keyboard_arrow_uptop

It’s become an annual tradition on the South Side of Chicago for someone to inevitably take issue with the PECOTA projections for the Chicago White Sox. I’ll admit, this year I can see where those people are coming from. With all the improvements the team has made—overhauling one of the worst bullpens I can remember watching on a regular basis; bolstering both the rotation and lineup with proven commodities—it’s a little odd to see a projection of 78 wins, just five more than last season’s disappointing total. However, it certainly isn’t the first time PECOTA has had Chicagoans scratching their heads when it came to the White Sox.

What stands out in my memory is the pre-2007 projection of the Sox to finish with 73 wins, prompting then-general manager Kenny Williams to say this:

"That's a good sign for us because usually they're wrong about everything regarding our dealings."

Of course, the White Sox would surprise many—except for PECOTA—by following up 99- and 90-win seasons with 72 wins. Let’s ignore that Williams said "they’re" wrong, suggesting either that he believes PECOTA is a group of people or that the ever-changing group of writers at Baseball Prospectus have never written accurately about the South Siders. Despite having a sort of warped view of what PECOTA is and does, he was actually on to something.

Year

PECOTA

Actual

Difference

2004

79

83

+4

2005

80

99

+19

2006

82

90

+8

2007

73

72

-1

2008

77

89

+12

2009

73

79

+6

2010

80

88

+8

2011

82

79

-3

2012

78

88

+10

2013

76

63

-13

2014

75

73

-2

2015

78

TBD

TBD

There’s a lot going on there and plenty that could be parsed out, but the bottom line is, the White Sox do manage to outperform their PECOTA projections more often than not—although three of the last four seasons have bucked that trend. Current White Sox general manager Rick Hahn seems to have much more respect for PECOTA and a better understanding of what it’s trying to accomplish, but that doesn’t change the fact that once again, he believes his team is being underestimated by our projection system. In the not-so-distant past, a study suggested that PECOTA was consistently low on the White Sox because it was hard to account for how strong of a job they do at keeping their pitchers healthy. Consequently, it consistently underestimated White Sox pitchers' inning totals, something Hahn has pointed to as well.

Once again, it’s the pitching projections that point to 78 wins for the White Sox in 2015. In particular, it’s the two starters behind Cy Young candidate Chris Sale: Jeff Samardzija and Jose Quintana, who are projected to combine for less than two and a half wins of value. In both cases, PECOTA is using data from seasons years ago that might not properly represent just how different each pitcher has become. It’s not a glitch in PECOTA, it’s just a reality; more often than not, using more data is the right way to go, but with Samardzija and Quintana it might cloud our view of what they’ve become.

From Hahn according to Dan Hayes:

"I believe their projections are a little light (on Samardzija) given what we have seen from him and what I think is reasonable to expect going forward. There certainly is a potential for variance in terms of likely results of what is going to happen."

Samardzija is an interesting case, and PECOTA being low on him (0.8 WARP) originally came as quite a surprise to me, but then I looked back at his history and remembered PECOTA’s long memory.

Many may have forgotten, but Samardzija’s story is a fascinating one. It wasn’t too long ago that the Notre Dame product was a below-replacement level reliever. Heading into the spring of 2011, he was on the verge of being released from a Cubs team that was about to lose 91 games. Things were bad for Samardzija, with one Cubs blog stating, “it’s hard to imagine another team wanting to use a 25-man roster spot on him,” and another headline reading “Why is Jeff Samardzija in the majors?”

But none of that was hyperbole. Samardzija was a poor reliever with a big name; he’d shown promise at one point, but it seemed almost impossible that he’d ever live up to those dreams. However, the big righty surprised many by having a solid 2011 season out of the pen, seemingly salvaging his career. But Samardzija wasn’t satisfied with a role as a reliever, and told the newly hired Theo Epstein-led regime that he wanted to be a starter, a decision many, including myself, found curious at the time.

I remember talking to scouts at the time and hearing that it was a bad move to try to convert Samardzija to a starter once again. He’d been yo-yoed between roles throughout his career—some blamed that for his career starting slowly—and had finally found success as a reliever, so why mess with a good thing? But Samardzija quickly silenced doubters during spring training with a much crisper repertoire: The mid-90s fastball was complemented not only with a solid slider, but a newfound splitter that many scouts still point to as his bread and butter.

What comes up most frequently when talking to scouts about what changed with Samardzija is his consistency, particularly his command. That was once a major issue for him, but his command now gets above-average grades (last season he posted a walk rate just a tick below 5 percent, easily the best of his career) and has a wide arsenal at his disposal, utilizing all five of his pitches at least 10 percent of the time.

Samardzija tossed 175 innings in his first full season as a starter, and has passed 200 each of the past two seasons. Add that up with his picture-perfect mechanics and clean health record, and it’s easy to envision he’ll blow by PECOTA’s projected 186 innings. If he can continue to limit the free passes and keep the ball on the ground (something he’d worked on with Cubs pitching coach Chris Bosio and has been improving on steadily over the years, peaking at a 50 percent rate last summer), it wouldn’t be a surprise to see him once again perform at an all-star level (or better) in 2015, combining with Sale to form one of the best one-two punches in baseball.

