Is Dave Roberts the worst manager in post season history?

Is Dave Roberts the worst manager in post-season history?

The answer to the headline is yes, Dave Roberts is the worst manager in baseball post-season history.  I’m not even sure who is in second place.  He over-managed the Dodgers in the 2017 World Series and lost a winnable series.  In 2018 he pulled starter Rich Hill even though he was throwing a one-hitter after 6 innings.  And now in 2019 he blows Game 5 of the NLDS by leaving in Clayton Kershaw despite having a rested bullpen who had been effective all year.

I think there are two reasons for the decision to leave Kershaw in the game.  The first is the idea that he burnishes his reputation as a manager by making unconventional moves.  Anyone can go to the 8^th inning specialist in the 8^th inning of a baseball game, but only an innovative manager would leave in a starter, even if he has declining skills and has been used as a starter recently. 

Secondly, I think he was hoping to create a narrative that would burnish Kershaw’s Hall of Fame credentials.  Kershaw gets called “The greatest pitcher of his generation” a lot, but his post-season record is 9-11 with a 4.43 ERA.  Roberts wanted to give Kershaw the same chance that the San Francisco Giants gave Madison Bumgardner in 2014 when he came in to relieve on short rest to clinch the Giants’ victory.

But there are a couple of differences.  One, MadBum was 24 at the time, not 31.  Kershaw has been seen as in decline, and while still effective he isn’t usually mentioned in the Cy Young debate.  Maybe at one time he was “lights out” but not anymore.  Also, as a starter, in 2019 Kershaw’s ERA in his first inning was 5.97, so he was used to getting off to a slow start.  As a starter that’s forgivable if you subsequently get on track; as a reliever, having a bad first inning is disastrous.  Relievers have to come into a game and hit the ground running; this is another reason why Kershaw is not a reliever.

Second, the Dodgers had better options.  The 2014 Giants didn’t have a strong bullpen, but the 2019 Dodgers did.  It wasn’t great, but it was good.  But instead of going to Kenley Jansen or Pedro Baez or Joe Kelly in the 8^th, they went with Kershaw.   I call that a vote of no confidence in the bullpen.

I think the experience with Bumgardner in 2014 started an unfortunate trend.  The next year Matt Harvey of the Mets insisted on pitching in the ninth inning of Game 5 of the World Series even though he had never thrown a complete game in his life, and he promptly blew the game and the Royals won the title.  Then you have the Dodgers not using their bullpen properly in 2017 and 2018.  I suspect something happened in 2016 but I am too lazy to look it up.

Bottom line—World series managers, please stop looking for excuses to use starters as relievers on short rest.  Once in a while it works, but that’s not the way to bet.  Before the series began, the Dodgers had a 62% of winning; after using Clayton Kershaw in relief, the chances were zero.

In Defense of the Wild Card Play-In Game

One of the things I find frustrating about baseball in the modern era (I am not sure what I mean by that; post-steroid, pre-robot umpire?) is the unmooring of a concept that I understood intuitively even before I took a single course in statistics—that over a 162 game season, crap evens out.  A bad bounce on a ground ball, an unfortunate gust of wind propelling a fly ball into the stands, an ump having a bad day behind the plate, could all influence the outcome of a single game, but over the course of a season the breaks will even out and the better team will win more games.

One aspect of this unmooring is the replay rule, which demands that EVERY play be examined in super-slo-mo to assure that the ball was caught by the first baseman a full one-thousandths of a second before the batter’s foot hit first base.  Yes, egregious calls need to be corrected, if possible (assuming there can be a consensus on the definition of “egregious”) but unlike football, a missed play here or there won’t necessarily determine whether a team’s season is considered a success or a failure.

There are policy reasons for imprecision.  The “neighborhood” play at second was conceded for decades, because not caring that the pivot man in a double play touched second base while he had the ball was more important that fielders’ risking knee damage from a hard slide, or runners being beaned by thrown balls that travel exactly down the baseline to first base instead of from an angle outside the basepath.  But now if a shortstop is a fraction of an inch off the base when making the pivot, the other team will challenge the call and demand that someone ion New York City review it.

More recently, I have become annoyed with well-meaning commentators grousing about the one-game playoff between wild card teams in each league.  For example, on this FiveThirtyEight podcast the moderator says a one-game play-in game wasn’t fair and they should play a seven game series.

The World Series already threatens to extend into the first days of November; a seven game wild card series would guarantee a World Series starting in November and ending after the Ides of November.  That sounds iffy for a sport where, unlike the Super Bowl, the location of the finals can’t be determined in advance; given that the Twins play in an outdoor stadium IN MINNESOTA this seems insane.

The thing is this—the purpose of the wild card play-in game is NOT to decide which team is the better one.  They just finished playing 162 games and the team hosting the play-in game had some advantage over the other one that justified giving them home field advantage.  Playing one game is unlikely to better make this determination.  The A’s had a one-game lead over the Rays after 162 games, so they host; even if they lose the play-in game, they would have the same record but the A’s won the season series over the Rays, so they still could be considered the better team.

No, the purpose of the wild card play in game is to disadvantage wild card teams in the next round of the playoffs.  Presumably the teams in the wild card play-in game will start their ace, making him unavailable for the first two or three games of the Division Series.  The wild card winner will send their #2 starter against the division winner’s #1 starters, and their #3 starter against the other team’s #2. 

When the wild card was implemented it seemed like a good idea (and it was), until wild card teams started winning the World Series.   In the NFL, wild card teams have the disadvantage of no home field advantage, but in baseball if you split the first two games suddenly the home field advantage shifts, and now three of the remaining five games are played at the wild card team’s park.  Something was needed to subtly give the division winner a slight edge over a team that merely came in second (or even possibly third).

So don’t gripe about the Nationals getting lucky because a rookie outfielder botched a routine ground ball to right field.  Maybe the Brewers could have won a seven-game series, but they should have gotten the opportunity to play in one by winning their division.