Stats for The Environment — Context Variables
Park factors, weather data, and lineup position — the environment layer that makes MLB unique
You should read this if:
You bet MLB props and want to understand the mental models that drive outcomes.
The Core Insight
"Park factors are not just "hitter-friendly" or "pitcher-friendly." They break down by handedness, hit type, and even K rate. A 3-year rolling average by specific factor is the minimum for accurate projections."
The MLB Mental Model
Park Factors (granular)
HR by LHH, HR by RHH, 2B, 3B, K, BB — all separate
Predicts: Yankee Stadium inflates LHH HR but is neutral for RHH. Petco suppresses doubles but is average for HR.
Weather Variables
Temperature, wind speed/direction, humidity
Predicts: 10 degrees F = 3-5 feet of ball carry. Wind out at Wrigley is a different sport.
Lineup Position
Leadoff ~4.5 PA, 3-5 hole ~4.0 PA, 8-9 hole ~3.5 PA
Predicts: PA count is the volume floor for every counting stat
Time of Season
Stabilization timeline determines which stats are trustworthy
Predicts: April K% is real. April batting average is noise. July everything is real.
The Environment Stats That Matter
MLB is the only major sport with first-class environmental variables. Park, weather, lineup order, ump, and the calendar can each swing prop outcomes 10–30%. The book treats each of these as a real, modelable factor — not a vibes adjustment.
Park Factor
Granular contextRatio of runs (or HRs, or doubles, or Ks) at home vs. on the road for all teams, over a multi-year rolling window. 1.00 = neutral. Must be broken out by handedness and hit type — Yankee Stadium inflates LHH HR but is neutral for RHH.
Why it matters
A single "hitter-friendly" / "pitcher-friendly" number is lossy. Coors inflates RHH HR 22% but LHH HR 18%. Petco suppresses doubles but is average for HR. Granularity is the whole point.
How to use it
Apply granular HR factor (by hitter hand) to HR projections; apply granular K factor to K projections. Coors LHH HR factor 1.18 = +18% HR-rate multiplier on a lefty's HR projection there.
| Tier | Value | What it means |
|---|---|---|
| Elite | 1.15+ hitter or ≤ 0.85 pitcher | Extreme park — Coors (offense) / Oracle (pitching) |
| Great | 1.08 / 0.92 | Strong tilt — stack with weather |
| Average | 0.95 – 1.05 | Effectively neutral |
| Poor | Wrong-park signal | Park opposes your thesis — fade conviction |
Weather — Temperature
HR inputGame-time air temperature. Warmer air is less dense, so batted balls carry farther — roughly 3–5 feet of added carry per 10°F.
Why it matters
Per BIG-IDEAS.md: "Temperature alone can swing HR rates by 10–15%." This is a first-class variable, not a vibes factor.
How to use it
80°F+ game + hitter park + wind out = stack HR overs. Sub-60°F game + cold bats + wind in = fade HR overs.
| Tier | Value | What it means |
|---|---|---|
| Elite | 85°F+ | Ball travels max — +10–15% HR rate |
| Great | 75 – 84°F | Warm, supportive carry |
| Average | 65 – 74°F | Neutral |
| Poor | < 55°F | Cold, dead air — −10–15% HR rate |
Weather — Wind
HR inputWind speed and direction relative to park orientation. Wind out to center/left/right adds carry; wind in subtracts it.
Why it matters
Wrigley is the extreme case — wind out can double HR rates vs. wind in. At coastal parks (Oracle, Fenway), wind direction matters more than temp.
How to use it
10+ mph wind straight out + warm temp + hitter park = HR overs conviction stack. 10+ mph wind straight in = fade HR overs even in a hitter park.
| Tier | Value | What it means |
|---|---|---|
| Elite | 15+ mph out | Huge HR boost — wind doubles the park |
| Great | 8 – 14 mph out | Meaningful HR tilt |
| Average | < 8 mph any | Neutral |
| Poor | 10+ mph in | HR suppression — flyballs die at warning track |
Weather — Humidity / Altitude
HR inputDry air is less dense than humid air — so ironically it carries *less* than humid air of the same temperature (counterintuitive physics). Altitude trumps both: Coors thin air adds 10% distance on every ball.
Why it matters
Per DMP-APPLICATION-GUIDE.md: ~3–5 ft of ball travel per 10°F, stacked on top of park-altitude effects. Coors' "humidor" is specifically designed to mitigate altitude HR inflation.
How to use it
Coors at high elevation with any warm, dry day = stack power. Houston closed roof with high humidity = neutralizes heat boost somewhat.
| Tier | Value | What it means |
|---|---|---|
| Elite | Coors (5,280 ft) | Altitude-driven 15%+ HR boost baked in |
| Great | 1,000 – 3,000 ft | Modest altitude assist (Arizona, Atlanta) |
| Average | Sea level | Neutral |
| Poor | High humidity + cool | Dampens power ceilings |
Lineup Position
Volume floorThe book's PA-per-game distribution by lineup slot. Leadoff hits ~4.5 PA per game; the 3-5 hole averages ~4.0; the 8-9 hole averages ~3.5.
