Should the Nationals Sign Free Agent Ubaldo Jimenez?

There's been plenty of talk around the Natmosphere about whether the Nationals will sign or trade for a starting pitcher. Writers have recently floated the idea that Ubaldo Jimenez would be a good fit, but I don't think he would necessarily exceed the production that the team would get from in-house options from two of Ross Detwiler, Tanner Roark, and Taylor Jordan.

Recent Industry Thoughts

David Schoenfield of ESPN wrote Saturday that the Nationals should sign Jimenez. Schoenfield stated specifically that:

Jimenez is the one who can provide the most upside [among Matt Garza and Ervin Santana] and probably comes in a little less expensive. Plus he has a rubber arm, having made more than 30 starts six seasons in a row, one of just 13 starters to have done that.

Other Nats-centric publications were in agreement with the idea. Adam Kilgore of The Washington Post didn't address Jimenez by name, but it's easy to see how the 29-year-old would fit in given his reply in a recent reader Q&A:

I’m not suggesting the Nationals sign [other lower-tier starters] but they are average to below average options to consider if they nothing else pans out and they don’t feel completely comfortable with Tanner Roark or Taylor Jordan in the rotation….The Nationals have the window and money to win now, so maybe they should make a very strong push for free agent options like [Bartolo] Colon and [A.J.] Burnett, enough to entice them to come to Washington.

There's absolutely value in having a major-league pitcher toe the rubber every fifth day. The problem is, Jimenez wasn't exactly the guy you'd want to run out there that frequently in 2011, and especially 2012. But to Kilgore's point, the Nationals -- even absent a proven #4 or #5 starter (whatever that means) -- can win in 2014. Is Jimenez the guy?

Bad Ubaldo

After a great 2010, the righty's 2011 was a bit of a letdown, and 2012 was terrible. In fact, Jimenez was one of the worst starters in baseball in '12, pitching almost 180 innings of replacement-level baseball. Like ERA? It was 5.40.  Like FIP? 5.06. Think he was a victim of a wonky HR/FB rate? xFIP: 4.98. Even the red-headed stepchild of pitching statistics, losses, gave him no love (he had 17 of those L's, and I'm not referring to loves).

Fangraphs' Mike Podhorzer described Jimenez's 2011 and 2012 well. He wrote earlier this year that:

...something happened in 2011. His fastball velocity dropped nearly two miles per hour and his effectiveness declined along with it. He was then shipped off to the Indians to finish off a disappointing season. The following year, his velocity dipped another mile and a half per hour and he posted an ERA above 5.00. The downward spiral continued into 2013, as his velocity dropped even further and he allowed 19 runs over his first 4 starts.

Yahoo!'s Jeff Passan offered some humanity to Jimenez's struggles, writing that the hurler would at his lowest points stand in front of mirrors staring at himself, as if he would through force of concentration or introspection wish success to happen. It was of no consequence: Major league hitters are not a compassionate bunch.

Good Ubaldo

Jimenez turned it around in 2013, and the baseball writing community has already done well to analyze why. For example, Alex Skillin of Beyond the Boxscore succinctly demonstrated that it was really a tale of two halves for the Cleveland twirler (all graph effort is his):

2013	ERA	FIP	wOBA Against	HR/9	K%	BB%
First Half	4.56	4.50	.331	1.19	21.7%	12.2%
Second Half	1.83	2.61	.282	0.42	26.7%	8.7%

Jimenez improved from 2012, to be sure, but Skillin noted via Doug Thorburn's article at Baseball Prospectus that Jimenez's release point this season -- even during the second half of the year -- was far less consistent than it was during his dominant 2010 campaign. And it was true.

But Passan's article offers some insight into why. The Dominican native worked with Cleveland pitching coach Mickey Callaway extensively both during the off season and while the 2013 season was underway. Callaway got Jimenez back to fundamentals by re-working his timing, delivery, and posture. "This is where," Passan writes, "Jimenez practices in front of the mirror these days." Perhaps it was Jimenez adjusting to his new delivery, and succeeding despite an as-yet imperfect release. It is understandable to think of what a refined, version 2.0 Ubaldo would be capable of in 2014.

This past season Jimenez put up +3.2 fWAR and +2.7 bWAR. More simply, he was durable, saw his strikeout rate rise to a career-high (25.0%), and wasn't burned by the long ball. If it was him instead of Dan Haren, D.C. could well have made the postseason. No doubt, talk of a possible acquisition is merited, and I'd wager Mike Rizzo has thought enough to take a look.

So Why Not?

