Friday, October 21, 2022

VERLANDER & THE TOP 50 STARTING PITCHER SEASONS OF THE 21ST CENTURY

We've been swamped and sidetracked in the past ten days, and owe you more entries in the "second half" historical survey--they'll resume tomorrow. 

But a friend wrote about the dominant performance of the seeming ageless, apparently indestructible Justin Verlander in Game 1 of the ongoing ALCS between the Astros and the Yankees, and chimed in with a question about where Verlander's excellent 2022 comeback season ranked in "recent memory" in terms of starting pitcher performance. 

And so, sans further ado, here is some data related to that...data that reminds us rather sadly about the passage of time and the all-too-common fungibility of starting pitchers. But the date also allows us to relive some tremendous seasonal achievements that all-too-easily slip away from our memory banks. Here we go:

We'll do this from bottom to top, in groups of ten. These are ranked by ERA+; we could get a totally different list order by using OPS+ (batter vs. pitcher hitting data normalized to league), but that has a less direct correlation to runs allowed, so we'll buck our own tradition and cleave toward analytical orthodoxy.

Many folks here still pitching--Burnes, Manoah, Noal, Kershaw, Sale (just barely), Ohtani, and (of course Verlander). In this group there is very little to distinguish one performance from another, particularly when you peruse the key rate stats (raw ERA, H/9, HR/9, BB/9. SO/9). Odd to see what we might think of as a non-descript Randy Johnson season here (#41), but that might be based on a lingering connection to W-L record. Note, though, that the only other pitcher in this group with a hard luck WPCT--Felix Hernandez--won the Cy Young Award in 2010.

We also see Shohei Ohtani's fine year in the context of the 21st century--excellent, but not earth-shattering. Some of that is due to the fact the starting pitchers--even the elite ones--are simply pitching fewer and fewer innings per year. Our table at right shows how that has changed as we enter the third decade of this volatile century. 

Again, the 31-40 group is tightly bunched, but there is more variation in the /9IP stats. Verlander's 2019 season shows up here at #31 with a tremendously dominant H/9, but he was very susceptible to the long ball in that freakishly homer-happy year. The rate stats for Derek Lowe (#39) twenty years ago look downright quaint in comparison to what we see elsewhere. Youthful greatness snuffed out is represented by Mark Prior (#33) and Jose Fernandez (#40). Still active from this list: Verlander, Ryu (just barely), Kershaw, Scherzer, Strasburg (just barely), and Sandy Alcantara (who has an outside shot at the NL Cy Young Award).

Very few active pitchers to be found in the 21-30 group, who cluster in the first decade of the 21st century (which seems like a radically different time, now, doesn't it?). The only active pitchers here are Gerrit Cole (#23), who will start the pivotal G3 of this year's ALCS against his former team, and Dylan Cease (#29), the White Sox righty who had a breakout year in 2022 (and needs, but won't get, a mound-mate whose last name is Desist). Cease and Clay Buchholz (#22), the mercurial "magic arm" of the Red Sox back in the Theo Epstein Era, are two guys who are noticeably wilder than their counterparts on this list. 

In case you're wondering, the minimum number of IP to appear on this list is 140, which is how Roy Halladay (#24) made it onto the list. On this list twice is Randy Johnson (#21 andn #27) for back-to-back years with 300+ Ks and plenty of wins (an aggregate 40-13 record).

And the Big Unit cracks the list one more time as the 11-20 segment takes us to ERA+ values exceeding 200. (We eyeball it here, but it seems highly likely that Johnson's 2000-02 seasons mark the only time anyone struck out 300+ batters in three straight years. (A quick check of Nolan Ryan's career stats indicates that he's actually the guy who did it first, from 1972-74.)

Pitchers who are still active in this segment include Sale (just barely), Greinke (ditto), Kluber, Kershaw, Hendricks, Trevor Bauer (actively in suspended animation), and Julio Urias, who we're figuring has the inside track on the '22 NL CYA. Zack Greinke's 2009 season is an outlier that we've discussed before; excellent HR prevention and an elevated strand rate allowed him to produce numbers that look better than they really were, offsetting his 48th best H/9 value.

And now: the Top Ten, as it stands after the 2022 season:

There's "our man" Greinke again, at #4, with his most brilliant season (2015, before he became a truly wandering man-boy). Several short seasons push their way onto this list: Clayton Kershaw's otherworldly but injury-shortened 2016 (#2), and the forgotten Rich Harden, who gave us just a brief glimpse of what might have been. (Imagine being traded during a season when you're compiling numbers like that...)

And. of course, there's Pedro Martinez, whom we failed to mention when he showed up at #14...because we knew we'd be seeing him twice here in the Top Ten--including what is almost certain to be the pitching "season of the century" (#1) that he rolled right out of the box in 2000. That OPS+ value is not a misprint, even though every time one looks at it, one swears that it must be.

And, at #5, there's Verlander, with a comeback season for the ages at the age of 39. (Only the embattled Roger Clemens, at #3, was better at a more advanced age.) He joins three other pitchers at this rarified level who are still active: Jacob deGrom, Jake Arrieta (just barely), and Blake Snell.

Before we sign off, here are some charts that show the distribution of these years by decades. 2022 produced six seasons now in the 21st century Top 50, making it the most populous year thus far. New entries to the list have been coming at an elevated rate since the "Launch Angle" era thrust itself upon us; 19 of the top 50 starting pitcher performances currently on the list come from 2017-2022...and that's with 2020 disqualified due to its "part-season" status due to the pandemic.

And how do our five segments (1-10, 11-20, 21-30, 31-40, 41-50) fan out across time? The chart at left gives us a look. The 2010 decade remains the most numerous, and has more than half of these top performances in all but one of the ten-game units. 

When we break that data into halves, we see that the 2020 pitchers are finding room for themselves primarily in the bottom half of the list. Verlander's stellar work in '22 put present-day pitchers into a slot in the Top 10.

One final chart--this one looks at the correlation of H/9 and BB/9 and its relationship to the raw ERA turned in by the pitchers on the Top 50 list. 

This chart operates a bit like a QMAX matrix chart, assigning two scores to each season based on how good each pitcher's hit prevention (H/9) and walk prevention (BB/9) was, and then assigned a slot in the matrix based on ranges (as displayed in the table at right. 

Because these are all such top-flight performances, the range of difference is not nearly as pronounced as it'd be if we were working with all of the starters in the 21st century, but note that the groups with the best combined hit and walk prevention (in the boxed area in the upper left of the diagram) is notably better than the groups with the more modest values for H/9 and BB/9. Despite what still persists as "received wisdom" among the original "neo-sabes", hit prevention still matters.