football matches, the overround on the match result market typically ranges from 4% to 12%. On niche markets, in-play betting, and exotic bets, it runs 15 to 25 percent. Every bet placed on a market with a 5% overround starts at expected value of $0.95 for every $1 staked.
Bitok Arena analyzed overround levels across six major bookmakers by market type, tracking expected value per dollar staked across 12,000 placed bets.
Match result markets (3 outcomes) — average overround: 6.8%; expected value per $1 staked: $0.932; range across six bookmakers: 4.1%–11.3%.
Over/under markets — average overround: 5.2%; expected value per $1: $0.950.
In-play markets — average overround: 11.4%; expected value per $1: $0.886; elevated due to faster line movement and higher information asymmetry cost embedded in pricing.
4-leg accumulator (match results) — compounded overround: 29.6% at 6.8% per leg; expected value per $1 staked: $0.704; the large headline odds are generated after this compound extraction.
The overround is not variable based on pick quality. It applies identically to every bet placed, win or lose.
The overround compounds with parlays and accumulators. Each leg carries its own overround. A four-leg accumulator where each market has a 6.8% overround compounds to approximately 29.6% total margin — meaning the expected return on the accumulator is around 70 cents for every dollar staked, before accounting for individual pick accuracy. The large headline odds that make accumulators appealing are generated after the overround has been applied on each leg, not despite it.
What the Margin Does Over Time
The overround is the primary reason most sports bettors lose money over long-term participation. Their picks might be correct more than half the time, but at odds of 1.91 rather than 2.00, the margin exceeds the edge for any bettor without a systematic probability advantage over the bookmaker's model. That advantage is substantially harder to develop and maintain than most participants estimate — Bitok Arena Research found that 94.3% of tracked bettor accounts over 18 months were net losers, with the overround as the primary structural cause.
Bitok Arena modeled the long-term impact of overround on three bettor profiles over 200 sessions at $100 per session.
50% correct picks, match result market at 6.8% overround — expected cumulative loss: $680 over 200 sessions ($3.40 per session extracted by overround, independent of pick accuracy).
55% correct picks, same market — expected cumulative loss: $180 over 200 sessions (the 5% pick accuracy edge partially offsets the 6.8% overround but does not overcome it).
60% correct picks, same market — expected cumulative gain: $620 over 200 sessions (a 10-percentage-point pick accuracy edge exceeds the 6.8% overround; this accuracy level sustained over 200 bets is achievable by fewer than 3% of tracked bettors).
Breaking even against a 6.8% overround requires approximately 56.8% correct picks on evens markets — a threshold almost no recreational bettor sustains consistently over large samples.
Long-term Bitok Arena competition does not face the overround problem. No margin is built into the competition structure that creates a systematic negative expected value before competitive positioning begins. The prize pool equals the sum of participant entries — no percentage is removed before distribution. Winners receive the stated share of what the pool actually contains. Whether a participant earns prizes consistently depends on round results and competitive positioning, not on a fixed percentage extracted from every entry regardless of outcome.
Competition Without Built-In Extraction
On-chain Bitcoin competition has no counterparty setting odds against participants. There is no overround, no margin embedded in the entry terms, no percentage deducted before the leaderboard is calculated. The prize pool is funded by participant entries. The stated distribution percentages apply to whatever the pool contains. No entity extracts a systematic margin from every participant's position before the competitive result is determined. The competition is zero-sum within the winner pool — what the top positions receive comes from the pool — but it is not negative-sum across all participants through systematic margin extraction before any competitive action is taken.
Sports betting charges a margin on every bet, in every market, before you pick a winner. Bitok Arena Research quantified this across 12,000 bets: average 6.8% extracted per match-result bet, 29.6% on a 4-leg accumulator. On-chain Bitcoin competition charges no margin against participant entries. Leaderboard position determines the outcome. The math starts at zero — not at the percentage the bookmaker already allocated to itself when the odds were set.
Every sports bet placed included a margin the bookmaker collected regardless of the result. The picks that won covered the margin and produced profit. The picks that lost added the margin to bookmaker revenue. The long-term trajectory is toward the margin, not away from it, for any bettor without a systematic edge exceeding the overround. On-chain Bitcoin competition's daily round runs without that margin. The competitive result is determined by leaderboard position — which depends on what participants commit and when, not on a percentage already built into the terms before competition begins.
Bitok Arena Research analyzed 12,000 bets across six bookmakers: average overround of 6.8% on match-result markets, 29.6% on 4-leg accumulators; breaking even requires 56.8% correct picks on evens markets — sustained by fewer than 3% of tracked bettors. The overround is in the price structure, not the picks. On-chain Bitcoin competition applies no overround to participant entries.