Like Samardzija, Quintana surprised many with his rise to success, but that’s about where the similarities end. Quintana was never heard from much as a prospect, released by the New York Mets midway through the 2007 season, then again by the New York Yankees following the 2011 season. The White Sox picked him up as more of a depth add, and while they certainly saw potential in him, what he’s become could hardly have been expected.

As one scouting director put it, Quintana was seen as more of a "kitchen-sink" guy when he was in the minors, never really wowing scouts with any one part of his arsenal. And while he still can be considered that type of pitcher, with no one pitch really standing out as dominant, everything he uses appears to have ticked up a grade, giving him multiple real weapons. The southpaw now throws a low-90s fastball, his curve and change are average to above, and he’ll even mix in a cutter (something White Sox pitching coach Don Cooper is known to teach his pitchers at a high level) on occasion—though use of that pitch dropped dramatically in his particularly strong 2014 season. With a wide variety of pitches to keep the batter guessing, combined with the natural deception in his delivery and his ability to throw each pitch where he wants, one suddenly sees a pitcher who was expected to be no more than rotation depth catapult to one of the better mid-rotation options in baseball.

Last summer was certainly a breakout for Quintana, going from solid rotation piece to borderline all-star, as he set career highs in both strikeout and walk rate while also doing a remarkable job of keeping the ball in the park. So after a career season, it’s not too jarring to see Quintana projected to be worth just 1.6 WARP with walk and strikeout rates right around his career averages. It’s rare for PECOTA to jump on a bandwagon after just one big season. However, Quintana’s projected 3.77 ERA in a depressed offensive environment would be a career-high mark, which suggests that PECOTA has doubts about the peripherals he has posted over his major-league career.

So no, White Sox fans, PECOTA doesn’t hate your team (and same with you, Orioles and Angels fans). PECOTA has some very sound reasoning behind its projected 78 wins for the White Sox, but it’s very easy, with just better-than-projected performances from Samardzija and Quintana, to believe they can outperform these expectations. Add in that this team seemingly has improved at nearly every position, in some cases drastically, and a five-win improvement certainly seems conservative. But that’s PECOTA’s job, to be conservative and, in general, more realistic than the rest of us. But sometimes we can see things that PECOTA can’t. That might once again turn out to be the case in Chicago.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
OrioleDog
2/18
The PFM forecast of 127 IP of 5.32 ERA from Noesi and just 34 innings from Rodon is systematically underrating them too.

I am curious to hear other White Sox fans take, but see intent to use Rodon early in Hahn's choice to so conspicuously leave the #5 slot terrible.
jfranco77
2/18
I think a 5.32 ERA is way too high for that spot in the rotation. If Noesi is really that bad he'll be replaced once the Super 2 deadline passes. If he's not then maybe they take it slower with Rodon.
sahadev22
2/18
It may seem like a joke that they're starting Noesi, but go read some of Coop's quotes on him, he legit believes they fixed something with the guy and he'll be more than serviceable. I think the ideal scenario (outside of all the pitchers doing well) is Noesi doing well and the White Sox finding a taker for Danks. That seems unrealistic though, moving Danks, that is. Not sure how they do it, but replacing Noesi with Rodon would put four lefties in the rotation, certainly not what they'd prefer. And this org is known for starting talented arms in the pen, so I wouldn't be shocked if they just kept it as is (if it's passable) and eventually bring Rodon up to solidify the pen.
Chickychickyfry
2/18
"I wouldn't be shocked if they just kept it as is (if it's passable)"

Either way, I think the point is that's it's pretty reasonable to assume they'll get better than replacement level out of the #5 spot. They'll either keep Noesi and Danks in the rotation if they're pitching well, or they'll replace the worse pitcher with Rodon.

If that's the case, and Samardzija/Quintana pitch close to as well as they did last year, that's an easy 5+ wins you can add to the projection.
mitchiapet
2/18
I think many Sox fans dismiss PECOTA because it hasn't been uncannily accurate, and wrongly so. Any idea how the South Siders performance vs. PECOTA compares to the other 29 teams?
WaldoInSC
2/19
Right, because what's noteworthy about PECOTA's White Sox projections isn't that they underestimate actual performance, it's that they aren't much better than flipping a coin. PECOTA has been wildly wrong four of the past 11 years, significantly wrong twice more and pretty dead on three times. Anyone one of us could have guessed roughly that badly.
Yonilip
2/19
Here's the thing, the White Sox aren't some sort of outlier in terms of PECOTAs accuracy. If you look at the chart in this article PECOTA has actually missed on their White Sox predictions by an average of 7.8 wins per a season which is actually BETTER than the average number of wins by which PECOTA is inaccurate for MLB teams generally, 8.9 wins. It's not that PECOTA is a system that is unusually bad at predicting things (it's not), but it's about the fact that even the absolutely perfect projection system would be unable, to very accurately predict something like a number of wins which is effected by such a multitude of variables each of which have almost infinite possible outcomes. It's a big reason why projection systems work MUCH better for individual players than for entire teams. Overall PECOTAs track-record for predicting entire-team performance shows it's ability to be accurate to be very similar to that of an intelligent and avid baseball fan making the same predictions without the use of an empirical projection system: They do pretty well on the teams that end up performing as the general baseball community (fans, writers, players, managers, execs etc...) would have expected entering the season, and miss big on the handful of outliers that perform significantly better/worse than expected each year.