Why it matters
PA is the volume floor on every counting stat prop. A hitter dropped from leadoff to 8th loses roughly 1 full PA per game — a 20%+ hit to their hits, TB, and runs projections.
How to use it
Always use the *confirmed* lineup, not the projected one. Watch for demotions or day-off lineups. Platoon specialists in the 7-hole vs. wrong hand should be graded vs. 3-PA expectations, not 4-PA.
| Tier | Value | What it means |
|---|---|---|
| Elite | 1-3 hole (~4.4 PA) | Volume ceiling highest |
| Great | 4-5 hole (~4.1 PA) | Premium RBI spot |
| Average | 6-7 hole (~3.9 PA) | Typical regular |
| Poor | 8-9 hole (~3.5 PA) | Volume-capped; fade counting-stat overs |
Umpire Strike Zone
K/BB contextHome-plate ump's historical called-strike rate and zone shape. Some umps call a huge zone; others a pinched one.
Why it matters
Not covered by the book but flagged in BIG-IDEAS.md as a real, modelable edge. Wide-zone umps inflate K% for both teams and suppress BB%. Tight-zone umps do the opposite.
How to use it
Wide-zone ump + control pitcher + zone-pounder matchup = K over, BB under. Tight-zone ump + wild pitcher = walks over, IP under.
| Tier | Value | What it means |
|---|---|---|
| Elite | Top-decile wide zone | Big K boost both ways |
| Great | Above-avg zone | Mild K tilt |
| Average | League-avg | No signal |
| Poor | Tight / pitcher-hostile | Walks over / IP under risk |
Stabilization Points
Reliability gateNumber of observations needed for a stat to cross from "mostly noise" into "reliable signal." The book's appendix calls K% (batter) ~60 PA, ISO ~160 PA, BA ~910 PA, BABIP (batter) ~820 PA, K% (pitcher) ~60–70 BF.
Why it matters
Tells you which signals to trust when. Treating a hot April BA like a real skill is how casual bettors get mowed down; treating a hot April K% like real skill is sharp.
How to use it
Weight = min(1, sample_size / stabilization_point). April K props are live; April hits props are largely noise. July is when everything is reliable.
| Tier | Value | What it means |
|---|---|---|
| Elite | K% (P): ~70 BF | Trust immediately — 2-3 starts |
| Great | ISO: ~160 PA | Trust by mid-May |
| Average | SLG: ~320 PA | Trust by June |
| Poor | BA / BABIP: ~820 PA | Not reliable until late July |
"K% stabilizes in ~60 PA. BABIP takes 820." — BIG-IDEAS.md #5
Times Through the Order (TTOP)
In-game decayThe systematic decline in pitcher performance each successive time through the lineup. 2nd time slightly worse than 1st; 3rd noticeably worse — K% drops, BABIP rises, HR/FB spikes.
Why it matters
Per BIG-IDEAS.md: "The third time through the order is where pitchers fall apart." Not covered by the book but treated as a first-class input in DMP.
How to use it
If a pitcher is projected for ≥ 6 IP he'll face the order a 3rd time — apply TTO penalty to K% and HR/FB for late-inning batters. Pitchers pulled at 5 IP escape TTOP entirely.
| Tier | Value | What it means |
|---|---|---|
| Elite | 1st time wOBA baseline | Full K%/ER edge intact |
| Great | 2nd time +5-8 wOBA pts | Slight decay |
| Average | 3rd time +15-20 wOBA pts | Significant — adjust ER/K projections |
| Poor | 4th time | Extreme — pitcher should be out by now |
Framework in Action: Coors Field Is Not Just "Hitter-Friendly"
Coors inflates HR for right-handed hitters by 22% but only 18% for lefties. It inflates K rate less than most people think. It dramatically inflates doubles and triples. A blanket "Coors = boost" misses the granularity. DMP uses park factors broken down by handedness and hit type for accurate adjustments.
When to Apply This Framework
- ✓Check park factor and weather before every outdoor MLB prop bet
- ✓Lineup position determines PA count — crucial for counting stat props
- ✓Time of season determines which signals to trust — use the stabilization calendar
When to Pass
- ⚠️Do not assume a single park factor number captures the full picture — use granular data
- ⚠️Do not ignore weather for power props — temperature and wind are first-class variables
- ⚠️Do not trust BABIP-dependent signals before they have stabilized
Key Takeaways
- ✓Park factors must be granular: by handedness and hit type, using 3-year rolling averages
- ✓Weather is a first-class variable in MLB — 10-15% HR rate swings from temperature alone
- ✓Lineup position is the baseball equivalent of minutes and usage combined
How DMP Helps
DMP provides granular park factors, real-time weather with HR impact estimates, confirmed lineup data, and seasonal confidence flags.