On Monday, reader Andrew left a comment on my article asking how I felt about different ways to use Greg Rybarczyk’s HitTracker data to evaluate a hitter’s power. This reminded me of an article that I started writing a while back but never got around to finishing, so I’m putting it together now.
As you probably know by now, I’m a big fan of using HitTracker as a tool for evaluating hitters, but there are some pitfalls that we need to look out for. One of the most common ways of evaluating a player’s power using HitTracker is to focus on his average distance. Today, I’d like to talk about why this could be a dangerous approach and how it can lead to some incorrect conclusions.
The easiest way to do this is by starting with a hypothetical example. Let’s say that two players, Player A and Player B, each hit three 450-foot home runs. In their next at-bats, Player A squeaks one over a 320-foot fence, while Player B has one knocked down by wind at the same fence. The average distance for Player A drops from 450 feet to 418 feet while Player B’s average distance stays at 450 feet. While there is now a large gap in their average distances, is Player A’s power any worse than Player B’s? And is Player A any worse of a power hitter now than he was before the hit? Of course not.
What we need to understand is that HitTracker shows average home run distance, not average fly ball distance. Because HitTracker only tracks home runs, all we get is average home-run distance, which can be very misleading, especially when you're dealing with a player that you believe has been lucky or unlucky (which make up a large percentage of the players that we bother evaluating in the first place). This also makes comparing the same hitter’s average distances from year to year difficult and potentially unreliable because if he hasn’t had the same level of luck, we’re not really comparing apples to apples.
Further complicating matters is that ballparks have wildly different dimensions. Batters who play in a park with deep fences are going to have higher average distances because all of their shorter flies are going to be caught before they reach the fence and won’t be included in the dataset. Conversely, batters who play in parks with shallow fences will have lower average distances because more of their flies will go for homers. And since every hitter plays half of his games in road ballparks with wildly different conditions, homers hit on the road add even more noise to the equation.
Yet another consideration is that not every player hits the ball to the same part of the park. For hitters who have extreme pull power and rarely hit homers to center, their average distances are going to be lower, but that doesn’t necessarily make them any less likely to continue hitting a lot of homers. Left-field and right-field fences are shorter than center-field fences, so a player with extreme pull power can get away with hitting the ball a shorter distance (and even if they hit deep ones, they’ll have more short ones fall in to drive the average down). Jose Bautista, for instance, leads all of baseball in home runs with 24. Despite this, he’s hit just one ball further than 450 feet, and his average true distance (397.7 feet) is roughly league-average (396.5), but there are few who would argue that Bautista is an average home-run hitter. Bautista plays in a park with relatively shallow left-field fences and hits the vast majority of his homers to left.
This isn't to say that average distance doesn't have any value—after all, it’s probably tells us something that we wouldn’t know otherwise. We just need to be very careful when using it to analyze an individual player and avoid placing too much emphasis on it. This goes along with the point I often make about knowing what you’re looking at. The more you know about what you’re looking at, the flaws it contains, and the proper context, the better off you’re going to be.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.Subscribe now