What did my throwing look like today?

Recovery, recovery, recovery.

This is one of the easiest ways to keep up with the grind of a long season. Keeping the low days low helps keep the high days high.

My chronic workload has lived in the 11-12 range over the last week. As I acclimate to the spring training schedule and prioritize my compete days, my ACWR has remained at or below 1.0.

After pitching yesterday, my CW was 12.1 and my ACWR was 0.98.

As you may notice from the plot above, intensity seems to vary by pitch type. In this outing, my two fastballs had the highest torque on average and the slider had the lowest.

Interestingly, all pitch types had similar arm speeds. Good for deception perhaps?

My throwing intensity (determined here as a % of the top 5 highest torque throws of the outing) was low pre-game and ramped up into my appearance.

During the season, when there is less certainty about when I will pitch, pre-game throwing becomes much trickier to manage.

On the throwing side, I made 80 total throws on the day for a total of 17.5 workload on PULSE. The vast majority of that workload came from my warmup and game pitches.

My WHOOP clocked me at 8:29 of sleep the night before the outing (and that's with the time change from daylight savings!).

Going to keep an eye on any potential correlations between sleep and velocity/performance as the year goes on.

According to Weather Underground's history feature, Clearwater was 79-80°F while I was pitching.

I'll be tracking this throughout the season to see how temperature affects in-game heart rate.

www.wunderground.com/history/dai...

While not as consistently high as my big league spring training outing the other day (I guess the backfields ain't as stressful as the show), my heart rate still peaked at 178 bpm.

12:54: first pitch (157 bpm)
1:02: last pitch (178 bpm)

Looking at this example from later in the article:
- Red dots are FF total movement
- Orange dots are SI total movement
- White dots are spin-implied movement
- Black dots are non-Magnus movement (total - spin-implied)

Are you saying we can find the white/black dots using Alan's work?

Long story short is that we want to recreate this plot.

tangotiger.com/index.php/si...

Goal is to use observed tilt and active spin to figure out the Magnus-implied movement and subtract that from total movement to get SSW/other movement.

Is there a way to do this with publicly available data?

Just to confirm, arm angle is determined by drawing a line from the shoulder to the ball at release and finding the angle relative to a line parallel with the ground?

FanGraphs defines a + stat as a “stat with a baseline of 100, where 120 represents a performance 20% better than average and 80 a performance 20% worse than average.”

blogs.fangraphs.com/fangraphs-sp...

OG Driveline Stuff models were scaled this way too.

www.drivelinebaseball.com/2021/12/what...

That was one of the original purposes of using accelerations rather than movement that I saw described by @choicefielder.bsky.social and @jack-lambert.bsky.social: model parameter orthogonality.

This chart from Lambert demonstrates the problem well, and your idea would also work as a solution.

Looking further into your work, the transverse spin components are a function of the transverse acceleration components (amagx, amagy, amagz below).

If we know the true (observed) spin components, could we back-calculate what the Mangus (spin-induced) accelerations were using these equations?