A Quick Chat About Stat Stabilization

The 2020 season has been anything but ordinary. As such, some players are performing at a higher clip than what we’re accustomed to seeing while a variety of established stars are greatly struggling. Look at the success from Mike Yastrzemski (2.2 fWAR, 154 wRC+ in 47 games) and the overall scuffling from last year’s NL MVP, Cody Bellinger, (0.4 fWAR, 90 wRC+ in 45 games) as two pertinent non-Astro examples. Odds are that Bellinger will eventually start hitting like he has shown in the past sooner rather than later while the jury is still somewhat out on the (promising) grandson of a Red Sox legend. That said, it is a stark difference which catches the eye in the year of 2020.

I guess what I’m trying to point out is that a 60-game regular season is going to generate some unexpected numbers and trends. In baseball, we also know that it takes varying levels of playing time or events before stats aren’t considered small samples. Walk rate, for example, usually takes 120 plate appearances to stabilize for a hitter. For pitchers, they need about 170 batters faced. BABIP, for a hitter, requires 820 balls in play while a pitcher requires 2,000 of the same batted ball events. FreezeStats has a handy breakdown of this information, which is of great value in a short season like this one.

Much has been made of some of the small samples in this 2020 season, especially about the Astros for obvious reasons. Sooner or later, I’ll get around to a post about Jose Altuve’s issues. Got to get in the right frame of mind for that one. In any case, the usual metrics that we’ve come to rely upon are prone to noticeable swings if a player does well or performs badly. For example, Kyle Tucker had a .338 on-base percentage through September 6; his recent 1-for-23 stretch lowers that figure to a .305 on-base percentage. To stabilize an on-base percentage for a hitter usually requires 460 plate appearances. Tucker, for context, only has 154. The potential swing in stats one way or another is pronounced this year, which makes analyzing players more difficult than before. To truly see if a trend is proven true, we’re likely going have to see it in 2021 as well.

Another stat favorite of mine — if I can even call it that — is Andre Scrubb’s walk rate. In fact, he is one of four relievers with at least ten innings pitched this season with a negative strikeout-minus-walk rate of -1.3 percent. Put into another way, Scrubb is walking 9.93 batters per nine innings while striking out 8.84 per nine. But he’s only pitched in 18 ¹⁄₃ innings and faced 80 batters this season. That said, it does take 170 batters, or roughly 41.5 innings, for a walk rate to stabilize. Scrubb is nearly half way there to stabilization for that one metric, but it won’t come this season. Control issues aren’t new for the right-hander, though, as evident by his minor league numbers. Unless there is a drastic improvement for the 25-year old, I think we already have a rough idea what he will be in the major leagues if a career-long issue holds true.

The level of uncertainty surrounding numbers, as you can gather from above, is quite high in 2020. Nothing about this year has been normal. That said, there is one metric for a particular catcher that will require more careful attention, which involves the walk rate leader of the Astros offense: Martin Maldonado. Yes, you’ve read that right as Maldonado currently holds the highest walk rate (15.8 percent) of any Houston hitter with at least 10 plate appearances. That is something as, again, it only takes 120 plate appearances for walk rates to stabilize and Maldonado has 130 entering Tuesday. His career walk rate entering this season: 6.9 percent. I plan on doing a deeper dive into Maldonado and his approach as a hitter at some point in the near future, but figured it was pointing out for the sake of this conversation.

In any event, some will point to this season’s stats as proof of certain narratives. Again, the Astros will have that issue for obvious reasons. At the same time, stats require a certain number of plate appearances, at-bats, pitches thrown, or batted ball events before they truly mean anything substantial. That is not say that we can’t suspect certain trends are unfolding, but be open to new developments in the future.