Is Tommy Pham Lucky?

Is Tommy Pham lucky? It’s hard to say he’s not- he lives in a first-world country, and is a world-class athlete who is making ten times the median US household income to literally play a game. Is Tommy Pham lucky? It’s hard to say he’s lucky- he has a manager who delights in refusing to play him, he’s a 29-year-old who might never get a chance at the free agency dollars that a player of his caliber whose service clock randomly started 3 years earlier might attain, and here is a sentence from the Wikipedia article on keratoconus: "In a small number of people scarring of the cornea occurs and a corneal transplantation is required." In short, deciding if someone is lucky or unlucky is so dependent on context as to be impossible.

Is Tommy Pham lucky when it comes specifically to the outcomes he’s achieved while batting? On this question, at least, I can give you a solid ‘it really depends’. This question, at least, is something we can investigate further. One way you could look at it is to check his BABIP. On the surface, the .374 he’s running this year is above both his career .354 average (.336 excluding this year) and his .335 projection the rest of the way. He’s a pretty fast guy- maybe we could look at the BABIP for qualified MLB centerfielders to see if that does something for us? A cool .316, and that’s after flattering the sample size a bit by removing Curtis Granderson, who is a centerfielder only in the prodigious imagination of Sandy Alderson. From a BABIP lens, yeah, he’s getting a little lucky.

Luckily for me, the story doesn’t quite end there, and there are more ways of looking at his output this year than just how he does when he puts the ball in play. The way he puts those balls into play also matters, and Pham hits the ball incredibly hard. I’m not even exaggerating- really, really hard. Here are the top 10 players in the major leagues by exit velocity on fly balls, with a minimum of 25 results (all data in this segment is courtesy of Statcast):

  1. Aaron Judge

  2. Khris Davis

  3. Joey Gallo

  4. Giancarlo Stanton

  5. Miguel Sano

  6. J.D. Martinez

  7. Gary Sanchez

  8. Tommy Pham

  9. Mike Trout

  10. Trey Mancini

That’s 8 extremely large gentlemen, a tiny Athletic who hit 40 dingers last year, and Tommy Pham. Think that’s a fluke? Here are the top four hitters from 2016 in exit velocity on line drives and fly balls, minimum 24 results (I massaged the numbers a little to put an interesting name on there):

  1. Aaron Judge (24 results)

  2. Nelson Cruz

  3. Pedro Alvarez

  4. Tommy Pham

Now, did I cherry pick a sample set that makes Pham look good on both sides by using fly balls in 2017 and a blend in 2016? I absolutely did. Use a blend in 2017, and Pham drops all the way to 40th, just between Wil Myers and Justin Bour- still impressive company.

Hitting the ball hard is all well and good, but does it mean Pham should expect better results, and does it mean he’ll keep hitting the ball hard going forward? Well, essentially, yes and yes. To get the easier of the two out of the way first, exit velocity on air balls in one year is extremely well correlated to exit velocity on air balls the next year (63% R^2). This makes a lot of sense- this feels like an inherent characteristic of players. Does it predict good results? As a proxy for this, I checked the correlation between 2016 air EV and 2017 wOBA. Encouragingly, there’s a medium (.34) correlation between the two, which is about as good as 2016 wOBA (.33), though it can’t hold a candle to 2016 xwOBA (.41). Next, we want to add a simple guess at how much contact people are giving up by swinging so hard, since we need to account for the fact that good ol’ Randal is in 30th place on the 2017 Air EV list. By adding Whiffs per Swing (not literally swing-and-miss, but a good proxy for how often a player connects when he intends to), we’re now even with xwOBA in correlation (.4). Here’s a leaderboard for how the multiple linear regression of Air EV and whiffs per swing predicts wOBA going forward:

  1. Aaron Judge .378

  2. Ryan Braun .375

  3. Khris Davis .372

  4. Ryan Zimmerman .371

  5. Manny Machado .368

The top 20 is dotted with All-Stars. Here’s 13-17 (relevant to us) with their predicted wOBA’s:

13. J.D. Martinez .360

14. Marcell Ozuna .359

15. Tommy Pham .359

16. Joey Gallo .358

17. Bryce Harper .358

Not bad company to keep. As an interesting wrinkle, plugging in the same formula to Pham’s abbreviated 2016 production gives an output of .358 (neat bonus- his 2017 xwOBA is also .358). He’s oddly consistent, though this year his EV and whiffs per swing are both down, perhaps reflecting a swing that is a little more under control. To put that in context, that’s going to look like a wRC+ in the 120-130 area (the vagaries of park adjustments prevent me from coming to a precise answer here).

Here are a few more Cardinals of interest:

Randal Grichuk .351

Matt Carpenter .343

Paul DeJong .342

Yadier Molina .336

Dexter Fowler .332

Jedd Gyorko .331

Stephen Piscotty .317

So, why use this metric over xwOBA? One major reason is that it’s a lot easier to calculate. xwOBA is based on a computationally intensive bucketing of every batted ball over a year’s worth of data- it’s not something I could cook up in a Google Sheet in ten minutes. Additionally, this might be an even cleaner way of thinking about the process of hitting. While xwOBA is divorced from direct outcomes, it still comes with a bit of inherent luck- the bucketing process creates some odd bright-line divisions between angles, and batted ball mix can produce some extreme and non-intuitive answers, which are hard to check because of the above-mentioned computational intensity. How hard you hit the ball and how often you hit it, on the other hand, are pretty tough to fake. Assuming roughly even predictive power, it’s nice to have a tool that I can understand a bit better.

So, how lucky is Tommy Pham? Well, his wOBA this year is .393. Turns, out, he’s getting lucky. He’s been lucky enough to be the 23rd best hitter in baseball this year by wOBA. And yet, he’s projected to be the 15th-best hitter in baseball going forward by wOBA, even without continuing his current batted-ball luck. So is Tommy Pham lucky? Absolutely- just like everyone else on a single-season hitting leaderboard. Is Tommy Pham skilled? More skilled than lucky, I’d say.

Note: all statistics are current through Sunday night's games, before Pham went off again on Monday.