A look back at the Roger McDowell era in Atlanta

Unless you’ve been living under a rock, sleeping off a very long playoff hangover caused by a Cubs-Indians World Series, you’re probably aware that the Braves announced that they were moving on from Roger McDowell as they head into their 2017 campaign. McDowell was the pitching coach in Atlanta for 11 seasons. His tenure began right as the streak of division titles ended, and he presided over the resurgent 2010-2013 team, as well as the 2014 decline and the onset of the rebuild over the last two seasons.

There have been rumblings (though I’ve yet to see a source come out and say it directly) that the Braves were looking for a different approach from the pitching coach position than what McDowell had provided to date. For example, take the announcement from Mark Bowman, which is highly laudatory of McDowell’s body of work, aside from its concluding statement:

While Mike Foltynewicz developed into a bona fide big leaguer by the end of this season, the inconsistencies and struggles of two other highly regarded prospects -- Matt Wisler and Aaron Blair -- led the Braves to be concerned about how Sean Newcomb and other prospects would fare on McDowell's watch.

Bowman also notes that McDowell has applied a "tough love" approach, and that perhaps such an approach was considered potentially not as effective for the next wave of vaunted Braves arms. I don’t really have a way of judging that statement — it seems fairly speculative, but I also have no inside information to work with.

In any case, what McDowell’s departure does yield is a complete body of work for him as the Atlanta pitching coach. What it doesn’t do, and what still doesn’t really exist, is a great way to evaluate pitching coaches. It’s the same problem with evaluating managers, but even thornier: at least with managers, you can evaluate the run values of their tactical decisions, however fraught that is. With pitching coaches, all you really have to go on is pitcher performance, where it’s essentially impossible to fully dissociate true talent level from random variation and/or any effect a given pitching coach may have.

With that said, what I’ve pulled together are just some bits and pieces of high-level outcomes I found interesting that occurred throughout McDowell’s tenure. So as to not bury the lead, I’ll just state upfront that I was fairly shocked that McDowell was let go, as the quantitative assessments I could think to make of his tenure reflect on him very favorably. So if you don’t care about that sort of thing, I suppose you could stop reading here.

---

Teamwide pitching outcomes are, in all honesty, hardly a great assessment for a pitching coach. But, at the end of the day, if you are going to judge someone by the overall success or failure of their demesnes, Roger McDowell comes out looking pretty good. To wit, over his 2006-2016 tenure, Braves pitchers were:

• 4th in MLB (and in the NL) in ERA
• 4th in MLB (and in the NL) in FIP
• 3rd in MLB (and in the NL) in xFIP
• 6th in MLB (5th in NL) in SIERA
• 8th in MLB (5th in NL) in K%
• 11th in MLB (7th in NL) in K%-BB%

All of those look like pretty good marks, and the ERA estimators (as well as run-prevention itself) is pretty impressive, especially considering that two of the three teams consistently above the Braves (Dodgers, Giants) play in offense-suppressing parks.

Unsurprisingly, the effectiveness of the Braves’ pitching staffs tanked under McDowell’s watch as time went on. It’s hard to really lay this at McDowell’s feet, though, as the talent level was also highly variable, upended in large part by severe injuries to throwing arms and shoulders. (Astutely, you may here proffer the argument that it was McDowell’s coaching that was somewhat responsible for said injuries; I have no real evidence or basis to either support or deny that statement.)

Looking at the above, I suppose you could make the case that McDowell was axed because his performance was no longer up to snuff, but given the weak state of the roster in 2015 and 2016, I think that’s somewhat of a disingenuous argument. On the other hand, when the pitching staffs were good, they were really good (see 2009-2014), though again, a lot of that falls on the roster construction and talent acquisition rather than McDowell’s coaching.

---

So, how do we evaluate coaching of pitchers, then? I still don’t know, to be honest. But perhaps looking at individual pitchers in different environment could be useful. Specifically, I’m thinking about how well pitchers pitched before, during, and/or after being directly under McDowell’s tutelage. While performance varies from year to year, and pitchers age just like the rest of us, it’s still worth taking a look.

Weirdly enough (or perhaps not weirdly at all, depending on how you want to look at it), only 22 Atlanta pitchers accumulated 200 innings between 2006 and 2016. Of these, only Tim Hudson passed the 1,000 IP mark. (I’m using 200 innings as a cutoff because otherwise you start getting small sample size heavily influencing results, though even 200ish is prone to unfortunate variation, and nine of the 22 pitchers had between 200 and 250 innings under McDowell).

