Designated HitterMay 14, 2009
Johan Santana's Fast Start in PITCHf/x
By Harry Pavlidis

Johan Santana - have you heard of him? He's pretty good. The man is the ace of the Mets, was the ace of the Twins, and is one of the best left-handers in the game. He does it with a consistent, metronomic delivery that pumps out four difficult pitches.

Change-up (CU)1427152127581.27.06.8134.6
Two-seam fastball (F2)102216685692.07.67.8135.7
Four-seam fastball (F4)1992645134792.25.410.2152.1
Slider (SL)58232226084.50.53.6171.3

Notes: PITCHf/x data from Gameday, classifications by the author ("cfx"); data covers 2007 (partial), 2008 and 2009; mph is the average speed at 55 ft. from the back of home plate; pfx_x and pfx_z are the lateral and vertical deviation from the path of a spin-less ball (inches); deg is the angle of the spin axis


Santana's slider is one of the best in baseball, which is a fine indication of the consistency of his delivery. But that's all old news. What brings me here is to explore Johan 2009. He's off to a great start, even better than years past, which begs a simple question. What's he doing differently? If anything, that is.

It's early, and I'm only looking at games through May 6, so this doesn't include Johan's most recent start. Some trends have emerged that merit watching. That's about all you can do with most early season returns. Keep that in mind.

The biggest change is in pitch selection. Johan is throwing far more four-seam fastballs (or simply "fastballs") and far fewer two-seam fastballs ("sinkers"). Santana also appears to be throwing fewer sliders, a pitch he mostly uses against lefties. His change-up is primarily a gift to right-handed hitters everywhere (the gift of zilch, that is) but got a little extra use against lefties in 2008.


That's a siginficant increase in heaters. Another look is from a four-start moving average of pitch mix.


I made sure to include this chart, because, when you squint, you can see a giraffe. But why is he doing this? Santana's four pitches are all above average. The change is one of the best, and both of his fastballs and the slider are solid pitches.

CH  -3.7
F2  -1.7
F4  -2.0
SL  -1.5

If you're going to cut back on two pitches, they'd be the sinker and slider. I'm not sure why you would, neither pitch is hurting anyone but Santana's opponents. Breaking it down by season and, for good measure, batter hand, you do start to get the idea that the sinker and slider aren't what they used to be, while the change and fastball may be even better.


It's early, Santana is one of the greats and can beat you a few ways, so I'm not reading too much into this. I'm working with a short season and a partial data set (2007 didn't have full PITCHf/x coverage), too. But he's pitching well, he is throwing more heaters and fewer sinkers, and Santana's change-up is still a world beater.

Harry Pavlidis writes for Beyond the Box Score, The Hardball Times and Out of the Ivy. His own blog, Cubs f/x, feels neglected once in a while.


right now I'd say JS's only problem is he seems to shut down BOTH teams' offenses when he pitches!

Nice work. But pie charts? Why not present two separate line graphs, similar to those presented at the end of the article?

Pie charts seem more appropriate for showing proportions, such as pitch mix, than line graphs. I use a line graph, by necessity, to show the moving average over time.

What's wrong with pie charts, anyway?

I think it looks more like a camel, but maybe that's just me.

So are we sure that he's really throwing a two-seam and a four-seam fastball? They look suspiciously like one another in their movement. Could it be that they're the same pitch, just split into two groupings?