49ers vs Chargers — Week 4, 2018
The 49ers took over trailing 29-27 with 3:49 left, 1st & 10 at their own 7, 1 timeout in hand, against the Chargers.
What really happened
Final: 49ers 27, Chargers 29. Chargers won 29-27 — the drive turnover.
49ers ball, down 2 — a field goal wins it. Backed up at their own 7, 3:49 on the clock.
nflverse's historical win-probability estimate for this exact situation was 30.7% for 49ers. Our model starts at 35.9%. The two numbers are estimates, not verdicts — the gap between them is a reminder that endgame win probability is hard to pin down.
The real drive, snap by snap
- (3:49) (Shotgun) 3-C.Beathard pass short left to 22-M.Breida to SF 27 for 20 yards (93-D.Philon).
- (3:20) 3-C.Beathard pass short left to 46-A.Morris to SF 31 for 4 yards (52-D.Perryman; 20-D.King) [46-C.Landrum].
- (2:38) 3-C.Beathard pass short left intended for 46-A.Morris INTERCEPTED by 98-I.Rochell [33-D.James] at SF 23. 98-I.Rochell to SF 19 for 4 yards (58-W.Richburg).
The opening call, by win probability
At the first snap, the strongest opening call in our model was QB sneak (-1.19 pts), and the weakest was Flea flicker (-4.40 pts). A positive number means the call improved the offense's expected win probability; a negative number means it cost them.
| Opening call | Expected WP change |
|---|---|
| QB sneak | -1.19 pts |
| Quick slant | -1.51 pts |
| QB keep | -1.62 pts |
| Quick out | -1.64 pts |
| Quick hook | -1.69 pts |
| Quick RPO | -1.72 pts |
| Pull RPO | -1.73 pts |
| Screen pass | -1.84 pts |
| Curl route | -1.91 pts |
| Outside run | -2.03 pts |
| Inside run | -2.12 pts |
| Play action | -2.41 pts |
| Off-tackle run | -2.51 pts |
| Deep post | -2.51 pts |
| Deep comebacker | -2.73 pts |
| Spike it | -2.73 pts |
| Deep corner | -2.98 pts |
| End-around | -3.74 pts |
| Receiver pass | -4.37 pts |
| Flea flicker | -4.40 pts |
This is our own model estimate, not the real coach's grade and not a claim about what the “correct” call was. Football decisions depend on the opponent, personnel and game plan — the model only measures how a call changes win probability in this situation.