Loading…
Our engine ran 10,000 Monte Carlo plate appearances of Hunter Goodman against Blake Snell at Dodger Stadium. It gives a 17% probability of a hit per plate appearance, with 37.6% strikeouts, 8.1% walks and 2.7% home runs per PA. Average simulated length: 3.23 pitches per plate appearance. No raw head-to-head rate drives any of these numbers; the inputs are regressed multi-season rates for both players (methodology below).
| Outcome | Sim probability | Rate |
|---|---|---|
| Walk | 8.1% | |
| Strikeout | 37.6% | |
| Hit by pitch | 0.6% | |
| Single | 10.5% | |
| Double | 3.5% | |
| Triple | 0.3% | |
| Home run | 2.7% | |
| Batted-ball out | 36.7% |
Where the simulated plate appearances ended, and the hit probability conditioned on that count.
| Count | PA ends here | P(hit) at count | K / BB |
|---|---|---|---|
| 0-3 | 16.1% | 0.0% | |
| 1-3 | 14.3% | 0.0% | |
| 0-0 | 12.8% | 32.2% | |
| 0-1 | 11.1% | 30.5% | |
| 1-1 | 8.4% | 31.8% | |
| 2-3 | 7.2% | 0.0% | |
| 1-0 | 5.1% | 30.9% | |
| 1-2 | 4.6% | 27.3% | |
| 0-2 | 4.4% | 30.6% |
Career line: 8 PA, 1 hits (0 HR), 1 BB, 6 SO. .143 AVG / .393 OPS. Statcast-era pitch log: 8 PA across 3 seasons (2023–2025).
| Season | PA | H | HR | BB | SO | OPS |
|---|---|---|---|---|---|---|
| 2023 | 3 | 0 | 0 | 0 | 3 | .000 |
| 2024 | 2 | 0 | 0 | 0 | 2 | .000 |
| 2025 | 3 | 1 | 0 | 1 | 1 | 1.167 |
Read rate: 8 career PA is 0.0% of one full batter season. One extra dropped popup changes the career average by 14.3 points.
Re-running the same 10,000 simulated PAs with Dodger Stadium's factors (hit 0.97x, home run 0.93x) moves the hit probability from 17.7% in a neutral park to 17.0%, a -0.7-point move.
Blake Snell's 2026 pitch mix (22,713 pitches tracked).
| Pitch | Usage | Avg velo | Whiff / swing |
|---|---|---|---|
| Four-Seam Fastball | 50.7% | 95 mph | |
| Curveball | 18.1% | 81 mph | |
| Changeup | 16.9% | 87 mph | |
| Slider | 14.3% | 88 mph |
| Pitch | PA | H | K | Whiff / swing |
|---|---|---|---|---|
| Curveball | 5 | 0 | 5 | 0% |
Inputs: Hunter Goodman's and Blake Snell's last-three-season lines (3104 PA / 1445 BF weighted 1x/2x/3x by recency), shrunk toward league average, blended with a log5 odds-ratio matchup model, conditioned on real league count-level pitch and batted-ball frequencies, and park-adjusted with committed park factors. The count process is calibrated per pair so simulated strikeout and walk rates land within 0.3% of the model targets (this page: 0.03% K error,0.03% BB error).
Head-to-head history never enters the simulation; a batter-pitcher pair shares a few dozen career PAs at most, and honesty means the model's priors are the players' own multi-season tendencies, not a 30-PA fluke. Full engine notes on the methodology page.