Does a Heavy Workload Hurt Yadier Molina?

After Yadier Molina burst out of the gate with three homeruns in the first week of the season, the jokes about Yadi’s homerun pace commenced on Twitter. I know because I was one (of many) cracking wise about it. Then I was asked to identify the earliest he had hit his third homerun of the season. The answer prior this year was April 8, 2014, the eighth game of the season. All of this led me down a rabbit hole in which I discovered that Yadi has hit a seemingly disproportionate number of homeruns very early in the year throughout his career. Who knew? I wanted to write an article about his odd ability to get off to fast starts. But upon digging deeper, reality sunk in and it simply wasn’t true. Sure, he still hits for more power in the first week of the season- a not insignificant 3.11% HR rate through the first 7 games, compared to 1.93% overall in his career. But that modest power surge still only produced a .234/.286/.385 slash line, well below his career .284/.336/.404 rates. In other words, homeruns or not, Yadi is not especially a fast starter. But something else emerged after I collected the data on a game-by-game basis throughout his career, and it begs a question we ask almost every year when wondering which poor schlub will get stuck backing up Yadi. Does a heavy workload make Yadi less productive?

First, let’s establish that Yadi does in fact handle a heavy workload. We all know this. You don’t need a graph to illustrate it. However, we might take for granted just how much of a heavy workload Yadi has handled in his career. Since I have the data, let’s visualize it to show just how much time he’s spent behind the plate. Using Baseball-Reference’s Play Index, I’ve limited my search to players specifically playing catcher (Buster Posey’s time at 1B doesn’t count here, for example), individual seasons since 2005 (when Yadi broke in as a full-time regular). This graph represents the 300 individual catcher seasons with the most games played at the position. It’s a scatterplot comparing their number of games played at catcher compared to their relative OPS (OPS+, as it were) within the split- players playing catcher. Yadi’s 124 OPS+ on this graph, from 2016, indicates that he was 24% better than the league average hitter when they played catcher. And lastly, I’ve highlighted Yadi specifically. 2018 is obviously excluded here.

Catchers have racked up 129 games or more at the position in a single season 46 times since 2005. Yadi occupies 8 of those slots, including the 4th, 16th, 21st, 22nd, and 29th most games played. Every single season, Yadi has fallen in the top 300 in this group. He has played 1,629 games at catcher, by far the largest total in the timeframe. The next closest is Russell Martin, with a not-really-close 1,340. Suffice it to say, Yadi has handled an enormous workload. Again, we all knew this, but this should illustrate just how heavy that workload has been for a very long time. (Side note: see if you can identify Yadi’s 2006 on the graph. It’s not hard- it’s the one waaaay over to the left)

Now let’s take a look at his game by game performance since 2005 and see if it tails off the deeper it goes into the season. I’ve collected every game he has played and calculated his career numbers on that specific game in the season. For instance, he has 14 first games in the sample, and he has a .346/.424/.615 slash line in those games, good for a 1.039 OPS. In the 21st game that he played in these seasons, it’s a more modest .240/.333/.280, a .613 OPS. Since we’re looking for trends, and to even out the fact that he has less playing time very late in the season (four times played in game #137 and 138, three times in #139, and less moving on after that), I’m going to present this as a rolling 30-game average. Since this is all about how Yadi does compared to… well, the REST of Yadi’s career, I’m also presenting this as a relative OPS (OPS+) using his career OPS as the baseline. Here is his OPS+ as a rolling 30-game average, through X number of games:

That’s illuminating. We wanted to know if his productivity dipped as his playing time increased. The truth appears to be the exact opposite, with some caveats. First, we should probably toss out the giant burst for totals through 140-147. The smaller samples are having a larger effect than anticipated. Second, there is demonstrably more offense league-wide as weather heats up. Consider that in this graph, game #80 is approximately the beginning of July in most seasons. By game #110, the entire 30-game set in Yadi’s rolling average will have been played in swampy 85+ degree temperatures. In fact, you could almost attribute the entirety of Yadi’s burst from games 100 through 120 to warmer temperatures. Still, that only shows that he’s performing as he should- comparably productive in games 1 through 80, then taking on a significant bump in hotter temperatures. All of which is to say that we now know that Yadi’s production doesn’t collapse the deeper he goes into a season. If anything, he enjoys a warm weather bump in productivity.

How about the component parts of production? Specifically, let’s take a look at isolated slugging (ISO) and plate discipline. In this case, I’ll use BB/K for plate discipline. And like the last batch of info, we’ll put it on a scale relative to his entire career. Here’s how that looks:

The BB/K rate collapses over the course of the season, and it’s significant. By late July and in to August, he’s 20% below his career rates, which could certainly be seen as a sign of fatigue. Fortunately, that collapse is offset by an ISO that steadily rises throughout the season. As we saw in the OPS graph, he’s still quite productive later in the season- even more so than he is early in the season, as would be expected thanks to weather effects. In conjunction with the BB/K and ISO data, we now know that the power spike is what keeps him productive.

There’s one final piece of this puzzle- defense. Unfortunately, there’s not a lot of game by game defensive data out there. Even if there was game by game defensive data, catcher defense is... complicated. We have one piece of info in our arsenal- caught stealing percentage. Technically, I could also use passed balls, but the volume is so small- and potentially tainted by official scorers- making it not worth the time. With that in mind, here is Yadi’s caught stealing percentage done in the same fashion as the other graphs- OPS+ scale relative to career, 30-game rolling average.

This graph is a lot noisier with a lot more fluctuation over the course of the season. There are two things we can deduce. First and foremost, he is at his absolute best early in the season, at his most fresh. Second, there’s a marked collapse at the end. Since this is a 30-game average, the dive at the far right of the graph means his collapse is starting between games 110 and 115. Then again, the window doesn’t get really extreme until we get to game #142 or so (CS% from game 112 through 142). Prior to that, it looks like any of the other valleys in his season (see approximately games 68-74, games 84-88, or games 105-115). It’s not exactly a straight line down but there’s at least mild evidence here that fatigue piles on as the season continues. Even taking the later season spike into account (115 to 120), he peaks early in the season and never really hits THAT glorious level again.

In conclusion, his plate discipline slips mid-season and his CS% is a bumpy road from June onward. There’s no guarantee that either of these dips late in the year are fatigue, but they offer at least light evidence of it. It’s all made slightly moot by a power spike, probably weather-related. Add it all up and you have a player whose value- combining all aspects (power, plate discipline, throwing arm)- is fairly consistent over the course of the season. When is power is down, his caught stealing percentage is up. When his plate discipline dives, his power spikes. And at the end of the season when his caught stealing percentage slides, his plate discipline and power spike (or in the case of his ISO, it levels off at the top end of his range). There was, rightly, a lot of handwringing through the years about viable backup catchers and finding someone who could help keep Yadi fresh late in the season. As it turns out, it really hasn’t mattered that much.