Here’s a bold prediction: at some point in the next decade, we will have a pitcher who will break the record for most games started in a season. The record is currently 75, shared by Pud Galvin (1883) and Will White (1879.) Galvin threw 656 ⅓ innings, while White threw 680. The future record-holder will probably throw fewer than 150 innings.
They’ll be an opener.
Enough about the past and future, let’s instead focus on the recent past and the near future. White Sox manager Will Venable first deployed Grant Taylor as an opener this year on April 3. The White Sox won that game, to improve their early season record to 2-5. Venable had Taylor open again the next day – they won that game, too. Despite the early success, Venable’s tactics were met with skepticism by the fan base writ large. It was a mixture of “nerds are ruining the game” (balderdash) with a scoop of “anything the White Sox do is inherently a bad idea because it’s the White Sox doing it” (OK, understandable.)
The proof has since been found within the pudding. When the White Sox use any of their three best high-leverage arms to open a game, their record is 13-5. Openers work — when you have a good opener:
Chris Murphy is not on the level of Taylor, Hudson or Newcomb. So when he gave up back-to-back home runs in the first inning of Tuesday’s game to put the White Sox in a 3-0 hole, it’s totally reasonable to ask why he was opening. What isn’t in question is whether openers in general are a smart decision.
They are.
What’s the logic behind using openers in the first place? Like most ideas which lead to analytic nerds getting their heads dunked in toilets by the likes of Hawk Harrelson, there’s 100+ years of consistent data to back up the concept. Namely, starting pitchers see a precipitous drop in their effectiveness whenever they are facing a batting order for the third time. By using an opener, the opposing lineup’s best hitters do not get a third chance to see the bulk pitcher until late in the game, as opposed to the middle innings. And that’s if they get to see him a third time at all; when the strategy works as intended, the “starter” can tear their way through the middle innings and still hand over the ball to a late-inning, high-leverage arm.
This is where the anti-analytics people come out of their caves, wearing their leopard-fur loincloths and waving around crude hand tools in a blind fury, shouting, “Walter Johnson didn’t need an opener” like they’ve spotted an airplane flying over Pangaea. This is the only context where somebody can demand that Erick Fedde pitch as effectively as Johnson and receive widespread support. I’ll concede that Johnson did not use an opener during his career, and still did quite well for himself.
Let’s see how the OG Big Train did each time through the order:
Here’s the thing: A tired Johnson was still going to pitch better than any reliever. That’s why he has a career bWAR of 168.4. I am not saying that Johnson needed an opener. What I am saying is that any time a major league hitter gets a third chance to see the same pitcher in a game, they will always do better on average (I’m also saying that Johnson is better than Fedde, by the way). Openers aren’t a case of analytics running the game, but a case of adapting to facts that have been staring us in the face since the spitball.
If you want a more modern example, let’s look at Tristan Peters vs. Gavin Williams in the White Sox 7-3 victory over the Guardians on Monday. His first at-bat was a one-pitch pop out on a high inside fastball. His second appearance was a five-pitch strikeout with two runners in scoring position, featuring two swings over the top of breaking balls that were closer to the dirt than they were to the strike zone.
Peters seemed overmatched. In the interest of that long-dead ideal known as “full transparency,” it was that at-bat which caused me to have a full-blown crash out on main:
The third at-bat, of course, was that 18-pitch epic that became the turning point of the entire game. Peters had now seen the fastball, curveball and sweeper. That experience allowed Peters to foul off those pitches with metronomic efficiency. By the way, it’s not like Williams ever gave in; I don’t think he missed his spot on any of his first 17 pitches. Peters simply wouldn’t be beaten. It was an at-bat you could only have in your third at-bat against the same pitcher.
Now, a modest proposal. The White Sox starting pitchers are gassed. It’s undeniable. That’s why Venable should use an opener for every game from now until the end of the season.
Yes. Every game.
I’ll go even further – most managers pull their opener after their first five or six hitters, reasoning that their “starter” can facing the bottom third of the order three times without significant risk. Last I checked, though, the bottom-three hitters in every major league lineup are still major league hitters. The White Sox also have a bullpen full of starting pitchers past (Newcomb, Hudson) and future (Taylor.) Let them face the entire lineup once through. Pass off the baton after a clean nine batters faced and let the bulk arm come in for the heavy work with a totally clean slate. If the club’s starters can only give you four innings, those innings should be four through seven, not one through four.
After all, No. 2 seeds only need to go 11-5 in October to win a World Series title.
Numbers don’t lie.
