clock menu more-arrow no yes mobile

Filed under:

Has the Yankees’ offense been helped by good fortune?

What does expected wOBA have to say about the Yankees' red-hot start?

MLB: St. Louis Cardinals at New York Yankees Andy Marlin-USA TODAY Sports

Baseball is famously finicky. A terrible team might defeat an elite one on any given night. Screaming line drives are caught for outs all the time, and lazy bloops go for base hits on the regular. Some might even say that the game is difficult to predict. Most observers certainly wouldn't have predicted the Yankees would have owned the league's best record thirty games in (and second-best record today).

Of course, over time, we've gotten a little better at predicting the game, and we've also improved at accounting for the lucky bounces of the ball that can occur. Fans and analysts have known for years to pay attention to things such as BABIP, or discrepancies between pitcher ERA's and FIP's, to try and gauge if certain players or teams are disproportionately benefiting from baseball's quirks.

With the advent of Statcast, we can account even further for the possible consequences of luck. Armed with the exact launch speed and angle off the bat of every ball, we can estimate the chance that any particular ball in play falls in for a hit. Baseball Savant uses that batted ball data to publish leaderboards and breakdowns for hit probability and catch probability.

Now, using hit probability, Baseball Savant has unveiled a metric to try and strip out as much chance as possible when evaluating offensive performance: expected wOBA. Actual wOBA, or Weighted On-Base Average, is a strong all-encompassing offensive stat that captures the value of everything a hitter does in the batter's box, from drawing walks to smashing doubles and triples. Expected wOBA uses hit probability to assess the number of singles, doubles, etc. a hitter "should" have earned, based on quality of contact. It then incorporates a hitter's actual level of walks and strikeouts to calculate a full expected wOBA figure.

In theory, something like expected wOBA should be a better indicator of true player performance than normal wOBA, and perhaps a better estimator of what a player is likely to do in the future. So, how do the Yankees look by expected wOBA? Have they been unlucky? Have the balls bounced their way in the early going?

Through Monday's games, 292 batters had recorded at least 50 at-bats. I calculated the difference between expected wOBA and wOBA for each of those players and ranked them by that difference, largest to smallest. Here's how the Yankees fared:

Expected wOBA vs. Actual wOBA

Player Expected wOBA Actual wOBA Difference Rank
Player Expected wOBA Actual wOBA Difference Rank
Greg Bird 0.311 0.222 0.089 4th
Chris Carter 0.314 0.266 0.048 30th
Austin Romine 0.313 0.308 0.005 116th
Jacoby Ellsbury 0.341 0.353 -0.012 170th
Chase Headley 0.337 0.352 -0.015 183rd
Aaron Judge 0.463 0.485 -0.022 199th
Matt Holliday 0.377 0.407 -0.03 217th
Ronald Torreyes 0.279 0.322 -0.043 243rd
Brett Gardner 0.312 0.362 -0.05 253rd
Starlin Castro 0.344 0.409 -0.065 270th
Aaron Hicks 0.363 0.461 -0.098 290th
Total 0.347 0.37 -0.023

The first takeaway is that Greg Bird has probably suffered some poor fortune. His expected wOBA mark of .311 isn't great, but it's a vast improvement over his actual figure. His average exit velocity of 87.7 mph is right around league norms, suggesting his struggles so far have had less to do with poor contact and more to do with often not making contact at all (not to mention his injury issues).

Aaron Hicks also sticks out with the league’s third-largest negative differential between his expected wOBA and wOBA. Hicks has managed to run one of the highest wOBA's in the league despite making fairly soft contact, at least per his exit velocity (83.7 mph). That being said, his top-notch plate discipline has helped fuel an impressive expected wOBA of .363, far above his previous career standards, so Hicks has been a revelation regardless of his probable good fortune.

In totality, there's good new and bad news. The bad news is that the Yankees' offensive performance isn't entirely backed up by the underlying metrics. As a whole, the Yankees' eleven most prominent hitters have outperformed their cumulative expected wOBA by 23 points. This suggests that they've probably been a little lucky to rake the way they have, and they should be expected to regress a bit moving forward.

And yet, this shouldn't really be a surprise! The Yankees entered the year expected to be fairly middle-of-the-pack, and instead have legitimately looked like the best team in baseball. We didn't expect them to be the best team in baseball a month ago, so it's only natural that we probably shouldn't expect them to be the best team in baseball the rest of the way. This isn't to say they definitely won't dominate the whole year, but our intuition is that it's unlikely they will be this good all season, and that intuition is supported by the numbers.

The good news, on the other hand, is that even if we penalize the Yankees for overperforming their expected wOBA, they still have looked excellent. The best team wOBA in the major leagues across the past three seasons? That belonged to the Red Sox, at .346 in 2016. These Yankees' hitters have an expected wOBA of .347. Even after accounting for possible good fortune, the Yankees' offensive performance has been on par with the best team offensive performance in the game the past few years.

So if the Yankees' offense merely regresses to the levels they "should" be at, they will still be superb. There's no guarantee that the Yankees will be able to sustain even that level of play. Still, it's a good sign that even if we accept that they've probably been a bit lucky, the Yankees have played like an excellent offensive team so far, and look pretty deserving of the shiny second-best-record-in-baseball label they own at this point.