Dominating Strikes

Amateur baseball analysts are very enamored of the "swinging strike" statistic, a count of how many pitches a pitcher throws which are swung at and missed by the hitter. And for good reason! Throwing strikes is, obviously, an important thing for a pitcher; if you don't do it, you just end up walking people. But you can, strangely, throw "too many strikes," creating a situation where the hitters know they're going to get a hittable pitch and that they should therefore swing at it, and you aren't walking anyone but you're giving up hit upon hit upon hit. The solution is, of course, the swinging strike. If you have pitches that are good enough to induce swinging strikes, you can throw "too many strikes" and get away with it. They'll know a strike is coming, but not be able to do anything about it. And, of course, swinging strikes lead to strikeouts. So there are lots of strategic reasons for liking swinging strikes, but they're also just symbolically important. A swinging strike represents pure and utter dominance by the pitcher. You threw a pitch, the batter thought it was good enough to merit a swing, and yet they failed so miserably at the very most central aspect of hitting that they didn't even hit the ball at all. Not even weakly. Not even so weakly that they fouled it back. They just missed it. Poof. Fail. Pitcher 1, Hitter 0.

Called strikes do not have that same feeling of dominance. Well, not all called strikes. A first-pitch fastball down the middle is not particularly dominating. The hitter just didn't want to swing at it. We don't know whether the hitter would've hit it well had they swung at it. It also didn't do a particularly impressive job of disadvantaging the hitter. No more than, say, a foul home run. A called strike is as much a choice by the hitter as it is a demonstration of skill by the pitcher: for whatever reason, he didn't want to swing, even though the pitch was in the zone. Okay, fine. Except, sometimes that's a choice no hitter would ever make. Sometimes you know for a fact that the hitter wasn't okay with the result "called strike," that something more must have been going on. Because no hitter ever wants to take strike three called.* When that happens, you just plain lose the battle, and you have to go sit down. If the hitter had known the pitch would be a strike, and had been able to swing at it, they would have done so. We know that of every called third strike. Which means that either the pitch fooled the batter into thinking it was a ball or it did something to them that prevented them from swinging at it, though they wanted to. In other words, it necessarily involves some domination of the hitter by the pitcher. It's not as physical domination as a swinging strike; for all we know, had the hitter managed to swing they would have hit a home run off that pitch. It might have been very hittable, had it only been hit at. But it wasn't, though it needed to be.

So I'd like to see a new statistic, a slight modification of the "swinging strike" count. It is, obviously, swinging strikes plus called third strikes. That is a measure, to my mind, of a pitcher's ability to beguile hitters, to overpower hitters, and to do every combination of the two. The great pitchers often have a specific pitch that reliably gets them strikeouts looking. Greg Maddux developed the tailing fastball over the inside corner to a lefty. Anyone with a great curveball, like the ones that Justin Verlander and Clayton Kershaw have showed off in this year's post-season, will be able to drop it into the center of the strike zone, disrupting the hitter's timing and making them abandon any thought of swinging long before the pitch crosses the plate. The called third strike is the only other pitch besides the swinging strike where the pitcher unambiguously demonstrates their superiority to the hitter, and we should have a statistic that counts both methods of dominating hitters equally.

Fire Dan Warthen

The Mets' bullpen woes are, at this point, getting ridiculous. Their bullpen this year has lost 19 games, and while I don't think bullpen wins mean a damn thing and I tend to agree with the modern critique of wins and losses I do think that relief losses are meaningful. Relief pitching is an inherently situational thing, and the only way to get a loss as a reliever is to enter a game that your team is not losing and give up enough runs that they are. That's a bad thing, it's always a bad thing, and having 19 of the team's 44 losses come from the bullpen is absurd. For most of the year the starting pitching has been outstanding, primarily on the strength of Johan Santana and R.A. Dickey, but of late even they have been enduring a rough patch. In my opinion, it's time to get rid of pitching coach Dan Warthen.

Now, it's probably not necessary to go into much detail about why Warthen should be fired; he's been there since mid-2008, during which time the results have been sadly lacking, and for the last couple of years the pitching's been suspect. But just for fun I took a slightly thorough statistical look at the Mets' bullpen under Warthen, that is, from the second half of 2008 onward (a slightly rough proxy, since he got there in mid-June, but it'll do). Over that time-span, Mets relievers have had an ERA of 4.186. The total ERA for National League relief pitchers in that same time is 3.906. Mets relievers have struck out 7.7 per 9 innings, while walking 4.1 and allowing 9.0 hits; National League relievers as a whole have struck out 8.0, walked 3.8, and allowed 8.4 hits per 9. If Mets relievers had performed at league-average levels, they would have given up 55 fewer walks, 110 fewer hits, struck out 68 more batters, and allowed 68 fewer runs including 59 fewer earned runs since Warthen joined the Mets. That's around 6 or 7 wins below average for Mets relief pitching under Dan Warthen, and that's before you account for the fact that the bullpen typically pitches in higher-leverage situations so those runs below average might be even more damaging than that. (Oddly, the Mets seem to do a bit better than the league as a whole at not giving up home runs from the bullpen, though the difference is minute.)

So, long story short, the Mets' bullpen has been lousy under Dan Warthen. There's a pattern to it, namely god-awful in 2008 (which can arguably be blamed more on the loss of closer Billy Wagner, but it's just true that the team's famous bullpen troubles that year were entirely a second-half phenomenon), good in 2009 and 2010, bad in 2011, and god-awful in 2012. I'm not sure what to make of that pattern in terms of judging Warthen, but I think that at this point it's clear that, at the very least, he's not providing any value about replacement pitching coach. There's no reason not to ditch him. They should do so at soonest convenience.