How to Read Betting Splits

Ticket percentage. The share of individual wagers placed on each side. One hundred small bets and one large bet count as one hundred and one tickets, so this number reflects how many people chose a side, not how much is at risk.

Handle percentage. The share of total money staked on each side. A single large wager can dominate handle while barely registering in tickets.

Reading them together. When both numbers are similar, bet sizes are broadly comparable on each side. When they diverge, the side with a higher handle share than ticket share is receiving fewer, larger bets. That pattern is what people usually mean when they refer to sharp versus public money, though the labels are inferences rather than facts in the data.

What is reported. Some sources publish both percentages, some only tickets, and some only for selected markets. Coverage is typically best for major leagues and main markets, and sparse for derivatives and smaller competitions.

Whose bets they are. Splits almost always come from one operator or a small group of them. That operator's customer base is not the whole market, and sharper accounts may be limited or absent there entirely.

Why the source matters so much

Splits on their own describe a position. Their analytical value comes from comparison with what prices did.

Splits plus line movement. If the line moves toward the side with fewer tickets, the money on that side is larger or more respected than the count suggests. This is what reverse line movement describes, and it is only visible when both datasets are timestamped.

Splits plus closing prices. The market's final estimate is the closing price. Comparing where splits sat against where the price finished says more than either alone.

Splits across books. Where more than one operator publishes, differences between them describe different customer bases rather than disagreement about the game.

Splits over time. A single snapshot near kickoff hides how the position built. Sequential captures show whether money arrived early or late, which is a different signal.

Beware of the timing gap. Splits are often published with a delay, so a split captured at the same moment as a price may describe an earlier state of the market.

Storing splits alongside odds

Claims about fading the public. The idea that betting against heavily backed sides is profitable is widely repeated and contested in the evidence. Any test of it needs closing prices, a long record, and honest accounting for margin, and results depend heavily on the period and market studied. Treat a split as one input to a question, not an answer.

Inference about individual bettors. A high handle share does not identify who bet or why. Attributing it to sharps, syndicates, or a single account is speculation unless the source says so.

Market-wide conclusions. One operator's customer base is not the market.

Causation from correlation. Splits and line movement often move together because both respond to the same news. That does not establish which drove which.

Using them without recording them. Split data is frequently overwritten rather than archived. If you intend to analyse it, capture and store it yourself with timestamps and the source name, because retrospective access is usually unavailable.

A practical use. The most defensible use is descriptive: recording how positions and prices evolved together, per book, so that later analysis of market behaviour has both halves of the picture.

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