What Is Reverse Line Movement?

Reverse line movement (RLM) is when a betting line moves toward the side that fewer people are betting. If 75% of tickets are on the favourite at -3 and the line drops to -2.5 — making the favourite cheaper, not more expensive — that is reverse line movement. The usual reading is that a smaller number of bettors the book respects are on the underdog, and the book cares more about them than about the crowd.

That reading is sometimes right. It is also the most over-sold signal in sports betting, because it rests on unaudited ticket data, ignores the other reasons lines move, and arrives after the price has already changed. RLM can tell you the market disagreed with the popular side. It cannot tell you the bet is still worth making.

How RLM is supposed to work

The logic has three steps:

1. Most tickets are on Team A.
2. A book trying to balance its money would make A more expensive to slow those bets down.
3. Instead, A got cheaper. So something outweighed the crowd — probably large bets from sharp accounts on B.

Each step has a weak point.

Why step 2 is wrong more often than people assume

The whole signal assumes books move lines to balance the number of bets. The best evidence says they don't. Steven Levitt's study of an online book's NFL handicapping contest in the 2001–02 season, published in The Economic Journal in 2004, found that in the median game almost two-thirds of the bets were on one side, and in almost 10% of games more than 80% were. The book's spreads showed no sign of being set to even this out. Levitt's reading was that it held the position on purpose, pricing the popular side slightly too high.

Two consequences follow. First, a lopsided ticket count is the normal state of a betting market, not an anomaly. "65% of bets on A" describes an ordinary game. Second, if books don't move toward the crowd in the first place, a line that fails to follow the crowd is not surprising. RLM defined against a balancing model that books don't use will fire constantly.

Why step 1 is shakier than it looks

The ticket percentage has to come from somewhere, and you usually don't know where. Published splits rarely name the books, the sample size, or whether parlay legs and live bets are included. As of September 2026 we are not aware of any US regulator that audits them. They are also typically drawn from retail books — the places where the sharpest accounts are most likely to have had their limits cut. The crowd you are reading is the crowd a particular book chose to show you.

The previous guide, sharp money vs public money, goes through what splits can and cannot tell you.

Why step 3 has other explanations

A line can move against the ticket count for reasons that have nothing to do with sharp bettors at that book.

The book copied someone else. Tobias Moskowitz's 2021 Journal of Finance study describes books that "move on air": when a market-setting book moves, others follow without taking significant bets themselves. If a market-maker — Pinnacle, for instance, which says on its own site that it takes arbitrage bets because it is confident in its pricing — moves the number, a retail book with 75% of its tickets on A will move too. The split at that book played no part.

News. An injury update, a lineup, a weather forecast. The price moves toward the side the news favours, and the public may not have caught up. Calling that "sharp money" confuses information with who bet on it.

The book's own model. A book may simply revise its number. Opinion moves prices even with no bets at all.

Noise. Not every move is right. Moskowitz found that on average about half of a price's move between open and close was reversed by the game result. Some moves are crowds overshooting; some are books correcting in the wrong direction.

The part nobody puts in the headline: you get the worse price

Say a game opens with the favourite at -160 and the underdog at +140. Most tickets come in on the favourite, yet by the afternoon the line is -145 / +125. Textbook RLM.

What does the market now think? Convert and remove the margin:

• -145 → 145 ÷ 245 = 59.18%
• +125 → 100 ÷ 225 = 44.44%
• Total 103.63%; underdog's no-margin probability 44.44 ÷ 103.63 = 42.89%

Now compare two underdog bettors, taking the new price as the fair one:

• Bet at the open, +140 (pays 2.40 times the stake): 0.4289 × 2.40 − 1 = +2.9%
• Bet after seeing the RLM, +125 (pays 2.25): 0.4289 × 2.25 − 1 = −3.5%

The move itself is the value. The person who caused it captured it; the person who followed it paid the margin on a price that had already been corrected. If the move was right, you are betting into an accurate number with the vig still attached. If it was wrong, you are betting into an overshoot.

This is the same logic as closing line value, covered in opening vs closing lines: a bet is good when your price beats where the market settles. Joseph Buchdahl's analysis of 87,960 soccer odds pairs found that the amount by which an earlier price beat the Pinnacle close predicted realised returns almost one-for-one. Chasing a move after it happens is, by construction, the opposite of beating it.

What RLM can and can't indicate

It can indicate that some force outweighed the ticket count — a respected bet, a market-maker's move, or news. As a prompt to go and find out which, it is useful.

It can't indicate:

• that the side it points to will win;
• that the side it points to is still a good bet at the current price;
• that any sharp bettor was involved, rather than a copied line or a news update;
• anything reliable at all when the underlying split is unsourced.

Levitt also tested whether bettor consensus predicts results. He found weak, statistically insignificant evidence — not enough to trade on either with or against the crowd.

A more useful version of the same instinct

The good idea buried in RLM is "watch what the sharpest price does." You can do that directly, without the split. Compare your book to a market-maker's no-vig price, and if your book hasn't yet moved to where the market-maker already is, that gap is the thing worth acting on — before the move, not after it. The Academy's guides to devigging odds and closing line value cover the method.