Baseball Keeps on Breaking the Same Record

Earlier, Travis Sawchik wrote about the struggles of the middle class in this free-agent market. Sawchik called attention to those players predicted by the FanGraphs community to sign for under $45 million, and he found that the actual terms have been significantly lower than expected. Sawchik highlighted eight players in the group who have done better than the forecasts, despite the trend around them. Something those eight players have in common? They’re all relievers.

For all the talk of how slowly the free-agent market progressed — and, I suppose, continues to progress — relievers have seemingly been the exception. Relievers have come off the board in droves, and even back around the winter meetings, when all anyone could think about was Giancarlo Stanton and Shohei Ohtani finding new teams, relievers were finding contracts left and right. There was some obvious market enthusiasm, and to this point, free-agent relievers have signed for more money than free-agent starters. There are 24 relievers who have signed for multiple years, and that doesn’t count Mike Minor, who’s going to be tried as a starter again. One market is only one market, but still, what’s going on? It’s actually pretty simple to understand. You might even already know the answer.

Here’s a plot I’m sure you’ve seen in some form. Going all the way back to 1901, here is the league-wide percentage of all innings thrown by starting pitchers.

It’s a steady drop, and, in 2017, baseball reached a new low. Last season, starters threw just 61.9% of all innings, the result of teams exercising caution. Caution about injury risk, and caution about in-game over-exposure. You hear more than ever about the times-through-the-order penalty. This isn’t anything new. Relative to even the somewhat recent past, the starters of today are being babied, at least if you ask some of those old-timey workhorses. More of the work is falling to the relievers.

So now let’s think about that in a different way, with the help of the Baseball Reference Play Index. We’ll keep going back to 1901, even though I don’t care about those early years, for these purposes. I have three images to show you. First, for every year, here is the average number of position players per team. This counts them all the same, whether they played every day or pinch-ran just one time. All I care about is a player being rostered.

Nothing has really changed for a while. Sure, things have probably changed somewhere under the surface, but just looking at the overall picture, the average team last season used 22.5 position players. The average team in 1980 used 22.5 position players. The average team in 1953 used 22.5 position players. There was a big recent drop in 1994, but, that one is self-explanatory. This line has held steady. Nothing to see here.

Now, I mentioned a record in the headline. About that: In 2013, baseball reached a new all-time high in average number of pitchers per team. In 2014, the record was broken. In 2015, the record was broken. In 2016, the record was broken. And, in 2017, the record was broken. I imagine it’ll be broken again in 2018. There are two types of pitchers, as long as we’re going to oversimplify. Here are the average number of pitchers per team to have started at least one game.

Again, there’s really not much. There are some rises and falls, but the overall average over the past decade works out to 9.9. For the decade before that, 9.8. For the decade before that, 9.9. There’s a useful reminder in here that every single team should expect to need several starting pitchers. It’s not just about the opening-day starting five. But that’s not exactly a new message. Every team is probably going to need reasonable depth. Not every team is in possession of reasonable depth. So it goes. This all leads up to the obvious final plot: average number of relievers per team. The general definition is that these guys didn’t start even once. As a consequence, I’m missing out on the relievers who got random starts on bullpen days, but I’m not worried, because those games are infrequent.

And there’s the explosion. This follows perfectly from the first image, up above — as relievers have been handed more of the work, more relievers have been needed to do it. Over the past decade, teams have used an average of 13.0 relievers. The decade before that, the average was 10.5. The decade before that, the average was 8.2. Last season alone, the average was 14.7, up from where it was in 2016, which was up from where it was in 2015. Teams are increasingly shuffling through more relievers. You see guys going between Triple-A and the majors all the time. The scale of churn is unprecedented.

This requires that teams be more creative with the disabled list. This also requires that teams have a greater share of pitchers on the 40-man roster. A possible consequence of that is that position players are increasingly asked to be versatile, which might be why we’ve seen good hitters who don’t defend well finding limited markets. Mike Moustakas, for example, isn’t known for his versatility in the field. But that’s just a theory, and this isn’t a post about Mike Moustakas. This is a post about the demands on the bullpen.

Teams are calling on more relievers all the time. At any given moment, there are only so many good relievers to go around. So it makes sense you might see greater urgency to acquire the relievers who are well above-average. And beyond that, just as another consequence of relievers shouldering more of the load, you have relievers expected to throw more innings in each game. You also see a reluctance to use relievers too often on back-to-back days, and especially on back-to-back-to-back days. So bullpens now need to be deeper. There’s hardly such thing as a throwaway role anymore. Competitive teams want to be really strong with their back four or five, which requires investment. That can be even more greatly important in the playoffs, where managers have gotten aggressive with bullpen use. Just about every bullpen job now is important. That importance is reflected in growing salaries.

So relievers get to benefit. That has to come at someone else’s expense. Maybe it’s being taken out on back-of-the-rotation starters and positionless hitters. That’s something for another day. For this day, we witness the bullpen explosion. A reliever no longer has to be a closer to make good money. This is because teams need more good relievers than ever. I don’t know when this trend is going to top out, but it could still be some time.