Chiefs 30, Raiders 27 — Week 4: Thirteen Points in Thirty-Three Seconds

19 to 17, Raiders, three 35 left in the game, third and five at the Kansas City 40. The Chiefs' chance of winning at that snap was 33 percent. Patrick Mahomes threw deep left for Tyquan Thornton, 31 yards in the air, and Thornton carried it 24 more to the Las Vegas five. A Raiders unnecessary-roughness penalty moved it to the two. +5.2 expected points added — expected points added being what a snap was worth given the down, the distance and the field position, measured against what that situation was already worth — and 35 points of win probability, each team's chance of winning at that moment. It was the biggest win-probability swing of the game. Kenneth Walker scored from the two, the two-point try failed, and on the kickoff Bassa forced a fumble on the return and Cochrane recovered it at the Las Vegas 22. Walker ran 22 yards for a touchdown on the next snap. 30 to 19, and the Chiefs' chance had gone from 33 percent to 97 in 33 seconds of game clock. Las Vegas scored again and converted the two-point try, Xavier Worthy fielded the onside kick, and the Raiders' last drive ended with Nwankpa intercepting Kirk Cousins at the Kansas City 36 with 10 seconds left. It was the Chiefs' fourth straight win to open the season.

The run game leads, because it was the offense. 25 rush plays, a hundred 90 yards, two touchdowns, +5.8 expected points, and 48 percent of them successful, meaning they gained what the situation asked for. At the rate Kansas City ran over the first three weeks, those 25 would have been worth about +3. Walker carried 22 times for a hundred 77. One carry made the total. 30 seconds left in the third quarter, down 13 to 10, first and 10 at the Kansas City 16: Walker through the right guard, and he was not pushed out of bounds until the Las Vegas eight, 76 yards later. +4.9 expected points and 22 points of win probability. Two snaps later Mahomes found Thornton for the go-ahead touchdown. The other 24 rush plays came to +0.9. The early failures were short-yardage: on the first possession, third and two and fourth and two near midfield, no gain and one yard, and the ball went over on downs.

The charting answers half of last week's notebook question: whether the run game's yards come back at the same box count. The yards came back; the box did not stay the same. Las Vegas showed six or fewer defenders in the box on 14 of 23 charted designed runs, 61 percent, seventh-highest of the 30 charted offenses this week and well above the 44 percent of the first three weeks. 24 designed runs gained a hundred 89 yards, 7.9 a run, after 5.1 across the first three weeks.

The passing game opens on its success rate, because the chunks hide it: 13 of 32 pass plays were successful, 41 percent. The 32 were worth +2.7 expected points, sacks included; at the three-week rate they would have been about +7 and a half. Mahomes completed 15 of 30 for 225 yards, two touchdowns and no interceptions, and was sacked twice. Three completions — 41 yards to Worthy on third and 10 in the first quarter, then 42 55 to Thornton — made up a hundred 38 of the yards. After halftime he was five of 12. Thornton had eight targets, five catches, a hundred 11 yards and both touchdowns.

The charting says the ball was harder to catch than it has been. 19 of Mahomes's 30 throws were charted catchable, 63 percent, 20-second of the 30 charted offenses this week, after 75 percent over the first three weeks. Seven were contested, a receiver and a defender arriving together, 23 percent after eight. One was interception-worthy, a ball a defender should have caught, and three were thrown away. The rush was heavier: Las Vegas sent five or more pass rushers on 39 percent of his dropbacks, after 25 percent from the first three opponents, and he was outside the pocket on 12 of 33. Play-action, the other half of the notebook, came on 11 of 33 dropbacks, 33 percent, fifth of the 30 this week but under the 38 percent of the first three weeks. It did not carry the yards this time: five of 11 for 71 off the fake, 10 of 19 for a hundred 54 without.

The pass defense opens on a total that is not good: 49 Las Vegas pass plays worth +8.7 expected points, 365 yards, and 53 percent successful. On defense the good direction is negative, and at the rate this pass defense allowed over the first three weeks, 49 plays would have been worth about −6. No sack stood; the one Karlaftis got late was wiped out by a holding penalty. Kansas City sent five or more rushers on 22 percent of Cousins's dropbacks, 20-fourth of the 30 charted defenses this week and under its own 32 percent of the first three weeks. 36 of his 49 throws were catchable, 74 percent, above the 68 this defense had allowed. The contested rate held: 27 percent, the same as the first three weeks. What held up was third down. Las Vegas converted five of 14, and three of its four field goals came right after a third-down stop.

The run defense was the steadier unit. 26 Las Vegas rush plays went for 72 yards and −6.2 expected points — on defense a big negative is the good result — against a three-week rate worth about −1 and a quarter. 38 percent were successful. The box was lighter here too: six or fewer on 15 of 26 designed runs, 58 percent, after 44. The Raiders gained 2.8 a designed run, under the 3.6 this defense allowed in Weeks 1 to 3. The stop that mattered came with five 46 left, the Chiefs down 17 to 16: third and two at the Kansas City 30, Ashton Jeanty through the right guard, Tranquill and Gillotte, no gain, and Las Vegas settled for a field goal.

The other turning point came early in the third quarter, Raiders up 13 to 10. Matt Gay made a 41-yard field goal, a Kansas City illegal-formation penalty wiped it out, and Las Vegas went for it on fourth and two at the 18. Cousins threw short left for Jeanty, incomplete. +3.6 expected points to the Chiefs and nearly 10 points of win probability.

What repeats. Last week's notebook asked for the play-action rate and whether the run game's yards came back at the same box count. Play-action fell to 33 percent, still fifth in the league this week. The run yards came back, and the box got lighter, so the question changes. What goes in the notebook: Mahomes's catchable rate, 63 percent after 75, and whether defenses keep emptying the box against this run game.

The stat lines, scored half-PPR. Walker, 30.4: 22 carries for a hundred 77 two touchdowns, one catch for two yards. Thornton, 25.6 on five catches for a hundred 11 two scores. Mahomes, 17 even. Worthy, nine even, four of seven for 70. Travis Kelce caught two of four targets for 15, 2.5. For Las Vegas, Cousins scored 22.6 on 365 yards, and Brock Bowers 17.6.

For anyone joining: wins over Denver and the Colts at home, at Miami, and now at Las Vegas, the last of them from behind with three and a half minutes left. Four and oh.

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

_Data via nflverse. Charting: FTN Data via nflverse._