Giants Squandering Yusmeiro Petit’s Potential

By: Ben Leonard

Postseason hero Yusmeiro Petit was overlooked in the playoffs, not starting a single game. In spite of this, he still shined.  Being overlooked has been the only constant in a tumultuous career for Petit, bouncing around from the Marlins to the Diamondbacks, Mariners, and Mexico before settling in with San Francisco. He found his place as a long reliever for the Giants in their title run, posting a 1.42 ERA in 12.2 innings while striking out thirteen. The right-hander is talented and cheap, yet the Giants have penciled in Tim Lincecum as their fifth starter over him. Bochy and the Giants naively believe that Lincecum can turn it around as a starter after four straight years of decline, partly because of his salary, and partly because of his two Cy Young awards.

Quietly Dominant

Petit quietly established himself as one of the best right-handed pitchers in the game in 2014, building upon his solid 2013 campaign. For some context, here is a blind comparison with one of the top right-handers still on the free agent market.








A 10.23 1.69 9.1% 3.69 2.78 2.84
B 7.14 1.74 9.7% 3.21 3.59 3.56

As you may have guessed, Player A is Petit and Player B is James Shields. Petit had the seventh-highest strikeout rate in the big leagues, while Shields’ 7.14 K/9 pales in comparison to his extraordinary clip. Shields is commanding a five-year deal in the $100 million dollar range, while Petit only has a 1-year, $1 million dollar deal. Petit has much better peripheral statistics, is three years younger, but makes essentially $20 million less per year. Petit outclassed Shields in 2014 in every way except name value and established history.

Unlucky as a Starter

Many fans may point to Petit’s ugly splits between starter and reliever as reason to lose faith in the righty. He did indeed post an ugly 5.03 ERA as a starter, and a measly 1.84 ERA as a reliever. However, in such a small sample size, it can be easy to get caught up in standard statistics and draw flawed conclusions.

Petit’s rapid decline was largely due to a decrease in luck, not skills as a starting pitcher. Opposing hitters actually hit line drives more frequently against Petit as a reliever (22.4% vs. 20.2%). However, Petit was a victim of an inflated 13.1% HR/FB rate, compared to an unsustainable 2.1% as a reliever.Even his infield hit percentage spiked as a starter, sitting at 7.6% compared to 4.8%.  Petit threw the ball seemingly just as well, if not better, as a starter, but got nothing to show for it, shown by his 3.01 xFIP. Balls simply were placed better with him as a starter, in part fueled by an inflated .309 BABIP (vs. .261).

The Giants should expect regression to the average from Petit next year as a potential starter, somewhere right in between his 2014 splits. Overall, he had a 2.84 xFIP last season, a somewhat realisitic figure for next season. Petit has always slightly underperformed his fielding-independent peripherals, but not enough to believe it will deviate too much from the 3.00 range. Even a number in the 3.50 range will be more than enough for the Giants out of the fifth spot of the rotation, as it is far superior to anything Lincecum and Vogelsong have achieved in recent years.


Unlike his apparent superior Lincecum, Petit is a pitcher, not a thrower. He has demonstrated an ability to succeed without velocity. His average fastball last season travelled at just 88.7 MPH, good for 155th among pitchers who threw at least 150 innings. Petit has exceptional command and keeps the ball down very well, not relying on sheer arm strength to deceive hitters. This attribute translates well over time for the righty, who should be able to avoid the sharp decline that Lincecum suffered as his velocity declined.

His devastating changeup caused many swings and misses last season, another reason for optimism. Changeups are less susceptible to decline over time, as grip is more essential to their movement, not the snap of a wrist. It is more of a feel pitch, one that does not much rely on brute strength.

For these reasons, Petit can be considered a relatively long-term solution on the hill for the Giants. He is under team control until 2017, and will remain very cheap until then. The Giants have three more years in Petit’s prime, and need to take advantage of it while they can. Lincecum has excelled in a long-ish relief role in the past, especially in the 2012 playoffs. He was effective even as recently as the 2014 World Series, throwing 1.2 perfect innings. Lincecum can fill in for Petit admirably, but has no reason for to start games for the Giants other than his pricey contract. The Giants will pay Lincecum regardless; therefore, there is no reason not to field the best staff possible and start Petit.

Stats and info courtesy of FanGraphs and Baseball-Reference

Cover Image: By Lisa Suender on Flickr (Original version) UCinternational (Crop) [CC BY-SA 2.0 (], via Wikimedia Commons

This article has 3 Comments

  1. And, not to be a jerk, but your Player A/Player B comparison is a bit of a cheat. Petit posted those numbers in 39 games, 12 of which were starts, resulting in 117 combined IP. Shields pitched 34 games, all starts, and finished with 227 IP. So Shields, who averages 6.2 IP/GS, has a harder job, since he generally goes through the order close to 3 times. Petit, in contrasts, enjoys the advantage of facing each hitter only one time on average.

