Briefly Considered: Ubaldo Jimenez at Different Velocities

Right-hander Ubaldo Jimenez produced his third consecutive poor start for the Baltimore Orioles on Sunday, recording just three strikeouts against 27 opposing batters while posting a single-game xFIP of 5.23 (box). It was also the third consecutive start for his new club in which Jimenez produced an average fastball velocity below his 2013 average of 92.1 mph (or 91.9 mph average when accounting for all pitches classified as either a four- or two-seam fastball).

It’s both (a) largely unimaginative and also (b) not entirely inappropriate to begin any investigation into a pitcher’s poor performances with an investigation into velocity trends. A hastily performed study by the present author, for example, demonstrates that — amongst a large pool of starting pitchers from 2002 to -11 — that every mile-per-hour was worth about a tenth of a run of xFIP. Such a study, of course, accounts mostly for pitchers who have already adapted to their own particular velocity limits. One assumes that any one individual pitcher, when compelled to work with reduced velocity, would demonstrate an even greater increase in xFIP.

In the case of Ubaldo Jimenez, one would assume correctly. Below is a table which accounts for Ubaldo Jimenez’s defense-independent figures, as bucketed by average fastball velocity (in this case, the average of his four- and two-seam fastballs), from all 98 appearances he’s made since the beginning of the 2011 season.

One can measure it a number of ways, but generally speaking, every mile-per-hour for Jimenez since 2011 has added the equivalent of about 2% to his strikeout rate, about -1% to his walk rate, about 4.5% to his ground-ball rate, and has subtracted about 0.40 to 0.70 points from his xFIP. Beneficial, is what added velocity has been.

Of course, this isn’t to say that pitchers, including Ubaldo Jimenez, can’t adapt to reduced velocity. Indeed, such a thing is the reality for almost all of them who remain the league. Still, Jimenez appears particularly sensitive to velocity changes. During his fantastic September last year, for example, when he recorded a 2.21 xFIP over 41.1 innings, Jimenez averaged ca. 92.6 mph combined on his fastballs. During his worst month (July), during which he recorded a cumulative 4.97 xFIP, Jimenez averaged only 91.6 mph on his fastballs combined (i.e. about 1 mph less).

Does selecting the results from Jimenez’s best and worth months from 2013 represent a case of cherrypicking? Most probably, yes. Are there other variables which ought to be considered en route to producing an exhaustive study of the differences between Jimenez’s best and worst performances? Assuredly. What the data appear to suggest at the moment, however, is that the version of Ubaldo Jimenez who’s throwing his fastballs at a combined 91.5 mph or lower isn’t a particularly effective one. Meanwhile, when the present iteration of Jimenez is sitting at 92 mph or higher, that probably bodes well for his results.