Alan Nathan
banner
pobguy.bsky.social
Alan Nathan
@pobguy.bsky.social
Professor Emeritus of Physics, U of Illinois. Physics of Baseball: baseball.physics.illinois.edu. Mostly post about baseball. Sometimes Con Law, basketball, physics, WW1, Twilight Zone, ...
Need I say more...
October 18, 2025 at 5:18 PM
No Kings, Urbana IL...just getting started
October 18, 2025 at 4:51 PM
Today is the 65th anniversary of Ted's last swing (shown below), resulting in his 521st home run.
September 28, 2025 at 1:12 PM
Out-of-Place 'Devil Bird' Wows Spectators in Maine, the First Anhinga Ever Seen in the State share.google/htcC26o2eDqL...

Actually I spotted several yesterday in Camden ME. See one in my photo below.
September 4, 2025 at 1:33 AM
Here is a followup to my post about 500 ft being max distance, based on 2022 Statcast data.
July 17, 2025 at 12:17 PM
Here's another try w/most of selection bias removed. For the regular season, I am plotting only home runs and only those hit at Truist Park. I find the distance of those hit at the HRD exceed (on average) those hit during regular season by 16 ft. More for the longest home runs.
July 16, 2025 at 1:46 AM
Here is another way to look at the data. I use a GAM to fit fly ball distance vs EV,LA for regular season, then apply fit to HRD, then plot actual vs fitted distance (red line is when actual=fitted). Note for HRD, actual exceeds fitted by average of 33ft. Air density not taken into account.
July 15, 2025 at 9:00 PM
At the HRD last night, the ball was carrying better than typical. In the graphs below, I show home run distance vs. exit velo for LA in 25-30 deg range, both for the 2025 season and for HRD. The red line on both graphs is a crude eyeball fit to the upper limit of the regular season data.
July 15, 2025 at 6:12 PM
I expected best square-up to occur when AA matches both the VAA and centerline angle, in which case the LA would be equal to AA (if I neglect the spin of the pitch). It doesn't look like your analysis agrees with that.
June 2, 2025 at 8:55 PM
For 2024 data, I have plotted collision efficiency q vs. LA-AA. Note that q removes variation due to both bat speed and pitch speed (q=(EV-vbat)/(vpitch+vbat). It peaks close to LA-AA=0, just as I would expect. Note that I have reversed the horiz and vert axes from your plot.
May 31, 2025 at 10:00 PM
May 15, 2025 at 9:30 PM
As we enter the Passover season, I thought I'd share this letter from our local newspaper.
April 12, 2025 at 2:47 PM
Here's a screenshot of Dan Russell and me on our Zoom call yesterday discussing Torpedo bats and the testing we are preparing to do. I am holding a Torpedo-like bat that Louisville Slugger made for me back in 2000.
April 1, 2025 at 7:42 PM
@sunshinevvn.bsky.social My first attempt, using a reasonable guess as to the bat profiles, old vs. torpedo. Both bats 33"/31 oz. MOI's are not much different. Solid curves are EV, dashed curves are q. Max of EV shifts out ~0.5".
March 30, 2025 at 4:06 PM
So I did a study, using the same data and technique discussed here: tht.fangraphs.com/pitch-moveme..., the results of which are shown in the graphs. While there are systematic differences, they are of the order of 0.1 inches or less. (2/3)
February 25, 2025 at 8:00 PM
Very interesting article which I highly recommend. One conclusion is that it is better to mis-hit on the knob side of the sweet spot than on the barrel tip side. There is some interesting physics as to why that is so. See clip from here:
baseball.physics.illinois.edu/THTAnnual201...
February 16, 2025 at 6:47 PM
Just to convince you that "theory" (based on physics) is actually realized, take a look at the random sample of pitch data, sorted by pitch type. Upper plot shows movement (M) vs v, color coded by Cl. Lower is M vs Cl, color coded by v.
February 6, 2025 at 5:03 PM
No words necessary...
February 2, 2025 at 11:21 PM
Here's a nice example.
January 22, 2025 at 2:50 PM
How great is this! A menorah fit for a baseball fan!
December 29, 2024 at 10:57 PM
I have been maintaining that pitch movement is independent of velocity for a fixed ratio of spin to speed (i.e., for fixed Bauer units). Not only is this true based on physics, it is also supported by the data. Below I show movement vs. Bauer for different fixed value of velo.
December 27, 2024 at 6:01 PM
Day 2
December 27, 2024 at 12:09 AM
My version of the same cards. Wilt is badly off center, West a little. O and Elgin are pretty good.
December 26, 2024 at 7:51 PM
Looks like I can post a gif, so here is mp4 instead.
December 24, 2024 at 11:12 PM
Great example here. Note that this is a visual confirmation of the bat-tracking analysis shows quantitatively shows the same thing, as per @tangotiger.com
December 24, 2024 at 7:17 PM