The Michael Pineda suspension put the Twins in a real bind when it comes to their postseason rotation. Then again, the hallmarks of this 2019 team — aside from massive amounts of home runs — have been resilience and a creative use of the pitching staff.

The combination of resiliency and creative use of pitching was perhaps never more evident than this past Saturday, when the Twins swept a doubleheader from Cleveland despite not having a traditional starting pitcher in either game. Those throwing in relief in both games combined for 11â…“ shutout innings, keying the victories.

At the risk of putting too much emphasis on a small sample size and falling prey to recency bias, I'm starting to think the Twins' ideal postseason rotation isn't so much a rotation at all but a collection of arms designed to maximize the chance to get 27 outs with the least damage.

Assuming the Twins go on to win the division, which they seized control of with that doubleheader sweep, they will face either the Yankees or Astros in the division series. Both are extremely tall orders and will require every possible optimization.

Here's my 12-man postseason staff: Jose Berrios; Jake Odorizzi; Martin Perez; Taylor Rogers; Trevor May; Sergio Romo; Tyler Duffey; Zack Littell; Brusdar Graterol; Devin Smeltzer; Kyle Gibson; One of these two — Randy Dobnak or Lewis Thorpe.

Here's how I would line it up pitching-wise against either opponent in a five-game series:

Game 1: Berrios. He's still the Twins' best chance at a deep start and to match another team's ace. He gets the ball in Game 1. If he's on, you ride him for six to seven innings and hope to only use two to three relievers to finish the game.

Game 2: Bullpen game, with Martin Perez as the opener. You go with a bullpen game here instead of waiting for Game 3 because you get an off day for travel after this one and everyone can get rest following Game 2.

But Perez? The guy who has struggled for much of the second half of the year? Why him?

Opposing batters have an OPS of just .627 against Perez in their first plate appearance of the game this season. That jumps to .887 in the second plate appearance. Tuesday was a good example: Perez kept the White Sox scoreless for the first three innings, then gave up a run in the fourth and two in the fifth.

For two to three innings, he can be quite good. Limit Perez's exposure. From there, you throw every available top arm for an inning or two as long as you're in the game. This would have been a laughable strategy two months ago when the bullpen was a mess. Now the bullpen is an asset.

Game 3: Odorizzi. Entering Wednesday, he had worked at least five innings while allowing three earned runs or fewer in 22 of his 28 starts this season. He can be better than that when he's on — don't forget, Odorizzi pitched seven shutout innings vs. Houston and six shutout innings vs. the Yankees in back-to-back starts earlier this year — but his baseline start will keep you in the game.

Game 4: Bullpen game, with either Smeltzer or Gibson as your opener. The Yankees and Astros have crushed lefties and righties with equal efficiency this season. It comes down to who you trust more for two to three innings. It might be Smeltzer right at this moment, but Gibson has a little time to re-establish himself and, like Perez, is better when teams see him just once.

Game 5: Berrios starts and all hands on deck. There's another off day after Game 4. Berrios gets the ball with a healthy five days of rest if there's a decisive Game 5. But if he falters, every reliever will have had a day of rest and should be ready to be summoned early enough so the game doesn't get out of hand and the Twins have a chance to rally.

"Berrios and Odorizzi and use the bullpen until they're dizzy" isn't quite "Spahn and Sain and pray for rain." But it's kind of the modern version of a rainout equalizer when you lack quality rotation depth.