Texans 17, Colts 19 — Week 3: Three Turnovers Forced, Seven Points

Second and 10 at the Houston 39, four 45 left in the fourth quarter, Colts up 16 to 10. Will Anderson sacked Daniel Jones, the ball came loose, and Togiai recovered at the Houston 46. By win probability — the model's read on each team's chance of winning at that moment — that snap moved nearly 13 points Houston's way, and by expected points added, what a snap was worth given the down, the distance and the field position, measured against what that situation was already worth, it cost the Colts five and a half. Seven plays and 54 yards later Woody Marks went in from the one on first and goal with one 50 left, and after the extra point Houston led 17 to 16. The lead lasted 87 seconds. Indianapolis went 29 yards, and Spencer Shrader made a 53-yard field goal with 21 seconds left. The Texans lost 19 to 17 in Indianapolis, are oh and three, and led this game for less than a minute and a half of it.

That strip-sack was the third takeaway of the day for a pass defense that looked like last season's again. It faced 40 Colts pass plays — that count includes sacks, and there were four — and held them to −6.9 expected points; on defense a negative number is the good direction. That total is fifth-best of the 28 teams that have played so far this week. At the rate this defense allowed over the first two weeks, those same 40 plays would have been worth about +9 to the Colts. At last season's rate, when Houston had the best pass defense in the league by expected points per play, about −7.4. Sunday landed close to last season. The interception came on third and four at the Houston 28, 11 40 left in the third quarter, Colts up 13 to seven: Jones deep left for Josh Downs, and Blankenship took it at the Houston one. −2.8 expected points for the Colts, six and a half points of win probability.

Three takeaways, and seven points from them. The turnover margin was +3, tied with Washington for third-best of the 28 teams that have played this week, and it was not enough. The first came on the Colts' fourth snap: first and 10 at the Indianapolis 47, 11 28 left in the first quarter, nothing to nothing, Anderson with the sack, the ball out, Houston's ball at the Indianapolis 40. 16 points of win probability and −6.3 expected points for the Colts, the biggest swing of the game until the final minute. Three plays gained nine yards, and on fourth and one at the 31 David Montgomery went off right tackle for no gain: −4.2 expected points, and about eight points of win probability handed straight back. The interception left the offense at its own one, and three plays later Houston punted from its own five after a delay-of-game penalty. Only the last takeaway became points.

The offense is where the week goes wrong. 223 total yards, the fewest of the 28 teams that have played so far this week. 30 pass plays, three of them sacks, produced −3.1 expected points and 153 yards after the sacks, and 43 percent were successful plays, meaning they gained what the situation asked for. By expected points per pass play that is seventeenth of the 28 teams so far this week. At the rate Houston passed over the first two weeks, those 30 would have been worth about −2.3; at last season's rate, when the passing game ranked fourteenth, about +1, and that against a Colts pass defense that ranked nineteenth last season. No interceptions is the clean line. But Houston did not complete a pass of 20 yards until the fourth quarter, and on third and seven or longer it converted once in seven tries — C.J. Stroud's eight-yard scramble on third and eight.

The run game opens on its shape: 21 rush plays, 70 yards, and 13 of the 21 gained three yards or fewer. No Houston run gained more than nine. −1.6 expected points, fifteenth of the 28 teams so far this week per play, and 38 percent successful. Against the first two weeks that is an improvement: at that rate, which includes a Week 2 run game that ranked last in the league, those 21 plays would have been worth about −2.6. At last season's rate, when the run game ranked thirtieth, they would have been worth −1.6, almost exactly what they were.

10 possessions: two touchdowns, a field goal, five punts, the stop on fourth down, and the final 16 seconds. Four three-and-outs, the same number as in each of the first two weeks. Inside the 20 it moved: two touchdowns on seven red-zone snaps, after none on 10 in Week 2. The first came in the second quarter, down 10 to nothing, at the end of a six-and-a-half-minute drive that a face-mask penalty and a pass interference call on the Colts carried to first and goal at the one; Stroud threw short left to Foster Moreau. The field goal came from Houston's longest drive, 71 yards from its own 23, with a 28-yard throw to Xavier Hutchinson and 20 yards to Wayne on third and five. It reached first and 10 at the Colts 12, three plays gained six, and Ka'imi Fairbairn kicked from 24 with seven 36 left: 16 to 10.

The run defense held about to last season's level. 31 Colts rush plays, 89 yards, −1.2 expected points — at last season's rate, when this run defense ranked in the top 10, those 31 would have been worth −1 and a half; at the rate of the first two weeks, about −5 and a half. Sixteenth of the 28 teams so far this week. The Colts converted both of their fourth-and-ones, a Jones scramble at the Houston 19 in the first quarter and a four-yard Jonathan Taylor run at the Houston 48 in the third that set up a field goal.

Then the last 90 seconds. Down 17 to 16 with one 48 left, the Colts started at their own 35. On second and 10 Jones threw deep middle for 20 yards to the Houston 45, 10 points of win probability. Houston spent all three of its timeouts on the drive. On third and six at the Houston 41, Jones found Keenan Allen at the 35; it was ruled a first down and reversed on review to five yards, fourth and one at the 36. Shrader's kick was worth 36 points of win probability, the biggest swing of the game. Houston got 16 seconds, a 17-yard throw to Dalton Schultz to its own 48, and two incompletions.

For anyone new to this show, here is where that sits. Last season Houston went 12 five with the best pass defense in the league by expected points per play, a run defense in the top 10 and a third-down defense sixth, and an offense that ranked fourteenth passing, thirtieth running and 20-fourth on third down. Week 1 was a 36 to 31 loss at home to Buffalo in which the offense scored 31 and the pass defense ranked thirtieth of 32. Week 2 was a 20 to six loss at home to Cincinnati: the run game ranked last in the league, 10 snaps inside the 20 produced no touchdowns, and four drives ended on downs. Sunday put the pass defense back near last season's rate, and the takeaways arrived, three after none in either of the first two games. The offense did not follow: the fewest yards of any team so far this week, and a run game at last season's rate, which was thirtieth. Three games do not settle a season, and two of these three losses came by five points or fewer. The defense this team was built on showed up on Sunday; the offense turned its three takeaways into seven points.

Houston is home against the Cowboys next Sunday, October fourth.

That's your Texans — every game, by the numbers. This is Muffed: we do the same deep dive for all 32 teams. Follow the Texans show wherever you're listening, so next week's recap lands the moment it's out. No theater. Just the football.

And if you play fantasy, there's a Muffed show built around your exact roster — free at muffed.ai.

_Data via nflverse._