No Soup For Ubaldo

There probably isn’t a single baseball fan in the country who hasn’t heard Ubaldo Jimenez been called “lucky.”

For several weeks now, analysts have devoted countless hours and vast amounts of energy to debunking the theory that Jimenez is—as his 13-1 record and 1.15 ERA suggest—one of the best pitchers in the history of the game. And with good reason.

There’s no question Jimenez is a talented pitcher entering the prime of what will certainly be an impressive career. But he’s not an all-time great, and he’s certainly not the greatest of all time.

Jimenez’ 7.8 K/9 rate is impressive (though not legendary—he’s looking up at not only Tim Lincecum and Josh Johnson, but guys like Javier Vazquez and Felipe Paulino), but it’s not enough for us to turn a blind eye to his wildness (3.2 BB/9). A 2.44 K/BB ratio is nothing to sneeze at, but it’s nothing compared to Dan Haren (5.05), Roy Halladay (5.63), or the superhuman Cliff Lee (16.75).

As a result, Ubaldo’s FIP is a more mortal-looking 2.93. That’s nothing to sneeze at, and it’s the seventh-best mark in the game. But it’s more than two-and-a-half times his ridiculous 1.15 ERA.

And that’s before you consider Jimenez’ ludicrously low 3.8% HR/FB rate. That’s why his 3.61 xFIP is significantly higher even than his FIP—and that’s normalized for a pitcher in a neutral park, not one who plays half his games at the launching pad that is Coors Field. Substitute his xFIP for his ERA and ignore the wins (naturally, he wouldn’t have as many if he gave up more runs) and you’ve got a questionable All-Star, not a unanimous Cy Young.

So where is all this luck coming from?

The fishiest thing about Jimenez’s season so far is his 91.2% LOB rate. In other words, fewer than one out every 11 baserunners he’s allowed have ended up crossing the plate. The discrepancy between his strand rate and the norm (72 percent) is greater than the overall range of qualified pitchers’ LOB rates in 2008.

It makes sense that a better pitcher would strand more runners; the better the pitcher, the better the chance of making an out, so there is less opportunity for the other team to score. But Jimenez’ 91.2% figure places his performance well outside the reach of logic and fully inside the realm of luck.

Consider the case of John Candelaria, whose 88.8% strand rate in 1977 stands as the closest anyone has come to pulling a Ubaldo over a full season since at least 1974. The year before that, his strand rate was 72.5%; the year after, it fell to 76.8%. Simply put, you can’t sustain a number like that unless you’re playing Xbox.

Then, of course, there is the issue of Jimenez’. BABIP. I’m a firm believer that pitchers have some degree of control over where and how hard the ball is hit. I wouldn’t think it noteworthy if Ubaldo’s hit rate had merely slipped to .290, or .280, maybe even .270. But if you think the ability to induce weak contact is the reason his hit rate stands at an historically low .239 mark, I’m going to have to stop you right there.

It takes a lot more than talent for a pitcher to sustain a hit rate that low for more than a few weeks. Since 1989, only one pitcher has posted a hit rate at or below Jimenez’ current .239 mark over a full season without it ballooning 50 points or more the following year.

Now, some say that Jimenez’ hit rate is explained by the kind of contact he’s induced—his 13.8% line-drive rate is the third-lowest in the league, and his 54.9% groundball rate ranks fifth. But there’s no refuge in that argument, either.

Looking at tRA, which (unlike FIP) takes his batted-ball profile into account, Jimenez is expected to give up 3.09 runs per nine innings. That’s not a bad number by any stretch, but it’s not good enough to put Ubaldo in the history books. So even if you assume that his low line drive and HR/FB rates are the product of sustainable skill and not felicitous chance, Jimenez could be expected to give up nearly three times as many runs if he had neutral luck.

There’s no question Ubaldo Jimenez is a good pitcher, or that his is an arm to watch for years to come. But once the winds of fortune stop blowing in from the Coors bleachers, no one will mistake him for the best pitcher in the game.

Lewie Pollis lives outside of Cleveland, Ohio, and will be starting at Brown University in Fall 2010. Like at least half the people who will read this article, his dream is to be GM of a baseball team. For more of Lewie’s writing, click here.





