A Baseball Weblog

Monday, January 11, 2010

Sergio Mitre, Chad Gaudin, and the competition of not being traded

With the recent acquisition of Javier Vazquez, the Yankees have added even more depth to an already rich pitching staff. This means that the pinstriped careers of Sergio Mitre and Chad Gaudin are in jeopardy. Both Mitre and Gaudin spent time as the Yankees’ fifth starter in 2009, and had varying levels of success. Outside of a one-hit performance against the White Sox in August, Mitre was knocked around. Gaudin performed admirably, but in the Yankees’ rotation, he seemed to tire as he reached the later innings. If the trade rumors are true and materialize, the pitcher that stays could conceivably win a spot out of camp as an extra bullpen pitcher / spot starter. Now, my question is: which pitcher should stay? Gaudin had better production last year, but there are obviously other things to consider: could Gaudin net a bigger package in return than Mitre? Is there any statistical evidence to lead us to believe that Mitre will pitch better in 2010? How could the Yankees save money?

The first thing I’d like to do is roll out some PITCHf/x numbers on these guys (since that’s what I like to do). I’ll start with Mitre.




SpeedMax Speedpfx_xpfx_zspin_angle
SI90.9793.9-10.513.19252.86
CH83.5185.7-10.083.63250.02
SL78.5280.98.29-2.9370.95



Pitch#Pitch%Swing%Whiff%Wide Zone%Chase%Watch%Zone Whiff%
SI61069.7%43.9%9.3%57.9%19.8%39.1%6.5%
CH14116.1%56.7%37.5%52.5%25.0%18.9%40.0%
SL12414.2%31.5%35.9%46.0%35.9%56.1%24.0%

87599.7%44.2%17.8%55.3%22.5%38.0%14.7%



SLGCONRSvRSv/100xRSvxRSv/100
SI.634-10.40-1.710.860.14
CH.4851.100.780.730.52
SL.3890.580.47-0.67-0.54

.585-8.72-1.000.920.11


Sinkers, sinkers, and more sinkers. Mitre’s overall lack of success is reflected in the poor run values, but the expected values paint a different picture. The xRSv numbers are the defense independent variant on the standard run values and are based only on the batted ball type. Sergio’s 57.8% 2009 groundball rate is spectacular, and it is in line with his career mark of 59.7%. Groundballs are good at preventing runs (since they can’t become home runs), so it’s reasonable to believe that, with some better luck, Mitre could find some more success with his sinker if he maintains a similar ground ball rate. The other thing that struck me while looking at these pitch results was the whiff rate on Mitre’s offspeed pitches. I figured that, being a sinkerballing, pitch-to-contact kind of pitcher, Mitre would have a swing-and-miss rate well below the league average, but this isn’t really the case (2009 MLB average is around 19.5%).



Above is a graph representing the pitch height of each sinker that Mitre threw in 2009. We can see that most of his sinkers were about 2 to 2.5 inches off the ground, which is in the lower half of the approximated 1.5 inch – 3.5 inch strikezone. I must say that watching many of Mitre’s outings, he seemed to be missing location side-to-side within the strikezone a lot, which is oftentimes even more dangerous than missing outside of the strikezone. He did undergo Tommy John surgery in 2008, so maybe some more rehab time would sharpen his command.
In conclusion: I don’t see a whole lot that’s bad. He got lots of groundballs, could generate whiffs with his breaking pitches, and got burned on balls in play. I’m not saying that he’s going to be the best pitcher in the league or anything, but I do think that he has something to offer.





SpeedMax Speedpfx_xpfx_zspin_angle
FF90.3294.2-4.717.34212.34
FT89.6893.4-9.463.20251.27
SL79.6184.16.882.40110.44
FS85.9089.8-7.680.86262.42



Pitch#Pitch%Swing%Whiff%Wide Zone%Chase%Watch%Zone Whiff%
FF127250.6%39.4%14.2%58.1%16.4%43.3%12.2%
FT1787.1%47.2%10.7%47.8%29.8%30.6%6.8%
SL79331.6%49.6%36.1%49.1%37.2%36.5%22.7%
FS26910.7%48.0%18.6%45.4%31.0%27.0%13.5%

251299.1%44.1%22.2%53.1%26.5%39.0%15.1%



SLGCONRSvRSv/100xRSvxRSv/100
FF.587-4.27-0.34-0.80-0.06
FT.489-2.99-1.68-0.99-0.55
SL.4579.451.197.650.96
FS.3501.610.601.330.49

.5063.800.157.190.29


Gaudin works with four pitches --- two kinds of fastballs, a slider, and a hard changeup (I’m calling it a splitter here). Gaudin only got an average amount of groundballs (43.7%, ~42% league average) but got a substantial amount of swings and misses (22.2%) – despite only averaging 90.31 mph on his four-seamers. The whiffs contributed to his 8.5 K/9. His slider proved to be a valuable pitch for him, saving 9.45 runs over the course of the year and causing batters to swing and miss 36.1% of the time.

