MLB

Which Pitchers Are Outperforming Their Statcast Numbers?

Thanks to excellent production on balls in play, Cole Hamels is having fantastic seasons. This may not continue.

Earlier, we looked at pitchers whose production was out of sync with their Statcast data.

What about those hurlers on the other side of the coin, pitchers whose solid results do not line up with the launch angles and exit velocities they have allowed? Generally speaking, these guys are prime regression candidates, as Expected wOBA tends to predict future wOBA better than the past results themselves.

Of the 155 pitches with at least 250 bats against, here are the pitchers with the biggest positive gap between their xwOBA and wOBA, per Baseball Savant.

Parker Bridwell, Los Angeles Angels

wOBA: .304

xwOBA: .348

Diff: .044

ERA: 2.88

We probably do not need Statcast to see that Bridwell's performance is not sustainable, but it certainly confirms what many of us suspect.

The rookie’s good results owe mostly to a .264 BABIP and 86.1% strand rate, neither of which are reliable metrics in a small sample. His 5.9% walk rate is strong, but it is accompanied by a minuscule 15.2% strikeout percentage and a roughly average 1.25 home run per nine innings rate.

His FIP is 4.51 and his xFIP is 4.84, suggesting his underlying performance has been below average.

Bridwell’s average exit velocity is around league average (88.7 miles per hour) and despite just a .146 ISO against, he has been pounded on airballs. His average terminal velocity on flies and line drives is 93.4 mph -- this puts him in the 19th percentile among pitchers who have allowed at least 200 balls in play.

As for his low BABIP, Bridwell has the highest gap between xBABIP (.326) and his actual BABIP among the pitchers in this sample. We probably should not put much stock into 72 innings of an outlier BABIP in any case, but this offers more support for the idea that it's a mirage.

If you have him in fantasy and can get someone to bite on his shiny ERA, do it. Immediately.

Hector Santiago, Minnesota Twins

wOBA: .343

xwOBA: .374

Diff: .031

ERA: 5.63

Could you imagine if Hector Santiago was not getting amazing results on batted balls?

The 29-year-old has just a 6.4% strikeout-minus-walk rate and is allowing 1.9 home runs per nine innings in 70 ⅓ innings, giving him a 6.00 FIP, and putting him below replacement level.

That 5.63 ERA would be even worse if not for a .263 BABIP, which (you guessed it) is not supported by anything. His xBABIP is .291, and no pitcher with at least 200 balls in play has been “barreled up” at a higher than rate than Santiago. (Barreled balls are defined by Baseball Savant as having an expected batting average of .500 and an expected slugging percentage of 1.500.)

He is currently trying to return from a neck injury, but we probably should not expect much from him when/if he gets back onto the field.

Cole Hamels, Texas Rangers

wOBA: .294

xwOBA: .323

Diff: .029

ERA: 3.48

Hamels is by far the pitcher with the longest track record of success in this article. While his wOBA and ERA suggest he is having an excellent year, his xwOBA, FIP (4.52), and Deserved Run Average (4.46) say he has been about average.

A closer look makes it pretty easy to be bearish on the 33-year-old. His average fastball velocity has declined to 92.2 mph, its lowest since 2009, per Brooks Baseball, and his whiffs have all but disappeared. His swinging strike rate has dropped to 8.4%, while his strikeout rate is at 15.7%, which are both the lowest rates of his career by far.

He has still posted a healthy 7.8% walk rate and a 50.5% groundball rate, which has helped keep the ball in the park -- his home run/nine innings rate is 1.01. Still, his .236 BABIP is inflating the results.

Hamels has reliably been a low BABIP pitcher, as his career rate before the season was just .287. Then again, this number is lowered by some excellent early career results, as after running a .255 BABIP in 2011, his number in the 1,053 innings between 2012 and 2016 was .295.

As alluded to, Hamels’ xBABIP is .283. It is indeed lower than the league mean, and is in line with his career average, but it is also considerably higher than his real-life result.

There is also the matter of his power suppression, and as I said in the previous article, this is where Statcast really shines. The “x” stats do a lot better at predicting slugging percentage and home runs than BABIP, which really is as random as we’ve always assumed.

Hamels has allowed an impressive .146 ISO, but his xISO is .171, according to xStats.org, another site that calculates expected production based on exit velocity and launch angle. Hamels’ expected wOBA per xStats is .332.

He has allowed 11 home runs, but based on his quality of contact allowed, that site says he “should have” allowed 13.6.

That said, on the whole, Hamels has indeed been better than average on batted balls. His .341 xwOBA on contact ranks 31st among our 153 pitchers with at least 200 balls in play.

This .341 mark is still worse than his MLB-best .302 actual wOBA on contact, and not enough to truly offset his poor production on balls not in play -- his 7.8% strikeout-minus-walk percentage is one of the lowest in baseball.

The combination of Hamels’ name value and low, yet regression-prone ERA makes him an excellent sell-high candidate.

Jose Urena, Miami Marlins

wOBA: .312

xwOBA: .341

Diff: .029

ERA: 3.61

Like Bridwell, we can assume that Jose Urena is headed for regression without Statcast data. But this new information strengthens this inference.

His ERA is not supported by his 4.91 FIP or his 5.39 xFIP. The difference here stems mostly from his .245 BABIP, which seems to have come out of nowhere. His mark here over his first 145 ⅓ big league innings was .307. In 278 innings of Double-A and Triple-A pitching, his BABIP was .310.

If the 25-year-old had some ability to suppress production on balls in play, it has not manifested itself until now. And sure enough, it does not look like this is actually some new skill, but rather a mere sample size fluke. His xBABIP is .294 and the gap between his actual and expected results is third highest, behind only Bridwell and Hamels.