Of these 22 pitchers:

• Four were relievers (Peter Moylan, Eric O’Flaherty, Jonny Venters, Craig Kimbrel), which I’m excluding because, well, reliever variability, and because I set the cutoffs to specifically look at starters. [Down to 18 pitchers.]
• Four have only appeared in MLB as Braves (Julio Teheran, Mike Minor, Matt Wisler, Kenshin Kawakami). [Down to 14 pitchers.]
• Another seven have barely logged any non-Braves innings in their careers, whether that was pre-Braves, post-Braves, or both (Jair Jurrjens, Tommy Hanson, Kris Medlen, John Smoltz, Chuck James, Brandon Beachy, and Mike Foltynewicz). [Down to seven pitchers.]
• That leaves six pitchers with nice before-and-after comparisons: Aaron Harang, Shelby Miller, Javier Vazquez, Paul Maholm, Derek Lowe, and Tim Hudson, who is special because he also had a pre-McDowell Braves season under his belt.
• In addition, we have Alex Wood, who is in strange limbo territory because he has accumulated a fair bit of innings (130) as a Dodger so far.

What I’ve done below is hardly scientific, and just captures the pre-McDowell, during McDowell, and post-McDowell performances of each pitcher across ERA, FIP, and fWAR/200. It also notes the age range for each period of a pitcher’s career.

If you take more than a second to stare at the above, you’ll see it’s certainly suggestive of a few things. But what those things are, well, that probably depends on what jumps out at you the most. Some hypotheses (not conclusions!) could be:

• Long-tenured pitchers didn’t necessarily thrive under McDowell;
• Whatever McDowell was doing, he cursed pitchers that departed the Atlanta organizations, ruining their careers once they were out from under his heel;
• McDowell did a really good job of injecting some spring into the step of older veteran pitchers;
• Older pitchers who benefited from McDowell’s tutelage and went on to seek other opportunities couldn’t replicate their McDowell-engendered success with other pitching coaches, and their careers ended shortly thereafter.

Again, I don’t know if the truth is really in any of those statements above, and the sample is tiny. But still, it’s weird, right? Specifically, you’ve got six pretty different pitchers, each of whom ran aground shortly after leaving the McDowell-coached Braves. Pitcher-by-pitcher thoughts follow:

Tim Hudson

One of my favorite all-time Braves, Huddy had been a consistent 5-win-ish pitcher despite being a groundball guy, something not particularly favored by FIP (and therefore fWAR). The transition to the NL did not go awesomely, as his first season as a Brave (the year before McDowell became pitching coach was one of the worst of his career). While Huddy improved substantially after McDowell became the pitching coach, he generally did not recapture the type of success he had in Oakland.

As with most pitching coach-related stab-in-the-dark pseudo-analyses, there are multiple ways to look at this. One was that McDowell was simply worse for Hudson than other pitching instruction. But the other was that Hudson’s performance was always going to decline as he aged, and McDowell actually kept him pitching pretty well, relative to a "but-for" version of Hudson that didn’t have the same coaching. Without an alternate universe generator, we’ll never know if the latter is true, but it’s something to think about.

Something else to think about (which applies to the pitchers below): check out the pitcher aging curve below, courtesy of Bill Petti and Jeff Zimmerman via Fangraphs:

You can eyeball (and I did the math) that the difference between the average FIP of a guy at Hudson’s pre-McDowell ages and the average FIP of a guy at Hudson’s during-McDowell ages is about 1.00 points of FIP. Now, Hudson changed leagues, and the run environment was very different between the early and late 2000s, but I did some quick math that suggests that across 2000 and 2016, on average, a 1.00 increase in FIP (regardless of league, home park, etc.) contributes to about a 1.9 fWAR drop per 200 innings. Huddy’s FIP under McDowell didn’t rise that much despite his aging, and his fWAR definitely fell, but not quite by as much as you’d expect given that relationship.

Lastly, Huddy was worse upon leaving the Braves and McDowell. Aging was likely a substantial factor; so it goes.

Derek Lowe

I did not include the years where "Derek Lowe, closer" was a thing in the table above. The story is pretty similar for Lowe and McDowell as for Hudson and McDowell, although Lowe was around for fewer seasons, and much older when he first became a Brave. He also directly collapsed under McDowell’s watch, specifically in that star-crossed September of 2011, though his peripherals weren’t actually as bad as his overall run prevention line (yikes, .400ish BABIP-against for the month couldn’t have come at a worse time).

Leaving the Braves and McDowell didn’t help Lowe, though (you could say that for everyone on this list), as he stopped being a useful regular and retired soon thereafter.

(For what it’s worth, given how old Lowe already was when he first became a starter, I think he actually did somewhat worse, after adjusting for park and league, than you’d expect, after becoming a Brave. You can lay that at McDowell’s feet if you’d like, but you’d have to consistently then credit McDowell for Hudson and others.)

Paul Maholm

Unlike the other guys on this list, Maholm was never particularly good (well, maybe 2008-2009). The Braves also didn’t get anything particularly interesting or useful out of him in 2012 and 2013, though that was mostly because a bleh 2013 erased any goodwill from a strong post-Trade Deadline 2012 performance. Most notably, though, Maholm’s career basically ended as soon as he left the Braves. McDowell staving off the inevitable, or somehow birthing a shadow baby to assassinate Maholm’s pitching skill? You decide.

