Those Rascally Soft-Tossing Lefties

On August 13th, the Cardinals squared off with the Nationals in the first of a four-game set. Miles Mikolas took the hill for the Cardinals, who were 2.5 games back of the second Wild Card (5.5 back of the Cubs). Opposing Mikolas was Tommy Milone, a soft-tossing lefty who has bounced around the majors the past few years after some decent work with the A’s in the early 2010s. Well, when a low-velocity lefthander faces the Cardinals, you know what comes next. That’s right- fans start talking about how the Cardinals can’t hit soft-tossing lefties. There are all kinds of versions of this meme- the Cardinals can’t hit rookies, the Cardinals can’t hit bad pitching, the Cardinals can’t hit fifth starters, the Cardinals can’t hit junkball pitchers. The truth is, this isn’t a Cardinals thing, not even close to exclusively. If you listen to baseball fans across the country, the same belief comes through loud and clear. No one, absolutely no one, thinks their team hits bad pitching well. Something’s clearly amiss. Everyone can’t be right- otherwise the pitching wouldn’t be bad, you see.

I’m not going to attempt to explain the psychological phenomenon that makes it feel like things aren’t going as well as they should be going, that things are unfair, or that you’re getting unlucky. This is a baseball blog, after all, not a psychology dissertation. I’m going to focus on something far more prosaic. Are the Cardinals actually bad at hitting soft-tossing lefties? I designed a study that I think should help. First, I took a list of all the lefty starters that have thrown a pitch in the majors in the last five years. I bucketed them into three groups- hard throwers, average throwers, and soft tossers. Not that it’s all that relevant for this article, but I defined hard throwers as anyone sitting at 92.5mph or above and medium velocity as 90-92.4mph. This left me with 117 player-seasons out of 421 overall seasons by lefthanded starters with 80-handle averages on their fastballs (I counted each year as an independent sample). Average fastball velocity by lefties over this time was just over 91mph, so the cutoffs seemed reasonable.

After getting rid of the people who throw real major league fastballs, we were left with just the crafty lefties. Some notable names on this list are Mark Buehrle, Bruce Chen, late-career Barry Zito- it’s basically the list you’d expect. With this list in hand, I worked out the wOBA allowed by this cohort of pitchers in all the games they pitched in the last five years. Collectively, they allowed a wOBA of .327, comfortably above the major-league average of roughly .315. That makes sense- it would be cause for confusion and probably alarm if this group of pitchers weren’t worse than the overall league average. Why did I use wOBA in this study? Well, mainly because it’s my go-to all-purpose stat. Why is that? I’ve discussed it a few times, but essentially I find it to be both intuitive and comprehensive. By assigning every event a run value and then essentially just adding them up and dividing by plate appearances, you avoid a lot of the complexities that go on under the hood in some of the more complex all-in batting stats. At the same time, however, it weights the relative value of different events correctly. On-base percentage values a homer the same as a walk, and slugging percentage values two singles and a walk the same as a double. Neither of these are right- wOBA is just an elegant way of correctly weighting them. That justification aside, I then bucketed out how each team fared against this same group of pitchers. Feel free to skip this long list if that’s not your thing, but I might as well show my work:

The Cardinals sit in a comfortable 10th place against these soft-tossing lefties. Case closed, right? Well, obviously not. This isn’t anywhere close to a controlled study. The Nationals, for example, finished first in our sample; but they also featured dynamic offenses in each of the last five years. The Padres are second-to-last, but they’re terrible against all types of pitchers, not just this subset. We need more context. Instead, we need an index that handles how teams due against soft-tossing lefties relative to their own baseline. Luckily, it’s pretty easy to find each team’s baseline. We already have the overall MLB wOBA, after all. As I stated above, the league average wOBA is .315 over this five-year period, and our cohort of pitchers allowed a .327 wOBA in the same timeframe. Thus, a dead-average major league team would hit them 12 points of wOBA better. Let’s see how each team fares relative to its performance in all games:

As it turns out, the Nationals really do mash junkballing lefties. It’s not just an artifact of their above-average offense, though they do indeed have an above-average offense. Similarly, the Padres really are more lost at the plate than normal against lefty guile. (Quick aside- it’s SO hard to come up with synonyms to describe lefthanders who don’t throw very hard. I feel like I’ve run through all of them about ten times. It’s easy enough when you’re writing about a player- you can just use his last name every time. A class of players is a lot harder, though. Something to work on!) There are exceptions, though. The Blue Jays fare well by raw wOBA, but don’t do very well relative to their already-robust offense. The Mets are a below-average offensive team overall, but they came alive in this sample, with one of the largest improvements in the majors.

