In general, I’m a believer in the advanced metrics you find on baseballsavant.com or baseball-reference.com. While recognizing they aren’t the be all and end all of analysis, I use them as a piece of the puzzle, instead of the magic eight ball that answers all questions.
One of these that is often quoted on Houston Astros broadcasts is expected batting average (xBA) which purports to provide a player’s “expected average based on the quality of contact, instead of actual outcomes.”
It’s also one of the advanced metrics I have the least faith in. I understand how it’s calculated and can even partially understand why it’s calculated the way it is. The problem for me is that about halfway through the season I started seeing some ridiculous numbers that made me question the method. For example, I noticed a 400 foot home run that cleared the left field fence by 40 feet assigned an .830 xBA.
That’s where it got the best of me, and I decided to look into the curious case of Astros’ Yuli Gurriel.
Gurriel is currently slashing at .324/.388./490 with an OPS+ of 143 and a wRC+ of 145. By any measure an outstanding offensive season. Yet, when you look at Gurriel’s xBA it’s a pedestrian .269. That difference of 55 points is one of the largest I can recall and caused me to take a deeper dive into Gurriel’s numbers.
Gurriel is batting .330 against the fastball, despite the xBA sitting at .266, and he’s crushing breaking balls at a .341 clip despite an xBA of .265 on those pitches. Those are 64 and 76 point differences, respectively.
Gurriel has 74 hits in 224 at bats on the fastball for that .330 average and to get near the xBA of .266 he’d have to have about 15 (rounding) less hits in those 224 at bats. That means 15 “luck” hits in 224 at bats on fastballs. Seems like a lot.
Gurriel’s exit velocity, while not outstanding, is the highest of his career and while his launch angle is down from recent seasons, it’s higher than his career average.
So what would it take for Gurriel’s actual batting average to equal his current xBA? 0 for 74. Also, assuming 200 more at bats on the season, Gurriel would have to go 34 for 200 (.170) over the rest of the season.
Color me skeptical. Is it likely Gurriel will regress over the balance of the season? Of course. He’s 37 years old and his current .324 mark is 25 points higher than his previous career high and I do believe he’s hitting in some luck. Some being the operative word.
More from Climbing Tal's Hill
- Just how much better is the Houston Astros playoff rotation than the rest?
- Houston Astros: A Lineup Change to Spark Offense
- Astros prospect Hunter Brown throws 6 shutout innings in debut
- Always faithful Astros World Series champion Josh Reddick defends the title
- Michael Conforto declines Astros’ 2-year, $30 million offer
That said, we’re two-thirds of the way through the season, the majority of his at bats are behind him and you’re not taking any of his 117 current hits away.Sometimes when I post advanced metrics such as xBA I’m greeted with “you can’t measure everything with a formula.” People may be surprised to find out I agree with this.
Moving forward, I see a couple of possible outcomes: 1) A deep slump for the rest of the season as Gurriel regresses towards the mean and his xBA or 2) for some reason the xBA metric isn’t tracking the results of Yuli’s batted balls well.
I understand “averages” and that Gurriel is one of thousands of batters tracked by Baseball Savant and it all “averages out” over large samples. But Gurriel’s the only one I’m interested in for this article and his numbers are way out of whack.
In the end, I’m of the belief that regression will likely come, but not to the point that the xBA projects. So, this is one of the times I agree: You can’t measure everything with a formula.