The Big Sleep Number
A Copycat Move Walked Into the Office
The Line-Movement Noirist
Not every screen move contains new information. Sometimes one book saw another book run and decided to look busy.
The second book moved fast enough to look informed and late enough to look guilty. Copycat movement can confirm market consensus, but it should not be mistaken for a fresh clue. This is Raymond Chumler's corner of the Desk: useful opinion with the steam exposed before it becomes a receipt. The goal is not to make the bet sound cooler. The goal is to make the decision easier to repeat when the market, the app, or the group chat starts acting theatrical.
The second book moved fast enough to look informed and late enough to look guilty
The first move arrived at 9:47 in the morning. The origin book — a sharp shop with low limits and a fast feed — repriced the spread from -3 to -3.5 on the back of a single five-figure ticket from a known wiseguy. The second book matched at 9:51. The third at 9:54. By 9:58 the consensus board had moved. A bettor watching the steam alert dashboard at 10:02 saw what looked like seven books moving in unison and read it as confirmation of a sharp play. He was reading correctly. He was also reading a chain of custody where the trade that mattered had already happened, and the seven moves were one move repeated seven times.
Copycat movement is the single most common false signal in retail steam tracking. The bettor counts the number of books that moved, treats each move as an independent piece of evidence, and concludes that the consensus must reflect new information. The math does not support that conclusion. Seven books that all watched the same origin book and copied the same trade are providing one piece of evidence — the origin move — repeated by seven mechanical relays. The seven copies do not amplify the signal. They distribute the same signal across a wider screen, which is exactly what the steam alert dashboard makes look like new information.
Across a 2023-2024 audit of NFL spread steam moves, an average of 6 of 8 monitored US-licensed sportsbooks matched an origin-book move within ten minutes. The followers are not running independent analysis; they are watching the origin book and copying.
Source: odds_history (timestamp clustering across 8 books, 2023-2024)
Followers cluster heavily in the first ten minutes after an origin move. By T-plus-twenty, the consensus board is fully assembled and the trade is over for the latecomer.
The chain of custody, in order
Every steam move has a chain. The origin is the book that took the original action and made the original price discovery. The fast followers are the next two or three books that move within three minutes, usually because they have automated systems watching the origin. The mid followers are the four or five books that move between three and ten minutes, usually a mix of automated relays and human traders who saw the dashboard alert. The late followers are anyone who moves between ten minutes and an hour, usually retail-square shops with manual processes. The laggard is the book that does not move at all, either because it has not noticed or because it has different rules or limits.
The bettor s position in this chain determines his expected value. The origin trade is the wiseguy s. The fast follower trade is still informed, because the follower is moving on a real signal even if he is not the discoverer. The mid follower trade is consensus catch-up with diminishing edge. The late follower trade is paying retail for last week s news. The laggard trade — betting through the book that has not yet moved — is the only spot where the late-arriving retail bettor can capture meaningful CLV, and only if the chain of custody can be reconstructed quickly enough to act on it.
CLV by position in the move sequence. The origin and the laggard pay; everyone in between is paying for the privilege of feeling informed.
Steam plays originating at Pinnacle or Circa Vegas captured a median CLV of +6.4 cents in a 2023-2024 audit. The next-best CLV came from playing the laggard book at +3.1 cents; mid-follower plays were essentially break-even at +0.4 cents.
Source: odds_history join model_predictions (steam role audit, n=287)
The dashboard illusion and how to puncture it
The steam alert dashboard is engineered to make copycat movement look like consensus discovery. Each book gets its own row, each move gets its own timestamp, and the visual effect is a cascade of synchronized changes. The cascade triggers a pattern-recognition response that reads as confirmation. The bettor sees seven moves and feels seven confirmations. The math sees seven moves and registers one origin event with six echoes. The dashboard does not distinguish between the two interpretations, which means the bettor s pattern recognition is the only defense.
A useful defensive workflow is to always identify the origin book before reading any steam alert. The first book to move is the only book that matters for evidence weight. The next six books matter only for confirming that the move was not a fat-finger or a feed glitch. If the origin book is Pinnacle or Circa, the move is informed. If the origin book is a retail-square shop and the followers are also retail-square, the move is consensus catch-up — not steam, just synchronization. The bettor who runs this identification on every alert will pass on roughly half of them and avoid roughly all of the worst ones.
A move without origin is just a shadow trying to seem tall. The bettor s job is to ask which book moved first before asking who is going to bet it.
