8 min readOctober 29, 2026

How to Chase Steam Without Chasing the Side

When a sharp book moves and a slow book has not caught up yet, the price left behind is the bet.

Steam Chaser watches reference books like Pinnacle for line moves, then shows sportsbook prices that are still sitting on the old number and still look +EV against Upside's no-vig fair price, so you can check them before they disappear.

Follow Pinnacle, bet only the old price while it is still +EV, and let the play go once your book has moved.

What steam is, and why the old price is the bet

Steam is what happens when sharp money moves the sharpest books. Slower sportsbooks often keep the old price for a while, and that old price, not the side itself, is where the +EV bet lives. +EV (positive expected value) means the price pays better than the real chance of the bet hitting.

The video opens with a real move on Franz Wagner's points plus rebounds line at 30.5. FanDuel, Fanatics and ProphetX pulled the line within 30 seconds of each other and Novig's under jumped from -115 to -174, yet BetMGM was still showing the under at 30.5 for -105 about two hours later. Anyone taking the over at BetMGM that afternoon was buying the side the rest of the market had already run from.

Three prices on the same under

After the Wagner move, the under was available at Novig for -174, at DraftKings at 29.5 for -116, and at BetMGM at 30.5 for -105. Novig's new price worked out to about -15% EV against the fair price and DraftKings had the worst number for the under, while BetMGM's old price was the only +EV option, at about +4.8%.

What the test of Pinnacle moves found

The video tested every time a sharp book moved a player prop or moneyline by at least 1% in implied chance within an hour, where implied chance is the win probability built into the odds. Each time one of six big sportsbooks (DraftKings, FanDuel, BetMGM, Caesars, Fanatics and Hard Rock) still showed the old price and that price was +EV, it counted as a play.

Every play was then checked against the closing line: the no-vig price right before the game starts, meaning the market price with the sportsbook's built-in margin removed. Beating the close is the best early sign of a real edge, because it shows up long before wins and losses do. After Pinnacle moves, 76.9% of 1,547 old-price plays beat the close, compared with 64.9% for the same kind of +EV bet with no Pinnacle move, so the move itself was worth about 12 points.

How often each kind of play beat the closing line

Old price after a Pinnacle move
76.9%

1,547 plays at six big sportsbooks

Same kind of +EV bet, no Pinnacle move
64.9%

What a normal +EV bet already reaches

Right after Pinnacle moved the other way
57.2%

Taking a side Pinnacle had just moved away from

Old price after an exchange move
59.2%

Versus 62.6% for the same bets with no exchange move

Follow Pinnacle, not the first book to move

On big player prop moves, the exchanges usually moved first in Upside's feed, with Novig first 52% of the time and Pinnacle only 28%. Speed was not the same as signal: when an exchange moved, the old sportsbook prices beat the close 59.2% of the time, versus 62.6% for the same bets with no exchange move. Exchange prices flicker as single orders come and go, while Pinnacle changed price only about every 18 minutes on average, so a Pinnacle move means more.

Testing every pair of books, only two were worth following on player props: Pinnacle (worth about 11 points into the bigger sportsbooks) and FanDuel (about 5). That gives the video's rule: when Pinnacle moves a prop at least 1% in implied chance, a book you use is still on the old price, and that price is +EV, the play qualifies. If you bet at DraftKings or BetMGM, FanDuel's moves count too.

Which steam plays to skip

You might expect steam to matter most right before the game, but the test found the opposite. In the final hour, old prices beat the close 85.1% of the time, but so did any +EV price with no move at all (83.8%), because there is barely any time left for the line to move. Four to 12 hours out, the Pinnacle move was worth 14.8 points.

Game lines behave differently: into the big sportsbooks, nothing held up on moneylines, spreads or totals, and the slow prices were at Novig instead. Those go fast (about half of Novig's old prices were gone within 5 minutes of a Pinnacle or Polymarket move), and the feed shows the exchange price but not how much money is behind it, so check the size available before you act.

A line move is a reason to check, not a reason to bet

Seeing a steam alert and betting the same side wherever you play is the mistake the video warns about. Once every book has moved, you are paying full price for someone else's information, and by the time an alert reaches your phone the old price is often gone: almost half were gone within 15 minutes of the move.

How to set up Steam Chaser

Start by picking only the books you can actually bet at, since Steam Chaser only shows plays at the books you select. In the filters, reference books are the books whose moves count: the video uses Pinnacle and FanDuel, and on a FanDuel move it only takes sportsbook plays, because FanDuel's moves paid at DraftKings and BetMGM but not at the exchanges. Adding more reference books gives you more plays, but in the test those plays did no better than a normal +EV bet.

Leave the window (how far back it looks for a move) at 1 hour and the minimum move at the 1% implied-chance default, which is what the test used. Keep the minimum EV at zero with Pinnacle as a reference: even plays showing under 1% EV beat the close almost 71% of the time, versus 51.3% for the same tiny EVs with no move, because the EV number has not caught up to Pinnacle yet. For game lines, the video switches the market to game lines, keeps the same two reference books, and only takes Novig plays on the main number, since that is where the data showed an edge.

How to read a Steam Chaser play

The big green number on each card is EV, how much better the price is than fair odds, and right under it is the steam itself. In the video's example, Pinnacle had moved Pascal Siakam's under 4.5 assists from -145 to -153 (59.2% to 60.5% in implied chance) about 15 minutes earlier, while BetMGM still offered -125 against a fair price of -130. The card also suggests a stake, about $5 on a $1,000 bankroll at quarter Kelly, meaning a quarter of the bet size the Kelly formula would suggest.

Click the card for the full breakdown, including a chart where one line is Pinnacle, one is your book and the dashed line is the fair price; when Pinnacle's line moves and your book's does not, that gap is the +EV bet. The Where to bet list shows the fair odds on top and then every book: green means the price is +EV, Moved means that book has already moved, and Alt means it is on a different line, so it is not the same bet.

A steam-chasing workflow

1

Sort by most recent

The freshest moves are the ones slow books have not caught up to yet.

2

Check the reference move

Confirm the move came from Pinnacle (or FanDuel, if you bet at DraftKings or BetMGM) and that your book still shows the old price.

3

Move fast, then verify

In the test, half of Fanatics' old prices were gone in under 3 minutes, and Hard Rock was the slowest at about 40 minutes. Confirm the live price at your book yourself before deciding.

4

Bet only the old price

If your book has already moved, let the play go. The edge was the price, not the side.

Article FAQs

What is steam chasing?

Steam chasing means betting the old price a slower sportsbook is still offering after a sharp book like Pinnacle has moved. The bet is the stale price, not the side, so it only works while that price is still better than fair value.

Are line moves usually right?

In the video's look at 27,579 games, when a line moved three or more points of win chance, the side it moved toward was priced at 51.2% at the open and won 58% of the time. The edge, though, only exists while a book is still offering the old price.

Why not follow the exchanges, since they move first?

In the video's test, exchange moves on their own added nothing: old sportsbook prices after an exchange move beat the close less often than the same bets with no move. Exchange prices flicker as orders come and go, though an exchange move did help when Pinnacle had moved the same way.

Will every steam play win?

No. Every bet can lose, which is why the video measures plays against the closing line instead of wins and losses, and an edge only shows up over many bets. Upside shows the play; you decide whether to bet it at your book.

Open Steam Chaser

Pick your books and reference books, then see which sportsbooks are still sitting on an old, +EV price after a sharp move.