A Modest Proposal (For the Playoff Roster)

The playoffs are a funny thing. They’re the focus of tons of articles, tons of critical thinking, tons of narratives. Have you heard of playoff bullpens? Of course you have. Have you heard that the Nationals are bad in the playoffs, or that some guy or another is clutch? That one too. Despite all this focus, however, only a third of the teams in baseball even make the playoffs. Of those ten, six have gone home after four or five games. It’s very likely that you, as a fan, are only going to experience a tiny sliver of playoff strategy. You think the idea of optimizing your bullpen in the playoffs is neat? If your ace gives up three runs, that whole narrative is scrapped. Have a great platoon idea? A hitter could have an 0-15 slump and bam, the playoffs are over for you. In the trenches of actually experiencing the playoffs as a fan, there’s not a lot of time to be high-minded.

Luckily, the playoffs haven’t started yet, so it’s time for me to be high-minded. I’ll throw out the usual bundle of don’t-yell-at-me caveats: the Cardinals aren’t in the playoffs right now, they’re in the regular season. They could easily miss the playoffs. Nothing is certain, least of all in baseball. Knock on wood, fingers crossed. Great, we covered all the disclaimers. The reason I want to write about the playoffs now is because when they start, I’m not going to be a cool-headed analyst. I’m going to care if Cardinals Batter A gets a hit off of Evil Villainous Pitcher Y. I’m not going to care about what his WAR was this season. I’m not going to care whether a pitcher would be better used in long relief or in different leverage situations- I just want them to strike out every opponent. Today, though- today I can think about optimization. Today I can think about what I’d like to happen, before the red haze of rooting shrouds my eyes. Today, I’d like to talk about my ideal rotation and bullpen for the playoffs.

One thing that became evident to me right away when I decided to write about who should pitch for the Cardinals in the playoffs is that this team has both tons of depth and tons of pitchers with uncertain roles. To wit: the first thing I threw into my roster grid was two starters, like so:

That was just automatic. You can quibble about which of the two you like more, but those are the Cards’ top two healthy starters, end of story. The next thing I did was add three relievers who you could broadly call the scripted relievers, like so:

What I mean by scripted relievers is that if the game goes according to plan, these guys are there to handle the 7th, 8th, and 9th innings. If they pitch those innings, the game is going according to script. You could quibble, I suppose, with Dakota Hudson. Maybe you’re more of a FIP than an ERA kind of person, or maybe you just like relievers who strike out more batters than they walk. You and the Cardinals would disagree on that, though. Hudson has the second-highest entry leverage index on the team. In other words, the Cardinals put him into the game when things matter the most. The only reliever who enters in more important spots is Bud Norris. Jordan Hicks finishes a close third.

When I went to fill out the next block of pitchers, though, I was stumped. In the end, I settled for a category I’m going to call multi-inning relievers. It’s basically everyone else the team has tried in the starting rotation this year:

That’s a lot of names. They’re all over the map in terms of both performance this year and pedigree. For my money, Poncedeleon and Ross are a cut below the rest, but even if you want to minimize their roles a little bit there are four guys ready to go out there and give you two-ish innings at a time. Finally, we have the actual relievers:

If you’re paying attention, you’ll notice that I’ve listed seventeen pitchers without listing Michael Wacha or Adam Wainwright, who I expect to be more break-glass-in-case-of-emergency players than part of Plan A. In reality, only twelve of them will make a playoff roster. Some NL teams have even gone with eleven pitchers in the playoffs of late, but given the depth the Cardinals have, I expect them to roster twelve. To figure out which pitchers to cut, I tried to figure out a rough script for where the Cardinals might go to get nine innings a game of pitching. I think you’ll like what I came up with.

Before I get down to my exact plan, here’s a quick refresher of the playoff format. Due to travel days, the Divisional Series has off days after Games Two and Four. The Championship Series has off days after Games Two and Five, as does the World Series. As such, most of the playoffs is a two games on and one game off format. That’s really helpful for my plan, as you’ll see. I’m also assuming the Cardinals are using Mikolas in the Wild Card game, so I’m not planning on using him before Game Three. Here we go, starting with Game One:

I budgeted six innings for Flaherty here, but he has reached six innings in only about half of his games pitched this year. As such, John Gant is available for an inning stint here as well. The last three members of the bullpen, whoever they end up being, are also available. That’s a pretty conventional one- let’s take a look at Game Two:

Spicy! This setup has a ton of advantages, though. Carlos Martinez might be a short-stint pitcher only for the rest of the year, but he’s looked excellent in that role so far. Austin Gomber might be a little overextended as a playoff starter, but in two innings of max-effort relief he profiles quite well. I recently covered Luke Weaver’s potential as a multi-inning reliever and really love him in the role. Each of these three have skills that play up in two-inning bursts, and the righty-lefty-righty stacking effect can only be a benefit when it comes to disrupting opposing hitters’ timing. After that, there’s the predictable 7th/8th/9th triumvirate. Game Three is roughly what you’d expect:

As with Game One, I’m allotting six innings to Mikolas, but Gant is on hand if he needs to chip in extra. The back of the bullpen, as always, is available as well. This isn’t exactly rocket science- when your best pitcher isn’t available until Game Three, well, you pitch him in Game Three and hope it works out. What are we going to go for Game Four, you ask?

