SharpEdge

Line Moves

Every number here is DraftKings, the house book, from when the card opened on September 28 to this morning. The spread is the head start the underdog gets: Seahawks -3 means Seattle has to win by more than 3 for that bet to cash. The total is the over/under on combined points from both teams.

Seven games touched a key number this week. Key numbers: NFL games land on 3 and 7 far more than any other margin, so a half point on either side of those numbers is worth real money. The Rams went from -1.5 to -3.5 in Philadelphia, straight through 3, and that total fell from 46.5 to 42.5, the biggest total drop on the board. Indianapolis left 3 behind entirely, -3 to -5.5 (more below). Chicago went through 3 the other way, -3 to -3.5, even with Caleb Williams Out (hamstring) and Case Keenum projected to start. Buffalo came off 7 to -6.5, so New England +7 is gone. Seattle climbed onto 7, Houston climbed onto 3, and Green Bay slid from 3.5 onto 3 with Jalon Daniels projected to start for Tampa in place of Baker Mayfield (thumb, Out).

The move against us comes first. Closing line value is CLV: whether the line you bet was better than where the market closed. The best long-term sign a bettor knows what they are doing, because closing lines are the sharpest numbers there are. We took the Colts-Commanders over at 47.5 and it sits at 46.5. That's a full point of CLV going the wrong way, and it counts. Thursday's Browns +2.5 has no closing DraftKings number in our feed. The nflverse schedule line shows Cleveland +2.5, our number, so call it even.

Now the move with us. Houston opened -2.5 and is -3 now, which is the most valuable half point in football. If Houston wins by exactly 3, our ticket cashes and anyone laying -3 today pushes. Carolina +3.5 hasn't budged and is still on the right side of 3.

The Card

Houston -2.5 (Play), Sunday 1:00 PM ET. The case is Dallas's defense, which has allowed 0.191 EPA per play, dead last in the league. EPA per play is Expected Points Added: how many points a play moved the team toward scoring compared with an average play from the same spot. Positive is good on offense, negative is good on defense. Every profile number in this issue blends this season's first three weeks with last season, weighted 70% to last year. The model makes it Houston -4.7. That's only 1.7 points off today's -3, below the 2-point gap the backtest required, so I wouldn't bet it fresh. The number was the bet, and we already have it. I'm wrong if Houston's offense, 24th in EPA per play, can't take advantage of a bad defense.

Carolina +3.5 (Play), Sunday 8:20 PM ET. Detroit's defense allows 0.149 EPA per play, which ranks 29th, and Detroit is 0-2-1 against the spread this year. The model has Detroit by 1.3. That's a 2.2-point lean to Carolina, inside the backtest's 2-to-3-point bucket, the only one that cleared break-even (54.7%). It's the same number as when we bet it, so if you missed it, you missed nothing. I'm wrong if Detroit's offense, 7th in EPA per play, outscores a Carolina receiver room without Xavier Legette (knee, Out) and with Jalen Coker Questionable (quadricep).

Colts-Commanders over 47.5 (Lean), Sunday 9:30 AM ET. Indianapolis ranks 31st in EPA per play allowed and Washington ranks 26th. The model total is 50.0. We gave back a point, and Washington changed quarterbacks. Hold it and add nothing. New money gets 46.5, a better number than ours, but the backtest covers spreads only, so totals stay Leans.

Browns +2.5 (Lean), Thursday Oct 1, final. The pack hasn't graded it yet, so this issue doesn't either.

New plays: none. The closest call is Philadelphia +3.5. It's a 2.1-point model lean, in that 54.7% bucket, and it's on the good side of 3. But the model runs on team EPA and scoring margin, and Philly is missing DeVonta Smith, Marquise Brown and Dallas Goedert. When the number and the injury report disagree this loudly, I pass. Our feed also doesn't carry the DraftKings price on spreads, and I won't put a ticket in the ledger at odds I can't show you.

Sizing. Units: A fixed bet size so results can be compared. One unit is one unit whether that is $10 or $100. Plays get 1 unit. Leans get half a unit or nothing. Nothing on this card goes bigger, and at 2-8 nobody should be pressing.

Line Shop

DraftKings is the only book in this week's feed, so every number here is theirs and there's nothing to compare. Line shopping: Checking the same bet at several sportsbooks and taking the best number. Getting +3.5 instead of +3 is free equity, and it costs nothing but a minute. Do it yourself on three tickets:

• Carolina: anything better than +3.5.
• Colts-Commanders over: any total under 46.5.
• Houston: -2.5 if a book still offers it, because that half point through 3 is the whole bet.

