THE BOOK--Playing The Percentages In Baseball

Tom,

Thanks for that link to your study.  Really interesting stuff.  I am glad you liked the article and I hope to provide a lot of delirious moments ahead.

I posed this at USSM before seeing this:

Using the fastball data off of Fangraphs, I see, for one, most people lose velocity all of the time, but velocity seems to be stable around age 26-27, then dropping more and more and more as the pitcher gets older from there.  If I use only >= 400 BFP, I get an average of .35 mph lost per year, with it being ~.2 mph in their 20’s, .33 mph or so in their early to mid 30’s, and .5 mph in their mid to late 30’s, and .75+ per year after that.  That’s a total of 206 year to year samples.  If I increase it to 200+ BFP, I get .23 mph lost on average, but that undoubtedly includes moves to the bullpen, there is still over all loss (on average) for every age group, 23-24, 24-25, etc.  There does appear to be less loss in a pitchers late 20’s that say in their late 30’s, but that’d be expected.

The reason I used 400+ BFP (I should have used # pitchers, but I think that’s OK) is to weed out the moves to and from the bullpen, thus only year to year for starting pitchers.  The aging patterns are obvious that way.

here’s the data

Age Diff Count
21 0.4 1
22 -0.7 1
23 -0.14 5
24 -0.15 6
25 -0.591666667 12
26 0.088235294 17
27 -0.038888889 18
28 -0.359259259 27
29 -0.357692308 26
30 -0.029411765 17
31 -0.607692308 13
32 -0.245454545 11
33 -0.32 10
34 0.114285714 7
35 -0.3 5
36 -1.02 5
37 -0.566666667 3
38 0.3 1
39 -0.55 2
40 -0.283333333 6
41 -0.75 6
42 -1.05 2
43 -0.5 1
44 -0.6 2
45 -0.65 2

It just dawned on me, that since Fangraphs doesn’t seperate 2 seamers from 4 seamers that could alter the results, if you use that data.  It has more endpoints than the pitch f/x data, but it doesn’t include information about the movement.  I agree with Tango, aging for movement, as well as for pitch location (as in do pitchers start staying out of the middle more) is important, and would shed light on velocity changes.  Do pitchers throw more 2-seamers as they age.  I’d think so, but it’s be interesting to find out.

Good stuff Chris.

The Marcel aging curve, for performance stats (OBP, SLG, etc) is:

(age - 29) * x
where x=.006 if under 29 and .003 if over 29

You multiply or divide as logic dictates.

The cool thing about such a simple equation is that it accelerates the curve the farther away you are from age 29.

It would make sense that fastballs follows something like:
(age - 29) * x
where x=.0015 if under 29 and .0030 over 29 (numbers for illustration purposes only)

Chris, if you can best-fit your equation against this kind of scheme, I’d like to see the results.

***

As an example of the above, if you are 39 years old, you get: (39-29)*.003 = .03

So, your aging from age 39 to age 40 would be a 3% decline in effectiveness. 

If you were 19, you’d get a 6% increase in effectiveness.

This is a general rule, and of course, each component ages differently.  Speed at one extreme declines almost immediately.  Walks at the other extreme improves until almost retirement.

Yeah that is interesting Chris.  Again my data has the issue of being from the end of 2007 and the start of 2008 so if players who are around 28 wear down less than younger or older players that would skew my results.  Also, many pitchers will throw a cut fastball that can be mistaken as a fastball.  I am removing them as they tend to be slower like the sinkers do.  Not sure how much that would change the data though.

Josh—this is great stuff. One question: did you account for differences in temperature? I suspect that the bulk of pitchf/x data last year was in August/ September, when it would have been hotter than April/ May this year ... that could skew some of the data

John