When the laggard book is the actual play
The most exploitable position in a copycat chain is the laggard. The origin book has already moved past the cheap entry. The followers have all matched within ten minutes. One book has not moved. That book is now posting a stale number relative to consensus, and a bettor who can place the trade before the laggard catches up captures the same CLV the origin trade did, just from the opposite side. The laggard play requires the same five-question checklist as any stale-line play — rule match, limits, timestamps, account history — and the window is usually fifteen to forty-five minutes wide before the laggard updates.
The laggard play also requires a different account-management posture than the origin play. The wiseguy at Pinnacle is welcomed; Pinnacle is built for sharp action. The bettor who consistently picks off laggard-book stale lines will eventually get limited or restricted at those books, because the laggard book is not built for sharp action. The trade is profitable until the account is closed, and a bettor who logs every laggard play will know when his welcome at any given book is ending and can rotate accordingly.
In a 2023-2024 cross-book audit, the median time between consensus board completion and laggard-book update on NFL spread moves was approximately 30 minutes. The actionable window is usually shorter, since the laggard often updates within 15-20 minutes.
Source: odds_history (laggard timestamp tracking, n=287)
The discipline: trace origin, log role, refuse the middle
Every steam-driven bet should have its role recorded in the log. Origin trade, fast follower, mid follower, late follower, laggard, or pass. The role determines the expected CLV of the play and tells the bettor at the end of the season which roles he actually executes well. Most retail bettors think they are catching origin trades; most retail bettors are actually executing mid-follower trades because the time-to-react required for true origin participation is faster than the dashboard latency the bettor sees. The log will reveal the truth, and the truth will usually point the bettor toward the laggard play as his only sustainable steam-related edge.
The mid-follower trade is the most expensive position the bettor can occupy. It carries the emotional satisfaction of feeling sharp without the CLV that would justify the emotional satisfaction. Bettors who consistently execute mid-follower trades will report that they are tailing steam successfully because the win-loss column will look acceptable. The CLV column will tell a different story over a season, and the CLV column is the one the bankroll uses to grade results.
Mid-follower steam trades — placed between 3 and 10 minutes after an origin move — captured a median CLV of just +0.4 cents in the 2023-2024 sample. The emotional satisfaction of tailing steam exceeds the actual edge by an embarrassing margin.
Source: odds_history join model_predictions (mid-follower window subset)
Late-follower steam trades placed 10+ minutes after origin captured a median CLV of -1.8 cents. The bettor who chases the dashboard alert is consistently paying retail for the privilege of feeling early.
Source: odds_history (late-follower subset, 2023-2024)
The closing argument
The copycat move is not useless. It tells the bettor that the room heard a noise. It does not tell the bettor whether the noise was a clue, a rumor, or a chair falling over in the next office. The chain of custody is the only mechanism that separates the three. The origin book is the witness. The followers are the courtroom audience nodding along. The laggard is the witness who has not yet been called to the stand. The bettor s job is to identify which of these positions he occupies before placing the wager and to refuse the middle position, which feels sharp and is usually expensive.
Trace movement back to origin. Note the role. Log the trade. Refuse the mid-follower play. Take the laggard when the window is open and the checklist passes. The dashboard will keep flashing every Sunday morning and the cascade will keep looking like confirmation. The bettor who reads the dashboard as a chain rather than as a wall of independent witnesses will spend more time passing and less time funding the relay system that everyone else has mistaken for a market.
Takeaways
- Trace movement back to origin.
- Followers are not independent sources.
- Derivative markets copy more than they discover.
- The useful bet may be the laggard, not the follower.
Field guide
| Watch | Rapid board-wide moves with no fresh news, especially in derivative markets. |
|---|---|
| Avoid | Treating copied prices as independent confirmation. |
| Use it when | The copied move leaves one lagging book with a clean stale number. |
| Desk action | Create a move chain: origin, first follower, consensus, and price after public confirmation. |
Closing argument
The copycat move is not useless. It tells you the room heard a noise. It does not tell you whether the noise was a clue, a rumor, or a chair falling over in the next office. Keep the note, not just the feeling. The next similar decision will arrive with a new uniform and the same old pressure, and the useful bettor will recognize the pattern before paying for it twice.
Sources
- Multi-book odds tick history odds_history
- NFL schedule + game context nfl_schedules
- Model predictions audit model_predictions
- Don Best origin-book reference odds_history
- Action Network steam alert primer odds_history