A look at Ubaldo Jimenezby Josh Weinstock
July 20, 2011
Ubaldo Jimenez is the straw that stirs the drink. Or at least he has been in the past few days, being the subject of frequent discussion. The Rockies are apparently considering trading Jimenez, with the Yankees being named as the most likely suitor. Should the Rockies deal Jimenez, they would be losing a very effective and very unique pitcher. In the past four years, Jimenez has never been worse than 17 percent above average in terms of FIP. Although his performance is certainly notable on its own merits, Jimenez gets more attention for his unusual pitching repertoire. And rightly so.
Jimenez legitimately throws six pitches, maybe even seven. On an given day he throws a four-seam, two-seam, changeup, slider, curveball, and splitter. There are even reports of him mixing in a cutter. Combined with a bizarre pitching motion and elite velocity, he is somewhat of a pitching spectacle. You can examine his 2011 repertoire graphically below:
It's important to keep in mind that this is movement compared to a ball thrown without spin. The circles are centered at the respective means of each pitch. All classifications were done manually in the interest of accuracy. Gameday classifications had a lot of trouble with splitters; there were many situations where four-seams in 0-0 counts were classified as splitters, and other silly things like that. There was also one game of his in my database with terrible data, and you can see that in the small cluster of fastballs around (-1,-5). I did not classify any cutters as I was not comfortable with my ability to identify the pitch.
As you can see, Jimenez has quite the spread of pitches. This is what his repertoire looks like to a left-handed batter:
I had to eyeball the perspective, which is why it looks a little strange. I also apologize about not having the same colors in each of these two graphs, but that's an artifact of using two different plotting packages.
Ubaldo's rollercoaster season
Fresh off a career best 6.3 fWAR season in 2010, Jimenez looked like a different pitcher in the first two months of the 2011 season. With a few miles per hour missing and his command wavering, he predictably struggled, posting repuslive ERA's and merely unattractive xFIPs. In months since he's been much better, posting excellent marks in both traditional and advanced pitching metrics. Particularly of note is the huge drop in his walk rate, implying that his command has improved.
Ostensibly inconsistent, Jimenez has drawn comparisons to A.J. Burnett, which Dave Cameron addressed at Fangraphs. The comparison is kind of cheap. People see a pitcher with really good stuff struggle, and up come the accusations of mental instability and inconsistency. My personal theory is that this can be explained with a psychological principle called cognitive dissonance. The basic idea is having two conflicting thoughts in the mind at the same time causes an uncomfortable mental tension, which we then deal with through rationalizations and justifications in an attempt to re-align reality with our mental attitude.
In the case of Jimenez and Burnett, the two conflicting thoughts are clear. These are pitchers with nasty stuff, electric stuff—stuff that scouts salivate over. Then they struggle. They struggle, but they still have amazing stuff. But that doesn't make sense of course, pitchers with good stuff pitch well right? Of course they do, that's backed up by more than a century of baseball!
In this manner the mental tension is clear. This is then dealt with rationalizations like mental instability and inconsistency that are only applied to pitchers with good stuff. When pitchers without good stuff struggle, it's rarely a matter of questioning consistency or mental fortitude.
But I digress. Jimenez started the season very badly, and now is pitching very well. Often times changes in performance in small samples are dismissed due to the assumption that much of player performance is random. While I generally agree with such a doctrine, Jimenez provides an example where he has clearly changed his approach. Here is the difference in his pitch selection to open at-bats between the two halfs of his season:
As you can see, in 0-0 counts Jimenez has begun to pitch backwards more often, throwing four-seams about 11 percent less, with a corresponding increase in breaking balls. It seems that with a diminished fastball and unsatisfactory results, Ubaldo went to the soft stuff more often to open at-bats. It's also very possible that this is simply correlation and not causation, and that there is a different reason for his performance change. Indeed, this is not a trend that we see in all counts:
*Graph was made using code provided by James Keirstead in this R-bloggers post.
FF= four-seam, FT = two-seam, CH = changeup, FS = splitter, CU = curveball, SL = slider. This is a slopegraph displaying the usage of each pitch in each month. As you can see, overall his fastball usage was pretty consistent. I also would not pay attention to the increase in FT percentage, as that may be an artifact of my classifications as I did not feel very good about distinguishing between FF and FT.
References and Resources
*PITCHf/x data from MLBAM through Darrel Zimmerman's pbp2 database and scripts by Joseph Adler/Mike Fast/Darrel Zimmerman.
*James Keirstead's code on Github.
You can read more of Josh's work at FanGraphs, Beyond the Box Score, and itsaboutthemoney.net. Josh welcomes discussion through email and twitter. You can reach him at josh82093 at gmail dot com and on Twitter @J__Stock (two underscores).