Javier Vazquez

Javy Vazquez might have one of the weirder careers among 21st-century pitchers we’ve experienced so far. He was really good in general, had a ridiculously great career year in his lone season with the Braves, got traded for He-Who-Shall-Not-Be-Named, was absolutely terrible, then went back and had another good season with the Marlins... and then retired at age 36.

McDowell, of course, coincided with that career-best season. If not for Tim Lincecum that year...

Shelby Miller

We all know the Shelby Miller story pretty well by now, as it has huge implications for the future success of the Braves organization. Was McDowell responsible for Miller’s resurgence, or would he have figured it out himself? Insert a shrug here. But, the Diamondbacks definitely broke him after he worked pretty well in a Braves uniform, and paid a very dear price for the right to tamper with his pitching.

Hopefully Miller doesn’t end up like the other guys on this list, i.e., out of baseball after next year.

Aaron Harang

Kind of a mix of the Hudson/Lowe and Vazquez stories, Harang also came over for a one-year stint, albeit at the very advanced (pitching) age of 36. He had a remarkably solid season as a Brave, even though his outings were full of a weird, probably non-replicable phenomenon called "Haranging" where he managed decent FIPs (3.65, 3.76) in low- and medium-leverage situations, and then buckled down to stymie hitters to the tune of a 2.70 FIP in high-leverage ones. (Weirdly enough, this pattern actually holds true over his career, and appears to be driven by a low BABIP in high-leverage situations, though I’m skeptical that it’s a by-the-book-definition "skill" because he can’t seem to do it for seasons on command, i.e., see 2012 and 2013 compared with 2014 and 2015.)

Anyway, Harang’s career was haranging on by a thread prior to his Braves stint, as he was being not-particularly-useful as a major league hitter. Then came a resurgent 2014 that he wasn’t able to replicate in 2015 with the Phillies, and out of baseball he went.

Like I said, draw your own conclusions about these guys, and about Roger McDowell’s work with pitchers in general. Just make sure that if you’re faulting him for anything, you’re also crediting him as appropriate. And if you think he gets no credit for any pitcher’s success under his watch, then he surely shouldn’t take any blame as well.

Also, I would have really loved to add full seasons of Bartolo Colon, RA Dickey, and Jaime Garcia to the above table, as that would increase the sample of the table above by half. Alas, we won’t get to do that. And yes, it does strike me as oddly funny (and funnily odd) that the Braves removed McDowell and then signed a bunch of veteran pitchers to fill up their rotation. Huh.

---

One more thing, though...

Per the rumblings at the time McDowell was let go, the idea was that McDowell wasn’t gelling well with the young pitchers the Braves acquired. The sample size here is tiny, but there’s a bit of food for thought here as well. Specifically, the Braves have had three starters who were given FVs (future values) by Kiley McDaniel on Fangraphs (who now, of course, works for the Braves): Matt Wisler, Mike Foltynewicz, and Shelby Miller:

• Miller got a 60 FV grade in November 2014. A 60 FV corresponds to about a 3-win starter. Shelby Miller was pretty much that for the Braves, so you could read this as McDowell "making good" on Miller’s perceived potential.
• Wisler got a 55 FV grade in January 2015, aka, about a 2.5 WAR/200 starter. Instead, Wisler has been pretty bad, falling short of 0.5 WAR/200 so far in his career. The path towards success isn’t always as straightforward as immediately hitting your FV level, but still, it’s disappointing.
• Foltynewicz also got a 55 FV grade in October 2014. He’s still developing, and while his 2015 was very forgettable, he definitely improved in 2016. Will he hit that 2.5 WAR target, or perhaps exceed it? That remains to be seen. His 2016 was definitely a step in the right direction, though.

So we’ve got Miller-under-McDowell realizing his potential, and then being moved for a once-in-a-lifetime-rebuild-in-a-box-just-add-water package. We also have Wisler falling short of expectations, so far, and Foltynewicz proceeding unsteadily, though with some case for optimism. Was that not enough for the Braves, and that’s why they let him go? Could be, but only the Front Office, and perhaps McDowell himself, know for certain.

---

Coda: You can do this same analysis for Chuck Hernandez, the guy the Braves hired in McDowell’s stead. Hernandez has had three major league pitching coach gigs: the then-California Angels (1993-1996), the then-Devil Rays (2004-2005), and the Tigers of 2006 through 2008. I scanned through the stats, and the before-and-after numbers for the pitchers he worked with are not flattering, though he was around for Chuck Finley’s peak, so that’s something. Chuck Finley also retired after a 4.4 fWAR season, which is also something.

You’ve read over 2,500 words already, so let’s end on a poll.