SB Nation Dallas Editor's Pick
Feliz and Fastball Movement
by e2drummer on Jul 27, 2011 4:47 AM CDT
Email
PrintNeftali Feliz was absolutely dominant last year as a closer. This year... not so much. He currently ranks 13th on the team in pitching WAR, just behind Dave Bush, his K/9 has dropped from 9.22 in 2010 to 6.27 in 2011, his BB/9 has spiked from 2.34 to 4.58, and he has blown 20% of his save opportunities after blowing only 7% last year. How can a pitcher who is throwing the same pitches he threw last year have such different results?
While arguing an admittedly stupid opinion in the postgame thread, I went to look at Fangraphs and their Pitch F/X data on Neftali Feliz. I remembered reading the article "What's Wrong With Neftali Feliz?" by Dave Cameron on Fangraphs and being intrigued by his observation about the loss of some horizontal movement on Feliz's fastball. It was dismissed at the time as SSS, but two months later, the sample size has grown a bit larger. Here's the data:
Horizontal Movement:
Vertical Movement:
Neftali Feliz's Fangraphs Page
Looking at his movement charts, there's now enough of a sample size to say that he has lost significant horizontal movement on his fastball (FB). On average, it's showing he has lost a full 2 inches! of H-movement on his FB with no loss in V-movement. While some will point out that because his V-movement is the same as last season's, his H-movement probably shouldn't change, V-movement and H-movement are not necessarily tied together and the consistency of his V-movement between the past two seasons suggests that there may not be a funky error in the Pitch F/X calibration.
Looking at his breaking ball, his curveball/slider (CB/SL) H-movement shows a trend in the positive movement direction, same as his fastball. There could be two possible reasons for this: a. Pitch F/X has been calibrated differently this season, causing a positive direction trend in horizontal-movement for all his pitches or b. Feliz is spinning a tighter breaking ball while losing movement on his fastball. While both his FB and CB/SL show changes in H-movement, the limited number of changeups (CU) he has thrown this year show no change in H-movement from previous years, leading me to believe that the overall trend cannot be blamed on a system-wide error in the Pitch F/X data. Feliz is more than likely getting more H-movement on his breaking ball and less on his fastball.
With Feliz's struggles in the closer role this season and no other obvious reasons for the decline in his numbers, you have to wonder about the effects of his loss in horizontal fastball movement. He has not lost any FB velocity (95.8 average this season vs. 96.3 last) and he's still throwing his fastball at a similar rate (79.5% vs. 82.2%), but because of his heavy reliance on a dominant fastball, any change in movement should have a huge effect on his ability to get batters out. This is shown in the increased amount of contact batters have made on Feliz's pitches this season and the decrease in swinging strikes. The only thing it doesn't fully explain is Feliz's absolutely horrendous splits against RHB this season (0.43 K/BB), while his splits against LHB (4.00 K/BB) are great (even though they still regressed from last season). Those L/R splits are truly baffling. Maybe he's suddenly having an extremely tough time finding the outside corner to RHB?
Another obvious explanation for his drop in performance is that the league is adjusting to him in his second season. This is probably at least partially true, but I don't think it fully explains such a significant drop in his peripherals from last season. If it did explain it, we would have seen evidence of batters beginning to make adjustments last season.
The biggest question I have regarding the large drop in fastball movement is simply, what caused it? The only thing different from last season is that Feliz added a cut-fastball. Not being a pitching expert myself, I'm not sure how that would have any effect on the way a 4-seamer comes out of his hand.
I think we can all agree that something has changed about Feliz's pitching between last season and this season, even if we don't know exactly what it is yet. Whatever the problem is, it's up to people much more knowledgeable about pitching than me to figure it out. Hopefully, we see the 2010 Neftali Feliz back soon.
17 comments
|
7 recs |
Do you like this story?
Comments
He may need a "DL stint" soon.
"The Angels are like the villain in the movie that isn't dead until he's been stabbed 150 times in the bath tub, yet he still might come back up one more time." - Eric Nadel
Other explanations...
-A PITCHf/x calibration change coupled with an improved changeup.
-PITCHf/x is likely misclassifying cutters as 4-seamers and vice-versa on many occasions.
The first explanation is possible, but very unlikely
The second one doesn’t appear to be true. Some fastballs are mistakenly being misclassified as cutters and vise-versa, but that wouldn’t explain the loss of two inches in movement on ALL his fastball data, only some. His changeup also has almost the exact same movement as his 2010 fastball, but because it has a significantly lower velocity, there’s very little chance that Pitch F/X is wrongly classifying the pitch.
"Certainly, we were penalties, and we stopped ourselves." - Les Miles
What do you mean by "ALL his fastball data"?
