Where have all the walks gone?

Whatever happened to ball four? The walk? The free pass? The base on balls? It’s already something of a persona non grata in baseball. If a hitter walks, we sorta pretend it didn’t actually happen (e.g., “He went 1 for 3 with a walk.”) But it seems that lately, we’re seeing fewer of them. On a recent episode of my favorite baseball podcast, Effectively Wild, co-host (and um, my boss) Sam Miller pointed out that pitcher walk rates that are average now might have been among the league leaders a decade ago.

For example, Gerrit Cole and Carlos Carrasco have (as of this writing) have each walked 1.9 batters per 9 innings pitched. This puts them (among starters qualified for the league ERA title) in 22nd and 23rd place, respectively in MLB. In 1996, the two men would have ranked 6th and 7th on the same list with those kind of walk rates.  There are 94 pitchers qualified for the ERA title. The 47th guy (halfway down the list), Chris Heston of the Giants, has walked 2.5 batters per 9 innings. In 1996, that would have been good enough for 20th place, only a quarter of the way down the list.

In fact, here’s a graph showing the percentage of plate appearances that ended in walks league wide in each year from 1993 to 2014.

In recent years, strikeout rates have spiked, leading people to wonder about the cause of the strikeout scourge. Turns out that it’s not a new wave of pitchers who are striking everyone out, but a new wave of batters who are prone to strikeouts. In fact, the biggest contributor seemed to be that hitters were all too willing to take strike one, and that seemed to set up the rest of the at-bat for strikeouts.

It’s tempting to think that if strikeouts are up, walks are down. It’s not quite that easy. Walks and strikeouts actually don’t tend to correlate with each other. But where did all the walks go?

Warning! Gory Mathematical Details Ahead!

Hypothesis 1: Strike one strikes again!

When looking at the strikeout scourge, one pattern that became very apparent was that over the past two decades, hitters had become much more passive on the first pitch in an at-bat. Here are swing rates on the first pitch of an at-bat over the past two decades.

If the prevailing philosophy has become “I won’t swing until you throw me a strike,” then pitchers, as a whole, have responded with a resounding “OK, if you insist.” Called strike rates have gone up on 0-0 counts and more at-bats in general are starting 0-1. That means that the hitter is a third of the way to a strikeout, and if he’s busy striking out, he can’t walk.

Here’s what happens to walk rates in plate appearances that start 0-1 vs. 1-0 (plate appearances where the first pitch was in play are not counted).

Strike one cuts the chances of a walk happening in this at-bat by more than half. Oddly enough, one of the causes of the walk shortage might just be that hitters are taking too many pitches early in the count.

Hypothesis 2: It’s the Moneyball backlash

Outside of religious texts, I doubt that there has been a book more (mis?)interpreted than Moneyball. In the book (and the movie) A’s general manager Brad Pitt, erm… Billy Beane comes to the realization that being able to get on base, mostly in the form of walking, is under-valued and that the A’s can build a team around guys who walk a lot because at the time, no one was appreciating how much value walks have. Walk rates did hit something of a local peak in 2009 after a sharp decline at the beginning of the oh-oh’s, but since then, the bottom dropped out of the market. Maybe the “everyone should walk all the time!” meme just got old?

Was there a point where teams were starting to specifically bring in players who were good at walking? To test that, I looked at all hitters who retired (or were politely asked not to come back in one year) and their walk rates before they left and then the players who made their debut in the following year and their walk rate as rookies. Perhaps the differences over the years can be explained by new types of pitchers (or hitters) coming into MLB?

Here are the year-to-year differences between pitchers going out and pitchers coming in over the years:

It’s mostly random noise in terms of patterns, but we see that the pitchers who replaced the retiring players tended to give up lower walk rates. This shouldn’t surprise anyone on the pitching side.

Here’s the same graph for hitters:

What’s interesting here is that rookies also tended to have lower walk rates than the guys that they replaced. Rookies seem to want to put the ball in play. But we see some definite patterns in the magnitude of that shift from year to year. At points in the past, the new guys could be counted on to be about equal to their predecessors in terms of taking a walk. Over the last half-decade though, teams seem to be focusing more and more on guys who don’t want to walk and weeding out the guys who do.

So, maybe teams are moving away from walks. It doesn’t look like they were ever prioritizing walks specifically, but …

Hypothesis 3: What’s a 3-ball count for?

They don’t cover this in that one song about the old ball game, but it’s 1-2-3-4 balls you’ve walked. The laws of mathematics say that to get to a fourth ball, you have to have a third. Three ball counts are an interesting thing, especially 3-0 and 3-1 counts. The batter has a little bit of wiggle room. On 3-0 especially, he can sit back and wait for one specific pitch in one specific spot, and if he sees it, he can unload on it. If not, he can just let it go by. If it’s strike one, he has two more to play with. On 3-1, it’s some of the same calculus. On the flip side though, if players are trying to work a walk, they can be much more passive on these counts.

Here’s a look at how often hitters have swung on those 3-1 counts over the years (the 3-0 graph follows the same pattern, just with fewer swings).

As time went on, swing rates on 3-1 pitches fell, until about 2009 when they rebounded. It looks like sometime in 2009, someone flipped and switch and said, “Oh yeah, I guess we should swing more.”

Here’s how they did on making contact:

More swings, less contact. Seems that hitters felt the need to take a big swing. And there’s room for that on 3-1 because you know you’re going to get something to hit, and if you miss, it’s not the end of the at-bat. But it shows a shift in what hitters are thinking. 3-1 isn’t a chance to work a walk. It’s a chance to do something big.

With offense in general down, it’s hard to know whether it’s because of the lack of walks or maybe it’s the fact that hitters know that offense is down and just feel the need to play hero. A home run adds a run to the board and when run scoring is down, one run is worth a lot more and it’s worth taking a big shot.

Walk? Run!

Baseball has had a long and complicated relationship with the walk. Walks are undeniably positive events (for the hitting team), but they have always had the reputation of… well, it’s just a walk. It’s not that walks are the greatest thing in the world, but they certainly do have value. Like everything else, that value should be put in its proper place. But right now, it looks like the walk shortage is not just some random fluctuation. It looks like we’ve entered a period of time where batters, as a whole, are engaging in behaviors that make walks less likely, although not necessarily the same ones that make strikeouts more likely.

The trend toward taking the first pitch is certainly contributing to both, but it looks like more and more a walk is an outcome that players are steering away from. The three-ball count has become a time to take a big shot at a home run. And it seems that teams are looking for something other than the guy who can take a base on balls. And like the strikeout scourge, it seems to be mostly on the batters. So if your goal in life is to bring more walks back to the game, the intervention is going to have to start with the hitters.