Walking, Fast and Slow
When I was in Little League, it was a common refrain of coaches that not all walks were created equal, and that some were particularly egregious to permit. That is because walks, like all plate appearance results, do not exist in a vacuum, but rather are situational. While issuing a walk in just about any situation is displeasing for a pitcher, certain walks grate more than others. Lead off walks? Absolutely not. No out walks? Just as bad. Two out walks? What are we even doing here?
It’s been a long time since my 11 year-old athletic peak —walking batters left and right— but I find walks still have varying impacts and feels, all the way up to MLB. Generally speaking, the disparity in walk values is largely due to the in-game situation, and often has a tidy statistical explanation like win probability added. However, what follows is a brief overview of walks, not through the lens of their situation, but rather how long they took to come about. Specifically, the number of pitches that led to any given walk.
As an overview of walks in 2024, I made use of Baseball Savant data and later FanGraphs data. The following charts break down how many pitches all non-intentional 2024 walks included.
So, 6-pitch walks were the most common variety of walk in 2024, though 5-pitch walks weren’t so far behind. Below the same data is presented as a table.
In sum, the distribution of walks is concentrated among four-six pitch walks and long-tailed to the right; >70% of walks in 2024 occurred on six pitches or less. While all cumulating in the same result, a four pitch walk has a very different feel than even a six pitch walk, I would contend.
Which begs the question: would you rather watch your favorite team’s pitcher offer a free pass on four pitches or eight—reaching that undesirable end faster or slower? On one hand, a four or five pitch walk doesn’t exactly inspire confidence or illustrate control of a situation. The next batter up is probably taking his first pitch. However, an eight or nine pitch walk is likely dispiriting and additionally inefficient in terms of pitcher workload. That said, Blake Snell just cashed in in spite (because?) of a proclivity for “4 2/3 inning, 5 walk, 0 ER, 7K, 102 pitch” type pitching lines.
Now, baseball fans today are nearly as interested in the manner in which an outcome came about as they are in the outcome itself. A double that comes by way of a 110 mph rocket leaves us considering the outcome differently than a weak flare which inexplicably finds grass and results in a hustle double. An eight strikeout start has a different feel if the pitcher who authored that start was throwing two mph off his season average fastball velocity.
Given this emphasis on the how versus the what, I chose to briefly consider the manner in which pitchers permit walks. Very simply, I counted all those that were issued on five or fewer pitches (here calling it Five Pitch BB%, or 5PBB%) in order to see how much “fast walks” related to all walks.
Perhaps unsurprisingly, Five Pitch BB% very much correlated to BB%.
However, the correlation is far from perfect despite five pitch walks being, well, walks. Plenty of pitchers in 2024 had a very low 5PBB% relative to their overall BB% or vice versa. Below are those pitchers with the highest and lowest fraction of walks that were of the sub-six pitch variety.
So, one might suggest upon seeing this that when the Edwin Díazes and Jack Kochanowiczes of the world didn’t quite have it from a command perspective in 2024, they really didn’t have it. Those two led the way among 2024 pitchers with the highest fraction of their permitted walks that came on five or fewer pitches. Meanwhile, Carlos Estévez, Tyson Miller, and Anthony Molina —who at a glance didn’t have altogether vastly different walk rates than Díaz and Kochanowicz— almost always made a hitter work for their walk.
Does this mean anything though? Are “slow” walks any better than “fast” walks in the same way 110mph exit velocities are favorable relative to flares? I didn’t spend too much time investigating, but taking a cursory look it doesn’t appear as though “fast” walks are terribly predictive of future walks, or better than regular old BB% anyway. Below are 2024 first half BB% and 5PBB% versus second half BB%.
While it seems as though both first half BB% and 5PBB% correspond some to second half walk rates, BB% appears to be a better fit (as measured using adjusted R-Squared values). With that, 5PBB% may not be of prodigious predictive value, but still makes for something somewhat fun to consider. Personally, I would rather watch an eight pitch battle than a four pitch blip.
