Felix Data Dump

by Jeff Sullivan on Aug 17, 2008 8:16 PM PDT in Seattle Mariners Stats

A bunch of good stuff from PITCHf/x with maximum presentation and minimal interpretation. Unfortunately not split by batter handedness.

Table 1: Outcome by Pitch

Table 2: Pitch by Outcome

Table 3:PITCHf/x Data by Outcome, Split by Pitch

pfxX = horizontal movement, pfxZ = vertical movement, Loc_X = horizontal location (negative = inside to righties, zero = center of plate), Loc_Z = vertical location

Figure 1: Felix Changeup, Aerial View

Figure 2: Felix Changeup, Side View

Figure 3: Felix Curveball, Aerial View

Figure 4: Felix Curveball, Side View

Figure 5: Felix Fastball, Aerial View

Figure 6: Felix Fastball, Side View

Figure 7: Felix Slider, Aerial View

Figure 8: Felix Slider, Side View

Tweet Comment 24 comments | 0 recs |

Do you like this story?

Comments

Did you notice any pattern to suggest a clear difference between a 2 and a 4-seamer

in Felix’s fastballs? And we just going to go ahead and call that 89-90mph thing a changeup?

by Matthew on Aug 17, 2008 8:23 PM PDT reply actions

I'm calling it a changeup unless it's clearly something else just for the sake of simplicity

It’s not worth trying to figure out what it’s supposed to be.

I’ll play around with his fastball tonight or tomorrow when I get the chance.

by Jeff Sullivan on Aug 17, 2008 8:25 PM PDT up reply actions

This is the unsolvable riddle

I’ve been playing around with everything (velocity, pfxX, pfxZ, spin, break length, break angle, etc), and the best I can do is that, when I split Felix’s fastballs into velocity > 96mph (4-seamers) and velocity < 93mph (2-seamers), the former has a pfxX of -6.4, while the latter has a pfxX of -9.2. But there’s so much noise in there that I don’t think it’s possible to take a group of Felix fastballs and assign them as 2 or 4-seamers with anything better than reasonable accuracy.

by Jeff Sullivan on Aug 17, 2008 8:49 PM PDT up reply actions

Oh and next time just ask for batter handedness to be added.

by Matthew on Aug 17, 2008 8:25 PM PDT reply actions

I kind of assumed it would be in there and then didn't realize it wasn't until I was too far along

by Jeff Sullivan on Aug 17, 2008 8:26 PM PDT up reply actions

Great, now I feel inadequate for not providing it.

Do I get demoted to Data Tugboat now?

by Matthew on Aug 17, 2008 8:27 PM PDT up reply actions

At this rate you'll never make it to Data Battleship

by Jeff Sullivan on Aug 17, 2008 8:28 PM PDT up reply actions

And this isn't a sorority?

9 = 8

by JI on Aug 17, 2008 8:42 PM PDT up reply actions

Graham does use a lot of Greek letters...

Romanes eunt domus

by Last Fan Of Jose Lopez on Aug 17, 2008 10:50 PM PDT up reply actions

We can be ηβπ

by Graham MacAree on Aug 17, 2008 10:53 PM PDT up reply actions

Data to support what we've been saying for years

Fastball = his worst pitch. Needs to throw more change-ups.

A 6% swinging strike rate at his fastball, despite having the highest average velocity of any starter in MLB. His fastball command sucks.

by davidcameron on Aug 17, 2008 9:15 PM PDT reply actions

As a fan, this is so hard to admit.

by Teej on Aug 18, 2008 1:24 AM PDT up reply actions

It's amazing to see how much more break he gets on the curve when he gets SStrks

Although it could be a little of SSS and a little of the fact that when his curve is at its best is when he gets swinging strikes.

by Mariner John on Aug 17, 2008 10:11 PM PDT reply actions

Kind of intuitive too - the more the ball breaks, the harder it is to hit

The Jose Lopez Watch - 147 H, 17 BB, 39 G Left

by seattlebruin on Aug 18, 2008 7:23 AM PDT up reply actions

Nerd

by Graham MacAree on Aug 17, 2008 10:35 PM PDT reply actions

Bazing!

by seattlesundevil on Aug 17, 2008 10:50 PM PDT up reply actions

That curveball looks scary

Romanes eunt domus

by Last Fan Of Jose Lopez on Aug 17, 2008 10:56 PM PDT reply actions

What I found interesting was interpreting the batter's approach

Adding the Ball% and the SL% to see what pitches hitters are attacking

hitters take his curve 70% of the time, meaning this is an excellent pitch if he has command. Getting ahead 0-1 is huge, and throwing a first pitch curve that might be taken can add to his arsenal

fastball is taken 52, slider about 54. Interestingly enough, hitters are most aggressive on his changeup, watching it only about 45% of the time. Though hitters chase it most often they have the least success against it. With the highest groundball rates and almost tied with the slider for highest SS rates. It’s also least often thrown for a ball, although you’d have to look at the pitch FX location to determine if this is because it is most accurate, or if the strike percentage is inflated by the high swing rate.

by mariners124m on Aug 18, 2008 12:13 AM PDT reply actions