He didn’t throw a cutter last season. That would definitely add noise to his 4-seamer this season. Take a look at the spin movement charts. The FF and FC pitches are intermingled. There’s no question that some cutters are cutting into his averages.
Also, the first week of the season, the PITCHf/x system was calculating data incorrectly. I have still not updated my site’s data (been busy, you know), and that may also be the case with FanGraphs’ data.
by NoNameOnCard on Jul 28, 2011 12:30 AM CDT up reply actions
I'm saying that while
incorrectly classified cutters may mess with his overall average movement, there is still very little data suggesting his fastball has not changed at all. You’re looking for errors in the data, while I’m looking for any possible evidence that his movement hasn’t changed.
If he hasn’t lost any movement on his fastball and incorrect pitch classification is completely to blame, there would be a large number of pitches with the same movement as his fastball from last season with a few cutters mixed into the data. Instead, all we see is a very large number of fastballs that DON’T have the same movement as last season. The addition of a cutter that he uses very, very rarely would not explain that change.
Only a change in the Pitch F/X calibration at RBiA during the offseason could fully explain it. It would be interesting to see a comparison on the change in horizontal pitch movement for all Ranger pitchers that were with the team both this season and last season.
"Certainly, we were penalties, and we stopped ourselves." - Les Miles
I'm not looking for errors.
I’m pointing out that PITCHf/x should never be taken at face value.
Look at Colby Lewis’s charts. His fastball has also “lost” about 2 inches horizontally while his curveball and slider have “gained” about 2 inches. You think that’s coincidence?
by NoNameOnCard on Jul 28, 2011 11:39 AM CDT up reply actions
Also
if only there were a way to compare Feliz’s average horizontal fastball movement at home vs. on the road. That would help us single out whether it’s just a problem with the PitchF/X measurements at RBPiA.
"Certainly, we were penalties, and we stopped ourselves." - Les Miles
According to spin rate and spin angle measurements...
Feliz’s CH is different from last year. Small sample size, yes, but changes in spin angle and in spin rate are going to change the pitch’s movement. It’s not very unlikely that Feliz’s changeup has more horizontal movement this season, it’s very likely.
by NoNameOnCard on Jul 28, 2011 11:42 AM CDT up reply actions
the spin rate arnd spin angle data are calculated from the position/velocity/acceleration dataset, right?
If that’s the case, wouldn’t a change in calibration affect the spin rate and angle as well?
That is correct.
I wasn’t thinking straight.
by NoNameOnCard on Jul 29, 2011 11:52 PM CDT up reply actions
Good stuff
I have some emails about gaining inches I could forward to Feliz.
Some see a glass half empty, some a glass half full. I see a glass that's twice as big as it needs to be. - George Carlin
If only those products worked.
I’d make sure to use them to gain a few inches… on my fastball movement.
"Certainly, we were penalties, and we stopped ourselves." - Les Miles
My personal feeling is he doesn't have control of his fastball
as much this year as last year. If he cannot throw his fastball as strikes, it makes it tough to drop his slider/curve into two strike situations. You mentioned his BB/9 has moved up to 4.58 and that may be a reflection of him not spotting his fastball.
Generally a cutter should move slightly away from a right handed batter, somewhat like a flat slider, so normally it shouldn’t be confused with a four seemer. However, that may vary for some pitchers. It would be interesting to know if he threw a two seamer which usually would break in more towards a right handed batter and possibly sink more.
While it is possible his fastball may not have as much movement on it this year, my feeling backed up somewhat by your walks per nine innings stat is he just doesn’t have control over his primary pitch, and that allows batters to look for a fastball when he gets behind on a batter. By the way, this could be a temporary situation in that his last two outings have not been good, but he had a really long stretch before that where he did not pitch.
Regarding your second paragrap: sure, but pitcher do not throw their pitches consistently
There is generally a very large spread in movement of pitches thrown by a pitcher over the course of a season – horizontal movement of fastballs could vary like +/- 5 inches. Say your 4-seamer has an average horizontal movement of 5 inches, and the cutter moves 1 inches in the opposite direction. Apply the +/- 5 inches of variation among pitches, there will be plenty of overlap between the movement of the cutter and the 4-seamer.
I think the more sophisticated categorization system uses the spin angle and spin rates (which combines information on the horizontal movement, vertical movement, and velocity) calculated from the pitch f/x data, but even then, there is often significant overlap.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Lone Star Ball
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
In Association With
Big League Stew - MLB - Yahoo! Sports
- Get a grip: Reds’ Todd Frazier loses bat, still connects for home run (Video)
- Eric Hosmer redefines the Baltimore chop with infield double (Video)
- College player makes impossible over the fence catch to save tournament game (Video)
- Security!: Camden Yards guard collides with Royals’ Mike Moustakas
- The Juice: Nelson Cruz matches career high eight RBIs, Pujols stays hot in Seattle
