[princecharles]

I've finally compiled my MLB weather/pitching super subset data.

Hyper-specific humidity and barometric pressure cross referenced with successful pitchers who rely heavily non fastball (4 seam), and non knuckleball pitches.
Specifically curveballs, sliders, and two seam fastballs require 'thicker' air to be effective.
I actually took pitching lessons to further internalize what must happen for a 'movement' pitcher to be on his game.

Anyone else ever travel this path?

[Justin7]

Other groups are using humidity and pressure. I don't know whether they have it indexed by pitcher type.

Did you analyze this compared to market totals? Or MLs?

This type of analysis has a good chance of winning if you do it correctly.

[yak merchant]

I've finally compiled my MLB weather/pitching super subset data.

Hyper-specific humidity and barometric pressure cross referenced with successful pitchers who rely heavily non fastball (4 seam), and non knuckleball pitches.
Specifically curveballs, sliders, and two seam fastballs require 'thicker' air to be effective.
I actually took pitching lessons to further internalize what must happen for a 'movement' pitcher to be on his game.

Anyone else ever travel this path?

I've looked into it before, and I'm looking into it again now. You are correct that "thicker" air makes the balls break more and also helps out fly ball pitchers. However Air Density is way more about temperature and elevation. Contrary to common perception, humidiy actually makes the air "thinner". However people swear high humidity correlates to unders, and I do think there is a point of humidity where the ball get's "Saturated" (probably due to sweat) and therefore is heavier. Humidity at 40% versus 20% might make the ball travel farther (i.e. very minimally), but when humidity get's high you may get the opposite effect due to "saturated ball". Obviously this has some merit after learning what we learned from the Coors field humidor effect.
As far how to apply air density, I never came to any significant conclusion but that was pre pitch f/x. I'm of the opinion that certain pitchers match up with certain air densities. i.e. depending on your velocity and spin, there is an optimal air density for "you" that causes "late movement". From there you can go into all kinds of discussions on coefficient of drag and theories that a baseball goes through a drag crises at certain velocities/air densities. A statistics tard like me getting pitch f/x data normalized to draw any conclusions is the hardest part. Not to mention if you run the numbers, a 5 mile an hour wind is way more important than even a high temparture, add the fact that wind reacts differently in every park and you have something seriously hard to quantify. For example wind blowing in from right at Texas Rangers stadium actually causes "jet stream" effect that hits stadium behind home and causes an outward helping "wind" to right. In some stadiums wind has little or no effect at field level (i.e affects pitches) while in some I would guess it affects the pitches all the time (i.e old Candlestick).

I guess I don't really have any good advice, other than starting with converting your data points to air density, and tracking that,humidity, and wind direction/velocity. From their you'll have to draw you own conculsions.

[HeeeHAWWWW]

Other elements:


The ball becoming larger as humidity causes it to swell.

The difference between absolute and relative humidity.

The difference between ground temperature and air temperature on a sunny day.

[HUY]

I've finally compiled my MLB weather/pitching super subset data.

Hyper-specific humidity and barometric pressure cross referenced with successful pitchers who rely heavily non fastball (4 seam), and non knuckleball pitches.
Specifically curveballs, sliders, and two seam fastballs require 'thicker' air to be effective.
I actually took pitching lessons to further internalize what must happen for a 'movement' pitcher to be on his game.

Anyone else ever travel this path?

Have you made an independence test to see if the results are dependent or not on this weather data? I think this is the first question that should be asked.

[princecharles]

I've looked into it before, and I'm looking into it again now. You are correct that "thicker" air makes the balls break more and also helps out fly ball pitchers. However Air Density is way more about temperature and elevation. Contrary to common perception, humidiy actually makes the air "thinner". However people swear high humidity correlates to unders, and I do think there is a point of humidity where the ball get's "Saturated" (probably due to sweat) and therefore is heavier. Humidity at 40% versus 20% might make the ball travel farther (i.e. very minimally), but when humidity get's high you may get the opposite effect due to "saturated ball". Obviously this has some merit after learning what we learned from the Coors field humidor effect.
As far how to apply air density, I never came to any significant conclusion but that was pre pitch f/x. I'm of the opinion that certain pitchers match up with certain air densities. i.e. depending on your velocity and spin, there is an optimal air density for "you" that causes "late movement". From there you can go into all kinds of discussions on coefficient of drag and theories that a baseball goes through a drag crises at certain velocities/air densities. A statistics tard like me getting pitch f/x data normalized to draw any conclusions is the hardest part. Not to mention if you run the numbers, a 5 mile an hour wind is way more important than even a high temparture, add the fact that wind reacts differently in every park and you have something seriously hard to quantify. For example wind blowing in from right at Texas Rangers stadium actually causes "jet stream" effect that hits stadium behind home and causes an outward helping "wind" to right. In some stadiums wind has little or no effect at field level (i.e affects pitches) while in some I would guess it affects the pitches all the time (i.e old Candlestick).

I guess I don't really have any good advice, other than starting with converting your data points to air density, and tracking that,humidity, and wind direction/velocity. From their you'll have to draw you own conculsions.

Please excuse the late reply.
You're spot on in many areas, and a bit off in some.
Sweat typically doesn't have the time to saturate a baseball used in the MLB. Not enough time the average ball spends in play, and the most likey culprit for 'delivering' the sweat would be the pitcher, who has his rosen/talc bag to keep his pitching hand dry.

My ideas have continued to widen, as I'm looking at the popular prop, 'Will there be a run scored in the first inning?'.

My working theory is to isolate the extreamly valuable 'good for pitcher = bad for hitter' variables.

This has taken me a bit out of my league, as I'm generating 3D models of ballparks, then overlaying my key factors, some of which include angle of solar insolation (angle at which the sun is hitting the earth), then parse this through the resulting shadowing effect of the infield, which is a particularly strong variant in the early innings of summer west coast early evening start games.

Also, I'm having a lot more luck this year than last In aquiring feeds to web cams in some amazing ball park sections. In particular, the starting pitcher warm up bull pen was very difficult to get, and is a bit unreliable, but offers up much qualitative info.

Mind you, this MUST be generated real time, which actually is the easiest of my hurdles.

Write back if this interests you.