Out-by-out examination: When Buck should have used Britton

If there's one thing to take away from how the Orioles' Buck Showalter managed the wild-card game Tuesday, it's that you should call your mom right now and tell her you love her -- or your spouse, or your son or daughter, or your best friend. Don't put it off, and don't wait for just the right time. Even if you don't normally call your mom at work and remind her on an idle Wednesday that you love her, that's OK. She'll still be thrilled to hear from you. When you get to the bottom of your 11th inning, when you get to the Ubaldo Jimenez Pitching In Relief moment of your life, you don't want to be the person who hoarded all of your I love yous.

Or, to put this another way: Let's go through Tuesday's game, out by out, and see whether it would have been a good time for Showalter to have picked up the phone, asked for Zach Britton and told him that he loves him and/or would like to bring him in to pitch.

Bottom of the first, nobody out: Yes.

There's a reason the closer closes, and it's not because he's the closer: Managers want to get the most out of scarce resources. Pitchers can't pitch every inning of every game (anymore), and there are no teams that have enough really good pitchers to cover 1,500 high-quality innings in a season, so managers have to decide how to get the most out of those few pitchers who can definitely get outs. For some, that's best done by starting the game and throwing until they're exhausted, then, after appropriate rest, repeating the assignment. But for relievers, teams do it with "leverage." If Britton is going to be available for only, say, three innings over the next five games, it makes no sense to use those innings in a blowout, where he is unlikely to tilt the outcome of the game. Britton pitches only in close games because, in a world with scarce resources, there is a cost to using Britton, and managers want to avoid paying that cost unless they really need to.

This is a fine and rational system over the course of 162 games crammed into 180 days, and it works. Britton pitched 69 times this year, which is about as many as a modern reliever can be asked to handle. Most of those games are what we'd call "high-leverage" -- for example, coming into the ninth inning to save a 2-0 game against the New York Yankees. Forty-six of his other 68 appearances were higher-leverage than even that. More importantly, by limiting him to these 69 games and 67 innings, there was hardly a game that Britton wasn't available. In fact, it looks as if there was just one -- April 7, when Darren O'Day got the save because Britton had thrown 26 pitches the day before. Of the 20 most consequential moments the Orioles faced this year -- the individual plate appearances that carried the most leverage -- Britton was on the mound for 14.

This is why we have Zach Brittons. Forget what you might hear about the last out being the toughest; ignore the qualifications of the save statistic, which was designed merely to recognize the best relievers, not dictate strategy. The reason 150 years of baseball strategy have brought us to the This Is How We Use Zach Britton era has nothing to do with the ninth inning. It has to do with making sure that the best pitchers are used to maximum effect. So in a game like Tuesday's, where there is no tomorrow, where every pitcher (other than a few starters) is well-rested, where every pitcher will have (at least) a day of rest before the next game, there is no scarcity. And if there's no scarcity, then leverage becomes an illusion. Every inning is the most important inning that the Orioles will have because those are the only innings the Orioles will have. If the Orioles were down 10-0 in the ninth, it still would have been a good time to bring in Britton because Britton is good, and there is no cost to trying hard. The only way to leverage your best pitcher in a win-or-go-home situation is to make sure, no matter what, no matter what, no matter what, you use your best pitcher somehow. One way to make certain of this is to have him start. You can make a pretty good case for Britton starting, going an inning or two and then bringing in starter Chris Tillman. (Or, better still, bringing in Brad Brach or Darren O'Day and working backward until you run out of pitchers who aren't Brian Duensing.)

Bottom of the first, one out: No.

The Orioles decided that they didn't have enough good relievers to make it through nine-plus innings, so a starter would be needed for at least some of them -- whether at the beginning, the middle or the end is irrelevant, but some of them. Tillman was deemed the best option to do that. Once he's in the game, he's got to ride for as long as is necessary to get to the short-stint dudes. Not trying to scald you with this take, but no, Showalter should not have pulled Tillman with one out in the first inning and replaced him with Britton.

Bottom of the first, two outs → Bottom of the fifth, nobody out: No.

Same logic as above: If Showalter thought he needed a starter to get him to his bullpen and he thought Tillman was his best starter, then he's got to ride with him until the bullpen (the good parts of it) can plausibly take over or until the third-time-through-the-order penalty kicks in and reduces Tillman's effectiveness. One might argue that Britton, O'Day, Brach and Mychal Givens are capable of covering more than four innings in a win-or-go-home situation, but at a certain point, we just have to trust that Showalter knows his guys' physical limits better than we do.

Bottom of the fifth, one out: Probably.

Tillman would end up going 4 1/3, so we have it seems Showalter thinks his relievers are capable of throwing at least 4 2/3; innings in this game. So Britton is an option again.

Remember that I said leverage in a one-game playoff is an illusion because there's no scarcity? That overstates it a little. Britton can't throw nine innings, so there is still scarcity within the game, and some moments within the game truly are more important than others -- specifically, the ones in which runners are on base. Allowing a single (or a homer) will be more damaging in those moments than when there are two outs and nobody on. Once Tillman allowed a one-out double to Michael Saunders, the cost of another hit went up. A life vest is, after all, much more likely to be used when you're in danger of drowning than when you're lying in bed thinking about how comfy you are.

