2016’s Biggest xISO Disparities
Early last season, I introduced an equation that calculates a hitter’s expected isolated power, or xISO. Since I haven’t discussed it all so far this season, and we’re halfway through June, I figured now is better than never.
Before I proceed: Andrew Dominijanni expanded on my research about a month ago. He incorporated exit velocity, sourced from the new Statcast data hosted at Baseball Savant. The model better explains the variance in the data and has slightly better predictive (year-to-year) qualities, making it the optimal choice.
Andrew’s version of xISO generates a pretty simple calculation, but I’m feeling especially lazy today, and I can find hard-hit rate (Hard%), pull rate (Pull%) and fly ball rate (FB%) all on FanGraphs’ batted ball leaderboards.
If you ever want to calculate xISO on your own at any point, you can use the calculator I embedded in this post. One day, I’ll comb through FanGraphs’ Community archives and catalog Andrew’s and others’ great work in my omnibus post. It takes some extra time — something of which I have lacked lately. But I’ll get around to it. Promise!
Here are xISO calculations for all qualified hitters.
Five Biggest Overperformers
Robinson Cano, SEA 2B
ISO: .270 / xISO: .190
Does this surprise anyone? It’s nice that Cano rediscovered his power. It might be one of the feel-good stories of the year. He has exceeded his 2014 home run total and almost matched his 2015 total in roughly half the plate appearances.
While Cano definitely won’t hit 40-plus home runs like his current pace suggests, his “true” .190 ISO would be his best mark since 2013, when he still hit 25 to 30 bombs per year. He’ll fall off — maybe only a dozen more homers this year — but that doesn’t necessary constitute a sell-high at one of the shallower offensive positions.
Carlos Gonzalez, COL OF
ISO: .247 / xISO: .168
This is the latter half of 2015 all over again. Remember how miserable the start of that season was? With that said, CarGo has always flashed excellent power, averaging 31 home runs per 650 plate appearances throughout his career. This year likely won’t be much different. But his production against left-handed hitters can be characterized as anemic at best and really should be platooned. It is what it is. I would sell high no matter what, but anyone who holds a grudge will know I shouldn’t even bother trying to assess Gonzalez anymore.
Adam Duvall, CIN OF
ISO: .339 / xISO: .261
Duvall’s calling card has always been power and not much else. To attest: he has struck out more than seven times as often as he has walked this year. Duvall is 2016’s third biggest overperformer so far, but his .261 xISO ranked sixth among all qualified hitters. That’s a big deal. This is a Chris Davis and Chris Carter type of hitter, with all-or-nothing power. We know that these types of hitters can be frustrating to own, but vicious droughts typically precede bountiful springs.
David Ortiz, BOS DH
ISO: .379 / xISO: .307
Ortiz owns the league’s best xISO. ‘Nuff said.
Michael Saunders, TOR OF
ISO: .296 / xISO: .227
Saunders’ three home runs Friday might fully account for his xISO overperformance. Saunders’ peripherals don’t point to a profoundly changed hitter, but he hasn’t hit fly balls this often since 2011. It warrants a bump up from the .170ish ISOs he posted in Seattle and makes him roughly J.D. Martinez’s left-handed clone, down to the plate discipline. Not a bad comp for a dude who has always seemed to be on the cusp of providing meaningful fantasy value.
Five Biggest Underperformers
Yonder Alonso, OAK 1B
ISO: .076 / xISO: .176
Well… that’s not entirely surprising. Alonso doesn’t have power resembling anything close to prodigious, but he’s usually good to threaten double-digit home runs in a full season’s worth of plate appearances. He hits way too many ground balls, but he seems to be making solid contact, warranting probably a few more home runs than his output suggests. If you’re in a deep league, catching his outburst before it happens would be a prescient move.
Chase Utley, LAD 2B
ISO: .131 / xISO: .223
I don’t think Utley was expected to see this much playing time, but alas, the injury bug bites (as does the DFA bug for Carl Crawford). He’s hitting fewer fly balls than ever, but he’s also blasting line drives at a career-best rate. Most importantly, though, is Utley’s 40.7% hard-hit rate, which would be the second-best of his career. His plate discipline seems to corroborate the results — selling out for power and inducing quite a few strikeouts.
It’d be easy to dismiss his current power as more of the same — he posted .138 and .131 ISOs the last two years, so his .131 mark this year fits the narrative. But the peripherals suggest otherwise, and we could witness a resurgence of Utley’s career later this summer.
