How Weather Actually Moves a Projection: Dome vs Outdoor

Weather is the most talked-about variable in baseball analysis and one of the most misused. Temperature gets credited for swings it cannot produce, wind gets ignored when it matters most, and domes get treated as a boost when what they really change is something else entirely. Here is what the physics actually supports.

Only Three Weather Variables Do Real Work

Every forecast has a dozen numbers on it. For a batted ball, almost all of the signal lives in three of them: air density, wind, and whether the roof is closed. Everything else is either downstream of those three or too small to bother modeling.

Air density is the one that sets the baseline. Denser air means more drag, and more drag means a struck ball decelerates faster and lands shorter. Three things move density:

• Temperature. Warm air is less dense. A ball carries farther on a hot afternoon than a cold night.
• Elevation. This is the Coors story, and it is the largest single density effect in the sport. It is also constant, so it belongs in the park factor rather than the daily forecast.
• Humidity. Counterintuitively, humid air is less dense than dry air, because a water molecule is lighter than the nitrogen it displaces. Muggy conditions help carry slightly. The effect is real, and it is also small enough that it rarely changes a conclusion on its own.

Temperature: Real, But Smaller Than People Think

The rule of thumb that survives contact with the data is roughly 2 to 3 feet of extra carry per 10 degrees on a well-struck fly ball. That sounds trivial, and on any single ball it is. It matters because home runs live in a narrow band of distances, so a few feet moves a meaningful share of balls across the wall.

The practical version: a 25 degree temperature difference is worth a few percent on home run probability, not a few dozen percent. If you find yourself talking about a cold night like it neutralizes a hitter, the temperature is doing less work in that argument than you think. What a cold night actually does is push the marginal fly ball, the one that was going to be a warning-track out or a wall-scraper either way, over to the out side.

Wind Is the One That Can Actually Break a Projection

Wind is the largest daily weather effect by a wide margin, and it is the only one with enough range to swing a distribution rather than nudge it. A steady 12 mph blowing out to center is worth substantially more than a 25 degree temperature change. A 12 mph wind blowing in can erase most of a hitter-friendly park's home run advantage for that day.

Two cautions before you lean on wind too hard.

Direction only means something relative to the park

"Wind out of the southwest at 10" tells you nothing until you know which way the stadium points. Every park has its own orientation, so the same reported wind is blowing out in one city and across the field in another. Wind speed without the park's compass alignment is not usable information.

The reported wind is not the wind in the bowl

Forecasts report open-air conditions, usually from an airport several miles away. A stadium is a large concrete structure that redirects air. Some parks are notorious for swirling patterns that bear little resemblance to the posted forecast, and upper-deck construction has changed the internal wind behavior at more than one venue. Treat a strong reported wind as a strong signal and a light one as close to no information at all.

What a Closed Roof Actually Does

This is the part most analysis gets backwards. A dome is usually described as a hitter's environment or a pitcher's environment. It is neither, reliably. What a roof does is remove variance.

Indoors, air density is controlled, wind is zero, and the conditions are effectively identical from April to September. That does two things to a projection:

• The distribution narrows. The projection is not necessarily higher or lower, but the range of plausible outcomes around it is tighter, because one whole family of random inputs has been switched off.
• The park factor becomes more trustworthy. An outdoor park's multi-year factor is an average across every weather day the park has seen. A roofed park's factor describes conditions that will actually be in effect tonight.

A detail worth knowing: not every roof is a seal. Some venues have a cover over the field with open sides, which blocks rain and cuts wind without creating a controlled climate. Those parks behave like a sheltered outdoor bowl, not like a dome. And a retractable roof is a game-day decision, so a park in that column can be a dome one night and fully exposed the next.

The Double-Counting Trap

Here is the mistake that quietly ruins otherwise good weather analysis. A park factor already contains that park's average weather. Coors Field's factor embeds thin air. A Florida park's factor embeds summer humidity. A northern park's factor embeds cold Aprils.

So if you start from a park factor and then add a weather adjustment on top, you are counting the same physics twice unless your adjustment is measured against that park's own normal conditions. The correct unit is the deviation, not the absolute.

A Working Checklist

When you are looking at a matchup and want to know whether weather deserves a seat at the table:

1. Roof closed? Stop. Weather is not a variable today. Trust the park factor and move on.
2. Wind above 10 mph with a known direction relative to the park? This is the one weather input that can change a conclusion on home run and total base props. Check it before anything else.
3. Temperature more than about 15 degrees off that park's seasonal normal? Worth a small adjustment on contact-driven projections. Otherwise skip it.
4. Strikeout prop? Weather is close to irrelevant. Spend that attention on the lineup's contact profile and the umpire instead.
5. Rain in the forecast? This is not a projection question, it is an availability question. The risk is a delay or a shortened outing, which affects playing time rather than per-plate-appearance quality.

Why This Matters for Reading the Market

Weather is one of the few public inputs where the general conversation is systematically miscalibrated. Temperature gets over-weighted because it is the number everyone can see on their phone. Wind gets under-weighted because using it correctly requires knowing which way the stadium faces. Roofs get read as a level effect when they are a variance effect.

That asymmetry is where a careful reader finds disagreements worth investigating. When a market line has clearly moved on a hot forecast alone, the physics says that move is probably larger than it should be. When a strong wind is blowing in and nothing has moved, that is the more interesting situation.

Open a matchup on the dashboard and check the venue before you settle on a read. If the roof is closed, the weather conversation is over before it starts. If it is open and blowing, that is the number to look at first. PropPrizm is a statistical tool for informational and entertainment purposes and does not guarantee outcomes.