I wanted to come up with a stat for this. It’s partially because I love creating stats, but also to better display the spectrum of performance that each team has against the tender underbelly of major league pitching. My first thought was to call it the Soft Tossing Lefty Overperformance/Underperformance Index Statistic, or STLOUIS. Even for someone who abbreviates as shamelessly as me, though, that was a bridge too far. Instead, I took a hard left turn (pun most definitely intended) and settled on a more descriptive name- the Buehrle Number. Inspired by the most iconic soft-tossing lefty of the 21st century, a team’s Buehrle Number is simply its improvement in performance against this sample of pitchers minus the league average improvement. The Cardinals, for example, improve by 13 points of wOBA. The major league average is a 12 point improvement. Thus, the Cardinals have a Buehrle Number of 1.

The statistic doesn’t stop there, however. While we’re getting into small sample territory, we can look at a Buehrle Number for any set of hitters. NL and AL averages? No sweat. Divisions? Sure thing. Most interestingly, though, we can look at individual players. Are the samples probably too small to be meaningful? In a word: yes. In thirteen words: absolutely they are, but who cares, it’s still fun to look at them. So let’s see it. Here are Buehrle numbers for the eleven Cardinals who saw at least 50 pitches from eligible pitchers over the last five years:

I want to point out again that this is silly. The sample sizes on this are ludicrously small. Tommy Pham has 43 plate appearances in the Buehrle group. That’s incredibly far from approaching meaning. Matt Carpenter has the most plate appearances with 138. None of these are big enough for meaningful takeaways. Honestly, the team samples verge on not big enough to matter, and the Cardinals almost certainly aren’t meaningfully different from league average.

There are further things we could do with the sample. We could control for the exact quality of opposition that each team faced, because some of these pitchers are good and some are bad. That requires more computer programming than I feel comfortable doing, however, and I suspect that the results wouldn’t change in a statistically meaningful way. We could run it by FIP, ERA, batting average- any kind of stat that floats your boat. We could look at soft-tossing righties, or rookies, or pitchers with ERA’s above 5. Those are all studies for another day, however. What we have today is some evidence that sometimes our own beliefs aren’t all that true.

If you’re a Cardinals fan (and if you got this far, I assume you are), this is kind of a dull result. The Cardinals improve against soft-tossing lefties. They don’t improve that much, but no one in the league actually improves that much. Want an annoying summing-up of how much this matters? You’re in luck. Per nine innings facing soft throwers, the MLB average wOBA improvement is worth almost exactly .33 runs. Every third game, more or less, you’d expect an extra run. Over a full season of facing nothing but junk, that would be worth just over fifty runs. It sounds like so little! You’d think these guys would get absolutely crushed. The thing is, there just aren’t that many pitchers who are truly atrocious but still face a lot of batters. For every few 2015 Brad Mills (79 pitches, .589 wOBA allowed) there is a 2018 Brent Suter (1638 pitches, .320 wOBA allowed). The terrible ones don’t last long enough to make a huge impact on the sample.

If you’re a Phillies fan, I don’t know what to tell you. That thing you think is a monster under your bed? It really might be. Your team really has hit bad lefty pitching significantly worse- 29 points of wOBA worse than league average, in fact. Cheer up, though. Most of that was the old, bad Phillies. This year’s vintage barely resembles some of the teams in my sample. If you’re a Nationals fan- look man, it’s been a rough few years. Here’s the good thing, though. Unlike Jonathan Papelbon, your batters aren’t chokers, at least when it comes to facing pitchers they should be able to hit. One final note for all of you John Gant aficionados out there- none other than the man himself, John Gant, is second on the Cardinals in wOBA against crafty lefties, thanks to his home run off of 2018 Gio Gonzalez (both of his plate appearances in the sample came off of Gio in the same game).

The Cardinals won the game handily, by the way. Milone went 4 ⅓ and allowed ten hits and a walk while striking out only one. He even hit a batter. He got out of it with only two earned runs, more due to lucky sequencing than anything else. The myth of the soft-tossing lefty rolled on.

All statistics current as of games of Wednesday, 8/22.