The Unimportance of “Productive Outs”

It’s something we saw quite a lot last year. The Diamondbacks at bat, with a runner on third and one out. But the hitter was unable to get the man home, either striking out or not being able to deliver a sacrifice fly/RBI ground-out. Cue weeping, wailing and the gnashing of teeth in the Gameday Thread. But here’s the dirty little secret. While of course it “matters” at that moment - scoring the run is better than not scoring it - the reality is that there is little or no correlation between the teams that are good at making productive outs and overall success. This cropped up earlier this week, because of a Tweet from Jerry Weinstein, who was bench coach for Team USA at the 2022 Tokyo Olympics.

He went on, “We may be too focused on the big inning. We rely on match ups & those match ups often fail us. Yes the “big inning” drive the ball approach wins a lot of games,but the score the runner from 3B now approach can produce multiple runs over the course of a game. We just need to pick our spots better & commit to making productive outs when the situation calls for it.” However, as Kyle Kishimoto subsequently pointed out, “In 2022, team productive out rate and regular season wins had an r^2 of 0.061” - in other words, there was basically zero connection between the two stats at all, as the chart below shows. There’s absolutely no indication that “Championship teams know how to make productive outs.”

You may be stunned to realize that the Diamondbacks were actually the best in all of the majors in terms of productive out percentage last year (defined by the Elias Sports Bureau as “when a fly ball, grounder or bunt advances a runner with nobody out; when a pitcher bunts to advance a runner with one out (maximizing the effectiveness of the pitcher’s at-bat), or when a grounder or fly ball scores a run with one out”). Arizona was credited with a productive out in 32.5% of opportunities, over two percent better than the next best team, and well above the MLB average of 26.9%. This illustrates the power of selective memory: we focus on and remember the 67.5% of the time when the team made unproductive outs.

If you want further evidence, look at the 2021 Giants. Won 107 games, beat LA for the NL West? Well, late in the year, Grant Brisbee wrote an article: “The Giants will have the lowest percentage of productive outs in history, and you shouldn’t care”. They ended the year with just 21.9% of productive outs made, the lowest over a full season in recorded baseball history. Yet it didn’t even hurt them in close contests, where you’d expect it to show up most. San Francisco were 31-17 in one-run games, a very similar mark to their overall record. It’s because, as Grant said, “The Giants’ game plan is to hit the ball hard. It’s not to hit the ball to the right side at all costs... It’s why they’ve set a franchise record for home runs.”

The basic reason for the lack of correlation is simple: there’s one thing which is far better than productive outs, and that’s productive non-outs. And the players who strike out with a man on third, also tend to be the ones who are more likely to homer in that situation. We can see this if we look at who were the best D-backs at making productive outs in 2022. Discounting those with less than a handful of chances, the leaders were Jose Herrera, Sergio Alcantara and Geraldo Perdomo, who were almost at 50% combined (38 of 77). However, they were also very “good” at making outs in general, with all three having a on-base percentage between .250 and .285, well below the MLB average of .312.

The history of the “productive outs” fallacy goes back almost twenty years now, to an article Buster Olney wrote for ESPN, early in 2004. This prompted an immediate response from the sabermetric community, which was unimpressed. For example, Larry Mahnken of The Hardball Times wrote damningly, “It’s clear that Olney did very little research for his article, and what research he did do was data mining, trying to find stats that supported his claims.” Later that year, Anthony Passaretti of Baseball Prospectus concluded, “There is simply no evidence to support the notion that making productive outs is a legitimate, repeatable skill. Nor is their any reason to believe that they are the key to winning games.”

I won’t go into a particularly detailed analysis, because it’s something I wrote about back in 2012. I suspect this was probably at the peak of the Mark Grace, “Good things happen when you put the ball in play,” era. But what has happened in the decade since, is that productive outs have declined significantly. In 2012, at the time of my original piece, POs were made in 31.5% of opportunities, with just one of the thirty teams at 27% or lower. This year, the MLB average is only 26.9%, with nineteen teams at 27% or lower. A major factor is likely the increase in strikeouts, up from 19.8% in 2012, to 22.4% last season. Strikeouts and POs are almost mutually exclusive.

Yet, despite their decline, their importance remains an oft-cited mantra. Here’s three minutes of Harold Reynolds on his knees, worshiping at the alter of productive outs in the White Sox-Astros post-season series from 2021. Never mind that Houston simply had an OPS 66 points better than Chicago over those four games. Productive outs, baby!

As Jerry Weinstein’s Tweet also shows, the idea of Productive Outs is one which dies hard. Again, they do exist, and are better than the alternative. But hitting a 98 mph, dipping fastball is a hard enough proposition as it. I suspect trying to force batters to think about making a productive out is likely to prove counter-productive.