The Lean

Colts at Commanders, Sunday 9:30 AM ET, Tottenham Hotspur Stadium. This is the biggest spread move of the week, IND -3 to -5.5. Washington's injury report:

• QB Jayden Daniels: Out (elbow). Marcus Mariota is projected to start.
• RB Rachaad White: Out (shoulder)
• G Sam Cosmi: Out (concussion)
• S Nick Cross: Out (illness)
• WR Terry McLaurin: Questionable (hamstring)

The model still says Indy by just 1.4, a 4.1-point gap toward Washington. Ignore it, for three reasons:

• The recipe gives Washington the full 1.8 points of home field for a game in London.
• It has no input for who plays quarterback.
• The backtest says gaps of 3 to 4.5 points hit 52.1%, short of break-even.

Implied team total: How many points the betting market expects a team to score, worked out from the spread and the total. Bigger number, more touchdowns to go around. Indy sits at 26.0 and Washington at 20.5. Both pass defenses are soft: Indy allows 0.192 EPA per pass and Washington allows 0.2. That's the over case. No side; hold the over. I'm wrong if Indy leans on its run game (0.061 EPA per run) and Washington's run defense (-0.137 allowed) turns it into a grind.

Rams at Eagles, Sunday 1:00 PM ET. The spread moved two points through 3 and the total dropped four. Philly's report is brutal. All of these players are Out:

• WR DeVonta Smith (hamstring)
• WR Marquise Brown (ankle)
• TE Dallas Goedert (knee)
• T Fred Johnson (knee)
• LB Zack Baun (concussion)
• S Marcus Epps (groin)

The Rams are without Aaron Donald (back), CB Jaylen Watson (shoulder) and TE Terrance Ferguson (ankle), and TE Colby Parkinson is Questionable.

Philly's passing game was already at -0.105 EPA per pass, its defense has zero takeaways a game, and it's 0-3 against the spread. The Rams' pass defense allows -0.132 EPA per pass. Implied totals are Rams 23.0, Philly 19.5. The market moved the total right onto the model (42.1) and moved the side away from it (the model has the Rams by 1.4). Pass. The number sits on Philly's side of 3, and the injury report sits on the Rams' side. If Philly covers easily, the market overreacted to the missing pass catchers and I'll own the miss.

The Ledger

Hit rate: The percentage of bets that won. At standard -110 odds you need 52.4% just to break even.

That's two wins in ten graded plays, down 6.18 units, and there's no spinning it. The one number pointing the right way is average CLV of +0.4, which means we've gotten a better number than the close on average. Ten plays is a tiny sample, though, and CLV is a long-run signal, not a promise for tonight. For perspective, the model's walk-forward backtest over 2023-2025 went 51.7% across 437 plays against closing spreads, short of break-even. The model tells us where to look; the edge is the number we take. Nothing has been graded this week yet, including Thursday's Browns play.

Ledger Entries

Plain English

• Spread: The head start the underdog gets. Seahawks -3 means Seattle has to win by more than 3 for that bet to cash.
• Total: The over/under on combined points from both teams.
• Key numbers: NFL games land on 3 and 7 far more than any other margin, so a half point on either side of those numbers is worth real money.
• Closing line value: CLV: whether the line you bet was better than where the market closed. The best long-term sign a bettor knows what they are doing, because closing lines are the sharpest numbers there are.
• EPA per play: Expected Points Added: how many points a play moved the team toward scoring compared with an average play from the same spot. Positive is good on offense, negative is good on defense.
• Units: A fixed bet size so results can be compared. One unit is one unit whether that is $10 or $100.
• Line shopping: Checking the same bet at several sportsbooks and taking the best number. Getting +3.5 instead of +3 is free equity, and it costs nothing but a minute.
• Implied team total: How many points the betting market expects a team to score, worked out from the spread and the total. Bigger number, more touchdowns to go around.
• Hit rate: The percentage of bets that won. At standard -110 odds you need 52.4% just to break even.

Sources

DraftKings lines via ESPN (house book, openers from September 28); nflverse play-by-play and schedule data; official NFL injury reports; the SharpEdge model, backtest and ledger.

21+. Gambling problem? Call 1-800-GAMBLER.