I noted earlier that Gaudin seemed to have trouble after reaching a pitch limit. I was recalling three September starts against Toronto, Tampa Bay, and Los Angeles. To me, it seemed as if his move to the bullpen in the middle of the year disturbed his stamina for his return to the rotation at the end of the year. I decided to investigate his fatigue with a velocity chart for his starts as a Yankee, and what I found was inconclusive:



I do suspect that whatever the cause of the problems Gaudin ran into at the end of last year was, it is fixable with an increased amount of time in a starting rotation.

So, which one would you keep? If you base your argument on 2009 performance, the decision is pretty obvious. For some extra fun, I’d like to look at their respective 2009 seasons in terms of ERA and my three favorite ERA predictors --- tERA, FIP, and xFIP.


ERAtERAFIPxFIP
Sergio Mitre6.795.385.304.00
Chad Gaudin4.644.154.164.36

Gaudin is better in all categories except for xFIP. This is because xFIP normalizes home run rates, while tERA and FIP do not. Mitre’s HR/FB rate was an absurd 21.7%, significantly above the league average of around 11%. I’m thinking this would be an aberration that would likely regress in the future, but HITf/x would be particularly handy here to see how hard Mitre was hit.

Last year, I never thought that I would arrive at this conclusion, but I actually think the Yankees should keep Mitre and trade Gaudin. Considering Mitre’s extreme groundball tendencies, his abnormally (and quite possibly fluky) high home run rate, and his extended TJ recovery time, I think he’s a good comeback candidate for 2010. Gaudin may very well be more appealing to clubs since he is younger (Gaudin will be 27, Mitre 29), had a better 2009, and isn’t coming back from a major surgery. Also, by trading Gaudin, the Yankees would save money by paying Mitre his $0.85 million contract for 2010 instead of paying Gaudin, who already made $2 million last year. And we also have to remember that the pitcher who stays would probably only be filling a “sixth starter” role, due to the depth of the Yankees’ staff. I may feel differently about this in relatively short order depending on what actually happens in 2010, but for now, I'm leaning towards keeping Mitre.

Gameday PITCHf/x data is from MLB Advanced Media; it can be easily accessed via this tool. Other statistics are from Fangraphs and Baseball-Reference.



5 comments:

  1. Is there a statistic which differentiates hard-hit ground balls from routine or slowly-hit ground balls? Regarding Mitre, if there are a lot of hard-hit ground balls, it might not be a good thing.

    ReplyDelete
  2. Unfortunately, no. I do agree completely that it would be very important to see how hard the ball is being hit. HITf/x is getting worked out behind the scenes, and we should see some batted ball data (elevation angle, speed off bat, travel distance for each ball in play) before long. For now, all we have are ground balls and line drives. Mitre's line drive rate in 2009 was 18.2%, near the MLB average of about 19%.

    ReplyDelete
  3. Could you post a link to resources about Pitchf/x please. I'm interested in their methodology - actually how they obtain their results.
    There's a big difference between a hard hit grounder that will get through the infield unless it's hit right at someone and a Mariano broken bat dribbler.

    ReplyDelete
  4. http://mlb.mlb.com/news/article.jsp?ymd=20071002&content_id=2245402&vkey=news_mlb&fext=.jsp&c_id=mlb
    This is a pretty old article, from October 2007, but it gives a pretty good idea of how the data is picked up. If I find any other good primers, I'll post them.
    I know less about HITf/x, but I'm pretty sure that it's supposed to be equipped with the same tools as PITCHf/x and would track the outgoing pitch as opposed to the incoming pitch. It should explain a lot of things about "how hard" certain pitches are hit, and until it is more easily accessible, there are going to be holes in these sorts of analyses.

    ReplyDelete
  5. Continued -
    For now, we can only make inferences based on the data we have. If I noticed that there were an unusual amount of pitches high in the strikezone for Mitre, we would have something to go off of. I didn't find that this was the case. Also, his sinker had a lot of spin movement, which means that it wasn't "flat." Not that these two things alone make a good pitcher, obviously, and there are countless other ways to analyze a pitcher. Though I do think that those two things are important.

    ReplyDelete