This sort of usage -- bringing Britton in to quell a pop-up rally -- is hard to manage in the regular season, which is another very good, very rational reason to manage the way that managers do. It takes something out of a pitcher to warm him up in the bullpen, and managers really don't want to warm a pitcher up and not use him or to warm him up multiple times in a game. So teams develop predictable bullpen hierarchies whereby relievers (and their manager) can anticipate who will pitch in each situation and whereby unnecessary warm-up throws are minimized. Baseball is smart, and 150 years have refined these patterns to a point at which they make a lot of sense. But this obstacle is irrelevant in a wild-card game because the most important thing any pitching change can accomplish is ensuring that the best pitchers are in the game for as long as they can be. If Showalter gets Britton warmed up, and Tillman just as quickly gets out of the jam, no problem -- bring Britton in anyway! Bring Britton in! Just bring Britton in! One cannot fail except by failing to use Britton in an entire must-win baseball game.

Instead, Tillman gave up another double, at which point Britton could have been used. Instead, Tillman gave up another single, at which point Britton could have been used. Showalter went to Mychal Givens instead, which is just fine too -- Showalter needs to get 14 more outs from his relievers, and Givens is one of the four guys most qualified to do it. Of course, it's only just fine if you know for certain that Britton is going to pitch. Otherwise, Showalter was choosing to use a worse pitcher instead of a better one, and doesn't it sound odd when you say it like that?

Bottom of the sixth, none out: It's close, but probably not.

Now that Givens is in the game and has thrown only five pitches, it's probably worth letting him get another out or three. Twelve outs might still be too much to ask for from O'Day, Brach and Britton, especially if one of the three just happens not to have it on that night.

Bottom of the sixth, one out → Bottom of the seventh, two outs: No.

Givens is a good pitcher pitching well, and the Orioles still need somebody who's not Britton to get some outs. At this point, Givens appears to be one of the most qualified to do so.

Bottom of the seventh, two outs: Yes, but only because Showalter brought in Donnie Hart instead.

The situation -- two outs, nobody on, one of the Blue Jays' worst hitters up -- is the opposite of a jam, and it makes a lot more sense to hold Britton until the eighth and let him have two clean innings. But there are a number of ways to improve upon a sequence involving a 26-year-old rookie left-hander pitching to a right-handed power hitter in a ballpark made for right-handed power hitters. One of those is, naturally, bringing in the best reliever in baseball.

Bottom of the eighth, nobody out: Yes.

The best reason for using your closer in a tie game? If you hold him for the save, you get, at most, one inning out of him. If you use him in the tie, you quite possibly get two. Sounds crazy? This actually happened once! An ace reliever was brought into a game that was tied 2-2, and he pitched so well in preserving the tie that his manager sent him back out to pitch the next inning. He again held the lead, throwing 26 pitches over two scoreless innings, and later in the game his team scored four runs and won the game in 14 innings. (There was, in fact, no save, because of the final margin of victory.) Who was this brilliant manager? It was Buck Showalter! Who was this ace reliever? It was Zach Britton! Who was the team that failed to score against him two times? It was the Toronto Blue Jays, and it was just two months ago!

Bottom of the eighth, one out: Yes.

It's the postseason. Ace relievers routinely go five outs in the postseason.

Bottom of the eighth, two outs: Yes.

It's the postseason. Ace relievers routinely go four outs in the postseason.

Bottom of the ninth, none out: Yes.

Josh Donaldson, Edwin Encarnacion, Jose Bautista due up.

If somebody wants to make a very nuanced case that Brach is better than Britton, I'm happy to listen to that weird opinion. But Showalter didn't use Brach because he thought Brach was better; he did it because he thought there was going to be a bigger moment in this game than the ninth inning of a tie game with the best 2, 3 and 4 hitters in baseball coming up.

Bottom of the ninth, one out: Yes.

Now it's just Bautista, but there are two men on base. A single wins it. Two fly balls to right field win it. Showalter still thought there would be a bigger moment in this game somewhere down the line, even though there's something like a 65 percent chance that this game is going to end within the next four minutes. Showalter brings in O'Day, a great pitcher, a wonderful option to have, a pitcher for whom a manager can say with no shame, "I'd like to have him pitching for me in a tight spot." But the choice was not O'Day or wave the center fielder in and have him pitch. It wasn't even O'Day or Brach. It was O'Day or Britton, and the reason you hold Britton is you think there's a bigger moment coming, and there isn't.

Hot dog, was it fun seeing O'Day get out of that jam though.

Bottom of the 10th, none out → Bottom of the 10th, two outs: Maybe, but probably not.

About a third of extra-inning games end up going at least 12 innings, which means even if Britton has two scoreless in him, the Orioles are quite possibly going to need more good pitchers than Britton. Pulling O'Day after his one pitch in the ninth would have burned the next best Oriole when he still had more left. I can appreciate gambling on O'Day throwing a clean frame here, knowing it extends the best part of the pitching staff one inning deeper into a game that, for all Showalter knows, might go 18.

Bottom of the 11th, none out: Yes.

Duensing is so good that, at age 33 this season, he threw more innings for the Kansas City Royals' Triple-A affiliate than he did for a major league team. He's the guy you come up with when you're trying to make a joke about a generic reliever who probably retired four years ago. He pitched in this game instead of Britton.

Bottom of the 11th, one out: Yes.