That’s the second benefit of the triple-tandem start. With each of the three Game Two trio only throwing two innings, their recovery time is a good deal shorter. The Cardinals haven’t shied away from giving relievers multi-inning duty on two days’ rest, which is the setup here. If you’d like, you can even switch Weaver and Martinez to try to mess with opposing hitters a bit more. The three of them will all be pitching, so it doesn’t particularly matter in which order they enter the game. That’s basically it for the rotation- Game Five is just a repeat of Game One with Flaherty on five days’ rest.

At this point I’ve used nine pitchers, which leaves three to fill out the rest of the bullpen. My personal preference would be to carry two lefties and a righty to more aggressively play platoon matchups. I’d personally go for Chasen Shreve and Tyler Webb as the lefties, but if the team prefers Brett Cecil I’m not going to quibble. All three of them are reasonably interchangeable in my mind. I’d like John Brebbia for the last righty in the ‘pen, but I have a feeling that the Cardinals and I aren’t on the same page here. Until we see him pitch effectively again in the majors, he’s a risk, and the team will probably prefer a safer bet even if their upside isn’t as high. As such, I think this spot will go to one of Dominic Leone, Daniel Poncedeleon, Tyson Ross, or Adam Wainwright. Having Wainwright there would be cool for nostalgia’s sake, at the very least. This is probably the least important spot in the bullpen, so I’m not dead set on Brebbia. I do think he has the potential to be an impact reliever, but maybe that’s next year’s business.

Here’s the whole bullpen together:

That’s a pretty solid group. The magic of the triple-tandem start mitigates a lot of vulnerabilities. A tiring Weaver? A second look at Gomber’s flat fastball? Not happening. If you haven’t imagined Austin Gomber as a Lance McCullers-style breaking ball machine, I don’t know what to say to you. We may still be a long way off from this style of pitching working in the regular season, but the unique travel format of the playoffs makes it far more palatable. One weak point of my plan is that it doesn’t include a lot of days off for Hudson, Hicks, and Norris. They’re likely to get rest organically though- if any of these games gets out of hand, they can be switched out for one of the Gant/Shreve/Webb/Brebbia group. Additionally, any one of them can be subbed out individually if they need the rest. Overall, I like the way this schedule leverages the Cardinals’ best arms for the maximum possible amount of innings.

There’s one added benefit to the triple-tandem start that I’ve mostly glossed over. With a planned pitching change after every two innings, there’s no reason to let your pitcher bat in an important situation. A little thought in drawing up the batting order can help maximize your odds of having a pitcher’s spot naturally come up when he’s ready to be removed from the game. On the road, nothing changes at all. A pitching change comes after the top of the third inning. Batting the pitcher ninth makes it reasonably likely that his spot in the order will come up in the top of the third, when a pinch hitter can painlessly take his place. The next spot for a free pinch hitter comes up in the top of the fifth. That’s marginally less likely, but the most likely time for the ninth spot in the order to come to bat for the second time is in the fifth inning, so the odds are at least in the Cardinals’ favor. That’s great news- on the road, the place the team wants to bat the pitcher anyway is still optimal.

When the Cardinals are at home, the math gets a little more interesting. A free opportunity to pinch hit will come up in the bottom of the second inning, too early to expect the ninth spot to come up. The first place you’d think to bat the pitcher if you want to guarantee his spot comes up in the second is sixth. That’s a guaranteed plate appearance in the second inning, and a probable plate appearance in the fourth for a second free pinch hitter. The only wrinkle in this plan is that it probably isn’t ideal for your pitcher to bat with the bases loaded and two outs in the bottom of the first. It’s probably not ideal for them to bat in the first, period. Batting seventh is a hedge against that, but it comes with its own problems. The seventh spot doesn’t come up until the third inning about 10% of the time, which would result in a missed opportunity for a free pinch hitter. It seems like a close call here, so let’s do some further analysis.

To do the math on which of these spots would be better, I had to do a bit of estimating. This isn’t quite right, but I took the percentage of innings that the Cardinals scored at least two runs in as an estimate of how often the pitcher’s spot will come up in the first inning (I didn’t have data for how often the sixth spot in the order batted in the first). Not every two-run inning involves six batters, but most do, and hopefully the ones I’ve over-counted are offset by one and zero run innings where six batters come to the plate. This comes out to about 13% of innings (I excluded all extra innings and the ninth inning to eliminate innings with walk-offs, which could skew the math). This is a little higher than the naive odds of the first six batters being retired in order, which means I’d prefer to have the pitcher bat seventh.

So, there you have it. My master plan is to turn three starters into one Voltron-like mega-starter and hope the shorter bursts kick their results into overdrive. I’ve never seen anything exactly like it in baseball before, but that’s not a real impediment to the plan; the logic checks out to me, and it’s a way to turn the Cardinals’ unique depth of starters into a weapon. It probably won’t matter. Most playoff series don’t come down to the kind of narrow margins of efficiency the Cardinals will pick up with this plan. That doesn’t mean they shouldn’t try, though. This plan is, to me, the best way to attack the unique problem of postseason roster construction. Getting to see a pitcher bat seventh? That’s just a bonus.

One quick postscript: Good God am I sorry about the resizing on the tables. I tried a new format to make tables a little prettier on mobile but it looks like the payback on desktop is no joke. Tune in next time for another episode of Ben Fumbles With Technology.