Lewie Pollis is a sophomore at Brown University. For more of his work, go to WahooBlues.com. He can be reached at LewsOnFirst@gmail.com. Follow him on Twitter: @LewsOnFirst or @WahooBlues.

8 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Paul
14 years ago

But his FIP and ERA gap could narrow dramatically in the coming weeks as his luck evens out, and statistically he’d look better than he is. This is really no different than Grienke last season. He became human down the stretch, but was still outstanding. Your concluding statement is just bizarre. As noted, even when his luck evens out a bit you’re still talking about a guy who is legitimately in the discussion for best starter in the game.

Mark
14 years ago

How does the fact that his LOB% is unsustainable mean it’s “luck”? Cliff Lee’s K/BB ratio is beyond unsustainable, yet we don’t dismiss that as luck.

Wouldn’t the fact that his K/BB ratio is 3.43 with RISP, vs. 2.19 with bases empty have somthing to do with his very high LOB%? Or that his K/BB ratio with men on is 3.80? His xFIP is also much better with men on (3.14) than with bases empty (4.04).

MB
14 years ago

So what you are saying is he needs to starting giving up more hits (to raise BABIP), give up MORE jacks (to raise HR/FB% and lower his xFIP) and tidy up his control.

For extra measure, to improve going forward he also must stop stranding those pesky things called….baserunners…just let them score damnit!

MB
14 years ago

and nobody will at least consider him one of the best?

A pitcher who has seemingly conquered Coors Field, armed with a 99 MPH sinker, improved control and secondary offerings, a ground-ball machine and only 26 years old?

really

Chris
14 years ago

@MB: I think you are missing the point of the article. The author is not saying Ubaldo “needs” to do any of those things to “improve”. He’s simply saying that Ubaldo’s peripherals suggest that hes pitching better than expected and we can expect to see him give up more hits and home runs through out the rest of the season (i.e.- regress back to the mean/reasonable expectations).

You could also read the article to say, if Ubaldo had given up some more home runs / hits, walked fewer, and struck out more, then we would expect that pace to continue. For an example of an actual pitcher with good looking peripherals see Roy Halladay. He should be able to maintain his current success for the rest of the season (Which has been better to date by WAR Standards).

MB
14 years ago

I realize the point, I am just saying if he starts to do the things I suggest, he will no longer be targetted here for his luck.

Anon21
14 years ago

Mark: I think pretty much every analyst who has considered the question has concluded that Lee’s ridiculous 0.42 BB/9 rate is due in substantial part to luck. So I don’t buy your argument that Ubaldo is being singled out for “criticism,” (if that’s what you want to call it) on the basis of his obvious good fortune.

Moreover your ref-playing fails to respond to the argument’s central point about that LOB%: there may never have been a pitcher, and there certainly hasn’t been one in the last few decades, who has posted a percentage that high over the course of a full season. I just don’t think it’s plausible to say Ubaldo is maybe the best pitcher EVER at stranding runners.

Jake R
14 years ago

Using K/9 understates Ubaldo’s strikeout rate as it is suppressed by his BABIP performance causing him to face fewer hitters. Ubaldo’s K% sits at 23%. Just looking at the top of the K/9 leaderboard, this compares reasonably well with several pitchers posting much higher K/9. As an example, Dan Haren has struck out over an extra batter per inning than Ubaldo. He has, however, only struck out batters at a 0.5% better clip than Ubaldo, which is much smaller difference.

On the other side, Gavin Floyd and Ian Kennedy have virtually identical K/9 as Ubaldo. However, Floyd has a 20.7% K rate and Kennedy has stuck out opposing batters at a 21.1% clip.

I am mostly writing this to continue the crusade against per nine inning component statistics for pitchers. They are an inferior tool to per batter faced statistics and the next big step in accurately measuring pitcher performance will be the move to percentage rates for component stats, thereby eliminating the statistical noise from our measurement of strike out rates, walk rates, HR rates, etc. Until this change happens, it is important to remember that all of the per 9 stats are inherently flawed. If we regress Ubaldo Jimenez’ BABIP, we also have to regress his K/9 up and his BB/9 up and his HR/9 up before we can make any sort of argument based on the new BABIP. That FIP and xFIP fail to do this significantly detracts from their usefulness as statistics.