SSTN Mailbag: New Bat Speed Metrics!

What a week to be a Yankee fan. The Yankees swept the Twins in completely satisfying fashion, outpitching and hitting one of the hottest teams in the league. Judge is BACK, with a capital "B," Clarke Schmidt and Luis Gil continue to do everything in their power to make me look bad by projecting both to the bullpen, the bullpen has steadied itself, Clay Holmes is the best closer in baseball right now, the Yankees have one of the best catching duos in baseball, and the Yankees are getting contributions from up and down the lineup. Oh, and I didn't even mention that Jasson Dominguez began his rehab assignment (successfully so far, I might add), Gerrit Cole has thrown two good bullpen sessions, and Tommy Kahnle appears close to supplementing the bullpen as his dominant rehab stint draws to a close. You would have to be a complete curmudgeon to complain with anything happening around the Major League team right now.

Will the Yankees keep rolling like this? I think the upside for this team is a 100-win season, so it's certainly possible. Every team has its ups and downs throughout a marathon season, and the Yanks are clearly in the midst of one of the "ups." For now, I'm just going to enjoy it while it lasts. This team will likely need supplementation from either the farm or an outside club at some point this year, but this roster looks mighty fine to me.

As always, thanks for the great questions and keep them coming to SSTNReadermail@gmail.com. In this week's SSTN Mailbag, we're going to do a very deep dive into the new bat speed and contact quality metrics available through Statcast! Let's get at it:

David G. asks: MLB released a whole bunch of new metrics related to bat speed this week. We learned I guess what I would have already expected: Giancarlo Stanton and Aaron Judge swing faster than most. Stanton swings harder than anyone, but he's still a barely above-average hitter. What I don't really understand is how to put these numbers in context. What do they mean for player performance?

For nerds like me, the unveiling of Baseball Savant's new bat tracking statistics was like opening a present on Christmas morning. It has been well-known that teams have been tracking bat speed, swing length, and other contact quality metrics for years, but many of those numbers have not been publicly available until this past week. I admit, I spent entirely too much time late at night digging into all of the numbers when I probably could have used the extra couple of hours of sleep.

I want to start with some very clear qualifiers for these new bat speed and contact quality metrics. For one, we have no idea how long it takes for bat speed and contact quality statistics to normalize. It's not like Baseball Savant released years of data as they did with past metric releases; they only released data for the last two months. Given that fact, I expect that the high-level data has a fair amount of noise. We'll be better able to make some definitive associations based on that data by the end of the season, I would imagine.

I also don't want to jump the gun on analyzing swing speed patterns. I was fascinated by EJ Fagan's analysis on the subject, and I think much of what he wrote has merit, but I would soft pedal some of the comparisons of bat speed distribution between players. EJ makes the assumption that wider swing speed distributions reveal strategic planning on the part of the batter, but I'm not always quite as sure. I think there are some players for whom that analysis is bang on, but others I think likely get fooled guessing on a pitch, and adjust mechanically on the fly to make contact. Those imperfect mechanics lead to slower swings.

All of this is to say that there's a lot we don't know yet, but this data is absolutely worth following and analyzing. I've done some initial analysis, and have what I think are some interesting hypotheses.

One thing to begin with: at least right now, there is almost no correlation between bat speed and overall offensive performance. I ran a query to measure average bat speed against wOBA, and found that the data was completely and totally scattered. Now, that data may show correlation over a larger sample, but it doesn't right now. In fact, bat speed is less correlated with offensive performance than Exit Velocity, by a large margin.

However, there is some moderate correlation between bat speed and Exit Velocity (quick note: all statistics date back to Tuesday/Wednesday of this past week):

Honestly, I expected a tighter correlation between the two metrics, but these are still relatively correlated, which makes sense. Some guys swing hard, but don't have excellent contact quality and vice versa. What you may have noticed by now is that I have highlighted a few players on this chart, all of whom fall either well above or well beneath the trend line.

Here are the players that outperform their bat speed, relative to their average Exit Velocity:

Here are the players that seemingly underperform their bat speed relative to their average Exit Velocity:

So, what do we know about these players? What do they have in common, and where do they separate? As we can see, the players I've picked above the trend line all are either above or well above-average performers in terms of total offense. Players below the trend line tend to under-perform expectations relative to their swing speed.

I am getting into the weeds now, but some of you are likely unfortunately hooked on the game of golf. In golf, we measure ball speed relative to average swing speed. Ball speed, divided by swing speed, equals smash factor, or how well did you strike the ball; how much ball speed did you achieve relative to your potential. I like the squared up metric, but I wanted more of a thousand-foot view of swing speed versus exit velocity. When I measure "smash factor" (average exit velocity divided by average swing speed), I get a range of 1.16-1.38. Players who are getting smash factors of 1.16 are not getting much out of their swing speed on an average basis, while guys at the top of the scale are getting everything out of their swing speed.

Giancarlo Stanton has one of the lowest smash factors in baseball at 1.16. Luis Arraez has the highest smash factor, with Justin Turner just behind him. The players in each list fall where you would expect relative to performance. The players above the trend line all have high smash factors, the players below all have low smash factors. I wanted to see what correlation this has to performance:

Unfortunately, the correlation does not appear very strong. However, if we take out some of the outliers, we get a stronger correlation. I expect this correlation to become stronger as we evaluate more swings this season. The general trend pretty clearly moving in an upward direction: the better your smash factor, the better you perform offensively, though the correlation is not strong.

Again, I think we have a lot to learn about what these numbers mean and what they don't mean. I remain fascinated, and I think in the early going, we have a greater understanding of what makes some players great or frustrating. Luis Arraez swings slower than anyone, but he also gets more out of his swings than anyone. Judge and Soto marry fast swing speeds with crazy performance. Giancarlo Stanton swings harder than anyone, but doesn't have as much to show for it as you would expect due to contact quality. I can't wait to learn more.