Projections are not designed to be exact. Keeping this in mind, the Steamer forecast system pegs Jimenez for a +2.7 fWAR 2014, with a 3.64 FIP. Pretty solid, to be sure. And depending on your views on salary, feel free to look at MLBTR's projection of 3 years and $39 million as a baseline for what he will earn this offseason.

But let's leave money and draft pick compensation aside for now (of course, Rizzo has stated that he would prefer not to give another first rounder away this year) and focus on just how Jimenez did it this past year.

Schoenfield thought that nothing in Jimenez's profile screamed fluke, noting that his BABIP was over .300 and that "Jimenez simply threw more strikes than he has in years and let his natural movement take over."

Here is how Jimenez accumulated his strikes in 2013, courtesy of Baseball Reference:

Schoenfield is right on this score. Jimenez's strike percentage was 3% better than 2012, and 2% higher than his excellent 2010. Still, it was below the major league average. Of course, a strike may either be looking, or as a result of a swing. Jimenez was above his normal percentage for looking strikes, but actually 2% below his career average in getting "swung at" strikes ("AS/Str" -- inplay, fouls, and whiffs).

Things get interesting in the three right-most columns. 1st% tells us that Jimenez worked ahead more often than he usually does, got to 0-2 counts a lot more often (02%), and generated an excellent number of looking strikeouts (L/SO%). The latter two numbers are particularly interesting because of their upward jump from both Jimenez's averages and MLB averages.

The rub on Ubaldo has been that he can't consistently throw strikes. Yet this past season, he managed to get to 0-2 counts a whopping 217 times -- second only to his 2010 value of 219 0-2 counts, which, it should be noted, was achieved over 40 more innings pitched.

So, Jimenez got to 0-2 counts more frequently in 2013 than he ever has before. And after 0-2, batters showed a paltry .365 OPS. This value makes Jimenez 35% better than the league average pitcher on results after 0-2. Get to a pitcher's count early, earn favorable outcomes. This was a big part of his success.

Time to view this another way. Here's a more visually-pleasing comparison (lower is better for sOPS+):

Basically, Jimenez added 43 0-2 at bats from 2012 in 2013. That's 14 1/3 more innings of hitters in a position that statistically is the worst hitting circumstance in baseball, excepting three strikes (of course). Anyone would take that.

What does this tell us? While Jimenez did well to throw a few more strikes, he did really well to sequence those strikes to his advantage. And results after 0-2 worked out great -- Jimenez ranked 27th out of 107 qualified pitchers by sOPS+ on outcomes after 0-2. Think about it from a hitter's perspective.  If you're facing vintage, wild Jimenez, going 0-2 was not a concern. It was in 2013.

It would be a great story to conclude that Jimenez leveraged his refined approach into improved performance, especially with respect to 0-2 frequency and results. Actually, he probably did. But the performance part is a tricky concept.

Here's where Jimenez located his pitches with two strikes in 2012 (left) and 2013 (right):

Here begins the cautionary tale.  It looks like his location was pretty much the same on two strike counts in both 2012 and 2013, even though his results were not.

Over 1400 words later, here's the conclusion: In 2013, Jimenez benefited from a depressed BABIP on two-strike contact. For example, while he rocked a .314 BABIP after 0-2 counts in 2012, he saw a .234 BABIP after 0-2 counts in 2013. See for yourself (2012 again on the left, 2013 on the right):

In case you are wondering, his in-zone BABIP in 2013 (.290) was lower than last season (.330), too. And he actually gave up more line drives in the zone this past season on two strike counts -- 21.7% -- than 2012 (20.5%).

In sum, 2013 Jimenez saw more hard contact, but better batted ball outcomes on two strike counts. These results account for 28% of the batters he faced, a significant portion.

Conclusions

Great results on the back of good fortune for almost 30% of the hitters you battle is awesome when you've banked these results, not when you're going to the Bank of Lerner expecting the same in 2014. To be sure, Jimenez has done this sort of thing before. The real, forward-looking expectation should probably be somewhere in the middle.

And this hits the underlying question: if the Steamer projection is that middle ground, then is it true that Roark and Jordan are meaningfully inferior? Reasonable arguments can be made both ways; combined for 200 IP, their projections aren't horrifyingly worse (Roark here, Jordan here), although they are not quite as good. There's also money, the draft pick, and the chances Jimenez, away from Callaway's tutelage, again gets away from what makes him comfortable.

Either way, all Nats fans can agree that debating the back end of the rotation at Thanksgiving is better than worrying who will fill in behind John Lannan in the coming year. Enjoy the holidays.

• Note: All statistics courtesy of Baseball Reference, Brooks Baseball, and Fangraphs.