    As far as their 2014 stats are concerned, this make a significant difference. Last season, Petit’s held batters to a .175/.204/.272 slash line the first time through the order. The second time, those go up to .277/.299/.515. The third time through: .370/.414/.741. That is extremely alarming for a starter.

    Shields numbers show that he’s a far more effective starter: 1st time: .248/.285/.345; 2nd: .267/.318/.466; 3rd: .244/.277/.413. In short, Shields can and does go deep into games (by contemporary standards), but Petit appears limited as a starter because he becomes significantly less effective each time through. So Petit didn’t outclass Shields; he arguably performed better at a different, less demanding job. (Petit’s own splits bear this out, of course.)

    It’s easier to see the problem with the comparison if you take a more obviously inapt comparison using the same metrics (which are a bit cherry-picked, you must admit) you do for Petit and Shields.

    Player A: K/9: 10.31; BB/9: 3.8; HR/FB: 6.3 %; ERA 2.85; FIP: 2.83; xFIP: 3.14
    Player B: K/9: 10.29; BB/9:2.57; HR/FB: 7.5%; ERA: 3.15; FIP: 2.85; xFIP: 3.12.

    Again, I’ve used an identical comparison standard, but Fernando Rodney (A) is not nearly the pitcher that Max Scherzer (B) is, and Rodney certainly did not “outclass [(say) Sonny Gray] in everything but name value and established history”. (By the way, what’s wrong with established history?)

    I like what you’re doing here, man, and I wish you the best, but some of analysis is a little loose. I’m nobody, so take it for what it’s worth, but I think a bit more attention to detail and argument structure would go a long way.

  2. I tend to agree that Petit is a better potential option than Lincecum as a starter, but your analysis of Petit’s history as a starter is incomplete, and the omissions are significant in my view.

    You nicely point out that his struggles as a starter last year came as the result of some bad luck in a small sample size. However, there’s a lot more data on Petit as a starter than you report. Petit was a starter throughout the minor leagues, and he has a somewhat pedestrian 3.68 ERA overall, starting 170 of his 189 MiLB games. His ERA at AAA is 4.34 (108 of his 127 appearances were starts). In his 15 AAA starts at Fresno in 2013, he posted a 4.52 ERA. His peripherals were not much better — lower K%, higher H%, for example — and his career WHIP at AAA is a respectable but unspectacular 1.26.

    More importantly, he has 56 career starts in the Major Leagues, resulting in 297 IP. Here are some of his cumulative numbers: 4.94 ERA, 1.303 WHIP, 7.7 K/9. So that’s a decent sample size, and those are mediocre numbers. They are also very close to Lincecum’s overall numbers last season: 4.64 ERA, 1.394 WHIP, 7.7 K/9. In addition, Petit has been significantly better in 102 MLB relief innings: 4.24 ERA, 1.23 WHIP, 9.2 K/9.

    It’s also helpful, I believe, to remember that although it seems like Petit came out of nowhere, he’s been at it for a while, and he was once a highly valued prospect. In 2005 and 2006, he was the 46th and 69th rated prospects in all of baseball. So, after starting as an elite prospect, he pitched his way to journeyman status. In addition, you refer to the Giants having three more years in Petit’s prime, which is very cool, but he’s already 30, and it’s a pretty big stretch to call years 30, 31, and 32 part of the prime. Data suggests most players peak before their 30s.

    I’m optimistic about Petit — some guys do peak late — but there’s plenty of evidence that should make one dubious of his ability to become a quality Major League starter this late in his career. It definitely would be anomalous.

    Lincecum is, of course, very concerning, but not without hope. Although I concede it’s not much to go on, it merits mention that through July 20, 2014, Lincecum’s numbers looked like this: 20 GS, 120 IN, 9-6 (team record 13-7), 3.68 ERA, .229 BAA, 108 K, 1.25 WHIP. That’s very close to 2/3 of the season, and those are respectable 5th starter numbers. He tanked from July 25 on, of course, so that’s obviously scary, but this (admittedly arbitrary) split suggests that he might not be a lost cause as a starter.

    In sum, I think the Giants probably have a bit of time to play with the rotation-bullpen assignments and choose among Lincecum, Petit, and Vogelsong for that 5th spot. I’m just not sure there’s a clear way to predict who will be the best out of that group.

Comments are closed.