Matt Williams and What We Don’t Know

Their season ended Sunday, a week — if not a month — shorter than had been planned, and the Nationals jumped right into their off-season by firing reigning Manager of the Year Matt Williams. Williams has been a lightning rod for criticism since the Nationals started to falter in the middle of the year so it’s no surprise that he was relieved of his duties, especially since GM Mike Rizzo refused to say he’d keep Williams past the end of the season despite numerous opportunities to do so. Williams will go down as a bad manager and his tenure will be a negative on Rizzo’s resume. Still, from an outsider’s perspective, it’s difficult to know precisely how bad these situations have been.

It’s often said we like numbers here at FanGraphs — and that’s true, as far as it helps us understand the game better. But managers are one aspect of the game that have, to date, defied attempts at being quantified. Matt Williams put together a 179-145 record during his two seasons in Washington. That’s a winning percentage of .553, which comes out to a 90-win average. That’s pretty good! Matt Williams must be a good manager, then. The thing is, and you probably know this if you read FanGraphs, those are team stats, and ascribing them to one person, even if that person is ostensibly in charge of the team, is probably a mistake. There’s a lot more that goes into a winning team than the manager, a fact that was underlined today by the Nationals’ actions, and I suspect will be underlined again by Williams not being snatched up by another team any time soon.

I said earlier that Williams’ firing wasn’t a surprise, and it wasn’t because of a few things. Locally, Williams’ bullpen management has come under fire, as has his handling of the clubhouse. Nationally, the big stories were the Nationals’ loss of the division to an upstart Mets team that didn’t reach 90 wins until the end of the season and the fight in the dugout between Bryce Harper and Jonathan Papelbon. That is warmed over territory so I won’t cover it in any great detail other than to say the fight itself seemed to confirm the worst fears about Williams’ handling of the team in the clubhouse. Then the way Williams handled both the fight’s immediate aftermath and the public relations aspect afterwards erased the word “seemed” from the previous sentence. It’s one thing to be good behind closed doors as maybe Williams was (or not), but to appear clueless and indifferent and rigid while your team is exploding on national TV (no pun intended) within 50 feet of where you are standing is another much worse looking thing.

Even with all that, there’s a fair argument to be made that Williams was a passenger on this ship and his jettisoning is perfunctory. In 2014, when the Nationals won 96 games, made the playoffs, and Williams won his Manager of the Year award, the Nationals lost only 417 man games to injury according to Man Games Lost, 20th most in baseball. This season, when they missed the playoffs and Williams lost his job, they lost 1,030 man games to injury, seventh most in baseball. If the Nationals had lost 600 fewer man games to injury like they did last season, would the Nationals have finished behind the Mets and would Williams have lost his job? If the Nationals won more games, then Papelbon probably doesn’t snipe at Harper and probably Harper doesn’t jaw back at him. And on and on.

The Washington Post has covered the Williams angle extensively since the Nationals mid-season fall from first place, and today Chelsea Janes published a list of seven decisions that helped lose Williams his job. It’s instructive, I think, to go over the list a bit. But let’s make it fun! How many of the seven do you think were bullpen decisions? I’ll let you vote and then reveal the answer later so no peeking!

Before I tell you the answer, though, let’s quickly talk about what Williams did wrong. We know he was structured in his bullpen usage — I’d argue to a fault — though as with all bullpen decisions, it’s difficult to know for sure because we only see one result. It’s not a choose-your-own-adventure book where we can compare outcomes. We further know that reporting has indicated there were problems in the Nationals clubhouse. We also know that there are often problems in many teams’ clubhouses over the course of 162 games, and problems in the clubhouse don’t necessarily signify that the manager has made mistakes, even if that’s what reporting indicates. Terry Francona wrote a book about his time with the Red Sox, during which he won two World Series and was heralded as a managerial savior, and even on those world championship-winning teams there were in-season problems, often ones that sounded rather serious and would have been fodder for big newspaper articles had they got out at the time. And those are the ones Francona was willing to divulge in his book! Presumably there are others, possibly more sensitive examples that he didn’t include.

This isn’t to say Williams did a good job holding the clubhouse together. All indications are that he did not, but just to point out that those are only indications and even with that background we can’t really know what went on. Clubhouses are the responsibility of the manager but are composed of the players. Some players have outsized personalities that even the best managers can have difficulty controlling.

So, back to Janes’ article, which by the way I heartily recommend. How many of Williams seven deadly decisions were bullpen-related? The answer is…

[Drum roll]

Six.

Here they are the seven in summary:

1. Pulling a starter for the closer in the ninth.
2. Seventh inning reliever choice.
3. Not using his closer in a tie game on the road.
4. Leaving starter in to pitch seventh.
5. Not using his closer in a tie game on the road.
6. Asking Anthony Rendon to bunt on a 3-1 pitch in the ninth against the Mets.
7. Leaving Papelbon in to pitch after he choked Harper.

It’s instructive, I think, and interesting, that a manager as derided as Williams is still being judged (harshly) on his bullpen usage. How many managers refuse to use their closers in tie games on the road? Many, many of them! Further, can you fairly kill a manager for pulling a starter in the ninth and for leaving his starter in to pitch the seventh? Maybe so, but it seems iffy.

None of this is intended as criticism of Janes’ article. I think it’s representative of the Williams zeitgeist, however, and instructive in that, for a manager who apparently everyone despises, most of what we can get on him is that he won’t use his closer in a tie game on the road.

Was firing Williams a good thing for the Nationals? I surely wouldn’t be alone by arguing yes. It further seems that of the sins listed above, many came from the same book, one probably called something like Game Time Decisions for Managers for Dummies. Managing is easier if you can push buttons, and it takes someone more secure than Williams in his decision-making to go against established thinking.

The Nationals face an off-season of turnover as many of their starters are set to be free agents so it makes sense to bring in a new manager, clean break, timing, and all of that. Williams is interesting in that all the known aspects of his situation point to an obvious decision and yet when we dig deeper we discover that we really don’t know very much at all.