Asdrubal Cabrera, NYM SS
ISO: .132 / xISO: .218
Cabrera has been a steadfast middle infield option for half a decade, and the baker’s dozen of home runs he hits this year will be plenty valuable in deeper leagues. But note that Cabrera has arguably been making the best contact of his career, registering more hard hits and pulling the ball more than ever before. Like other names on this list, he’s suffering from some ground ball issues, but Cabrera’s contact “improvements” (depending on how you look at them) could warrant a minor power surge in the coming months.
Jed Lowrie, OAK 2B
ISO: .058 / xISO: .139
And another “Well…” goes to another Oakland infielder. The power Lowrie flashed in Houston (the first time) and Oakland (also the first time) has all but evaporated — even his .139 xISO would be among his career-worst. Were it not for his elevated batting average on balls in play (BABIP), Lowrie might be out of the job; then again, he also plays for Oakland.
Justin Upton, DET OF
ISO: .127 / xISO: .208
How poignant that we finish with one of 2016’s most intriguingly pitiful performances thus far. Upton has shaved his strikeout rate (K%) down to 25.8% in June (as of June 16) and has already doubled his home run total from the first two months. The .208 xISO is almost identical to his production the last three years in Atlanta and San Diego. I don’t know if Upton has recovered his plate discipline for good, but the power should show up in spades soon. This was a fairly obvious recommendation from the very start, no matter how bad he looked in April, but: buy low.
You commented on Lowrie’s BABIP, but didn’t speculate on whether there might be a correlation with his ISO/xISO split. Safe to assume that means you don’t see a plausible one?
He’s spraying to all fields nicely and hitting lots of line drives. It warrants an above-average BABIP, but I’m not so sure about .346.
Thanks for the acknowledgement, Alex!
To add to this piece, the version of xISO that I calculate is actually even more pessimistic on Cano (0.175) and Duvall (0.242). It also would have expected better performance for a couple of the underperformers, but it’s not quite as optimistic as Alex’s xISO for Alonso (0.155), Utley (0.169), and Cabrera (0.150).
One also notes that it seems like some of the overperformers are in good hitting parks, while some of the underperformers are in pitcher’s parks. I haven’t written it up yet or anything, but some preliminary analysis has shown no overall trend between the xISO-ISO discrepancy and park factors.
Not surprised that your data handles the outliers better. Actually makes perfect sense, given its improved fit!
Park factors are tough — in aggregate, they tend to show distinct trends, but per batter, they can be a little noisy. And then there’s the issue of pitcher factors endogenous to the park factors… it gets messy. So don’t be surprised if more in-depth analysis doesn’t turn up anything concrete.
Interesting thoughts on Gonzalez. It looks like he is hitting more line drives so maybe that impacts his xISO by a little. The decline in pulled balls in play is definitely a decent reason to sell.
I also want to note that an unpublished version of “my” xISO uses LD% and FB% separately instead of GB% to account for batted ball mix. In this case, fly balls have a greater “linear weight” than line drives, meaning more line drives might actually be bad for his xISO.
While my previous comment regarding park factors having no overall trend with the xISO discrepancy is true, I’m not prepared to go to the mat for that hypothesis with Coors. I don’t think we yet have analyses that are granular enough to say how, for example, pull line drives/fly balls interact with specific parks to affect outcomes.
Anecdotally, for Rockies, Story, Arenado, and CarGo and Parra are all exceeding their xISOs to varying degrees. But then there are guys like Mark Reynolds, Charlie Blackmon, and DJ LeMahieu who are underperforming. I don’t feel comfortable saying this is a proven Coors effect, but I just wanted to present a counterpoint on CarGo.
That’s some good work you’ve done! Really interesting stuff.
Thanks! Real credit goes to Alex et. al. who are willing to put their analysis out there on a regular basis and give dilettantes like me a place to start.
From a splits perspective on CarGo, he typically (aka career rate) hits a home run against LHP once every 24 AB, with paces of 44 and 32 AB in 2014 and 2015, respectively. Currently on a 15-AB pace… it’s just guesswork at this point, but I think you could tie the expected regression (or, conversely, the overperformance thus far) to that. Granted, they’re all small samples.
I’ve been suffering with Upton for a while now, so its good to see a projection which possibly points to him increasing his production. And I really need it because the OF depth of my fantasy team is just bad. I also have the severely underperforming Carlos Gomez and injuries to Brantely (thinking of dropping him), Sano, and Souza. Luckily Odubel Herrera was still available in my league, although he obviously doesn’t replace the power production of all the other guys.