One line to remember: every pitch decide “swing or lay off” first; use skills by situation; the two-option call only shows up after contact. The sections below follow the actual order of one plate appearance, and every claim comes with numbers you can re-check.
Default: swing every pitch. With the Balanced gear (the default), swinging everything + taking the 1B gives roughly a 48.7% hit rate for the whole plate appearance, and any extra-base hit can only come from swinging; “lay off” never produces contact — it only stacks balls toward a walk (with the Balanced gear, laying off every pitch walks 34.4% of the time and strikes out the other 65.6%).
The opponent's gear can be read: ① the pitch face echoed under every result; ② the opponent's swing/lay-off habits this game; ③ against an AI it will adapt to your habits in return. The trade-off (numbers = “lay off every pitch” vs “swing every pitch”, 1000 games per cell in bot self-play):
| Opponent gear | Laying off everything | Adaptive to the gear | Advice |
|---|---|---|---|
Mostly balls bbb | 100.0% | 100.0% | Lay off — force the walk |
Ball-heavy bb | 96.9% | 79.3% | Lay off more |
Balanced bs (default) | 3.0% | 45.2% | Just swing |
Strike-heavy ss | 0.1% | 48.0% | Swing |
Mostly strikes sss | 0.0% | 52.3% | Swing — never wait |
Balanced 48.7% · Strike-heavy 54.7% · Mostly strikes 58.8% · Ball-heavy 40.4% · Mostly balls 36.9%; in single-player (gears weighted Strike-heavy 30% · Balanced 50% · Ball-heavy 20%) about 48.9%. Strikeout rate (counting a foul as a strike): Balanced 20.1% / Ball-heavy 32.4% / Mostly balls 38.8% / Strike-heavy 11.4% / Mostly strikes 5.5%.
The five pitch dice change the per-pitch contact chance and the chance that laying off is correct (the more ball-heavy, the harder to hit and the easier to read):
| Gear | Swing — contact | Lay off — correct | What it buys |
|---|---|---|---|
Mostly balls bbb | 38.3% | 83.3% | Bait swings and misses · double-edged (gives walks) |
Ball-heavy bb | 43.3% | 66.7% | Punish wild swings |
Balanced bs (default) | 53.3% | 50.0% | Neutral and safe |
Strike-heavy ss | 63.3% | 33.3% | Punish waiting |
Mostly strikes sss | 73.3% | 16.7% | Extreme strikeouts |
The “lay off is correct” rate does not depend on the whiff rate (it is the number of ball-family faces ÷ 6); the swing contact rate is the whiff-rate-weighted average over the five pitch types (dead-center 10% · strike 20% · nasty strike 50% · ball 40% · wide 80%). So swinging and laying off are not zero-sum: with the Balanced gear, 53.3% + 50.0% = 103.3%.
When the hit die lands on the mystery face you choose take the 1B (an immediate single, safe) or roll the second die (another roll — extra bases, or a double play).
| Second-die face | Weight | Result |
|---|---|---|
| DP double play | 1 | Batter + the runner closest to home are out (before 2 outs); after 2 outs only the batter is out |
| BB walk | 1 | Batter to first, runners advance one base |
| 2B double | 2 | Batter to second, runners advance two bases |
| 3B triple | 1 | Batter to third, runners advance three bases |
| HR home run | 1 | All runners + the batter score, bases cleared |
Odds: DP / BB / 3B / HR are 1 face in 6 each and 2B is 2 faces in 6 — only 1 in 6 is a double play; the other 5 in 6 are a hit or a walk.
| Choice | Bat | Hit rate | Out rate | Walk rate | Expected bases |
|---|---|---|---|---|---|
| Take the 1B | No | 48.7% | 51.3% | 0 | 0.487 |
| Take the 1B | Yes | 48.7% | 51.3% | 0 | 0.974 |
| Roll the second die | No | 42.7% | 54.3% | 3.0% | 0.668 |
| Roll the second die | Yes | 42.7% | 54.3% | 3.0% | 0.974 |
| Outs | Situation | Pick | Reason (expected-run gap in brackets) |
|---|---|---|---|
| 0 | all 8 states | Take the 1B | Keeping the plate appearance alive is worth the most (gap 0.04 ~ 0.30) |
| 1 | runner on third (third / first+third / second+third) | Take the 1B | The runner on third scores and the double play is only a single out (gap 0.04 ~ 0.17) |
| 1 | the rest (empty / first / second / first+second / loaded) | Roll the second die | Extra-base value is higher — but only by 0.006 ~ 0.17 runs, a thin edge; taking the 1B is fine too |
| 2 | runner on third only | Take the 1B | Scores him for sure (gap 0.04) |
| 2 | the other 7 states | Roll the second die | After 2 outs the double play only retires the batter, so the extra-base upside stays (gap 0.13 ~ 0.61) |
| Priority | Skill | When to use | Notes |
|---|---|---|---|
| 1 | Bat (passive) | Whenever you have one — every hit upgrades one base (1B→2B→3B→HR) | Doubles expected bases (0.487 → 0.974) |
| 2 | Steal (active) | Runner on first, second empty, before 2 outs | 66.7% to second; only a failure adds an out |
| 3 | Sac (active) | Runner on third, before 2 outs | 66.7% to drive in a run; the batter is out either way |
| 4 | Mist (active) | 0 outs with runners on | 3/6 advance runners, 2/6 nothing (still uses the plate appearance), 1/6 double play |
| 5 | Hack (active) | Well behind with runners on | 1/6 home run, 1/6 triple, 4/6 outs |
| 6 | Cmd (active) | When the half-inning allowance is down to its last slot and another skill is still usable | 50% to reset the allowance (a free 3 uses) |
| State | Order | Idea |
|---|---|---|
| Runner on first | Steal > Sac > Mist | Move the runner into scoring position first |
| Runner on second | Sac > take the 1B > Mist | Sacrifice him to third (much better scoring odds) |
| Runner on third | Sac > Mist > take the 1B > Hack | Sac scores him 66.7% of the time |
| Bases loaded | Take the 1B > Hack | The 1B forces a run in; but with 2 outs and the bases loaded, rolling the second die is better (section 4) |
| Event | Odds |
|---|---|
| Plate-appearance hit rate (swing all + take the 1B · Balanced) | 48.7% |
| Plate-appearance hit rate (single-player weighted gears) | 48.9% |
| Strikeout rate (swing all · Balanced, fouls counted as strikes) | 20.1% |
| Walk rate (lay off all · Balanced) | 34.4% |
| Two-option face frequency (swing all · Balanced) | about 18.1% |
| Steal success / runner out on failure | 66.7% / 33.3% |
| Sac success (batter out either way) | 66.7% |
| Hack: hit / out | 33.3% / 66.7% |
| Cmd success (resets the half-inning allowance) | 50.0% |
| Mist: advance / nothing (eats the plate appearance) / double play | 3/6 · 2/6 · 1/6 (only with 0 outs) |
Produced by an offline bot-vs-bot harness (1000~2000 games per cell, 1 game = 9 innings; re-run 2026-09-27 at the current whiff rates). It compares relative strength and does not equal a human win rate; the dice faces and skill semantics were checked line by line against the live engine.
| Skill tier | Aggregate win rate |
|---|---|
| Use freely (Bat / Steal / Sac + Mist / Hack / Cmd) | 52.5% |
| Only Bat / Sac / Steal (most conservative) | 43.3% |
| No skills at all | 43.8% |