When rosters expand to 26 players for the 2020 MLB season, it is almost assured that the St. Louis Cardinals will come North with 13 pitchers on the roster. That is the most that will be allowed on the roster at a time, per the new rules, and so it would be silly (especially if you've been going with 13 pitchers on a 25 man roster for the large majority of the last three seasons) to keep anything less than that.
It could stand to reason that if you can maximize the position player half of the roster, then you should be able to deploy 13 pitchers to cover approximately 1,450 innings over the course of a Major League Baseball season in an optimal manner. For instance, one way to do that would be if you asked all 13 pitchers to pitch the same amount of innings. If you did, you'd be looking at them pitching just over 110 innings apiece. I would not consider that optimal, however. When you talk about a team wanting to maximize their pitching, a team should want their better pitchers pitching more often (their starters typically) and their worse pitchers a bit more protected, if you will.
I believe that there would be several efficient ways to optimize the pitching staff. Some of them are radical and some are not. I want to go over:
What it would look like to simply have your best five starting pitchers to throw every 5th game and how that would affect the bullpen pitchers.
What it would look like to optimize the starts that the best five starting pitchers get to enable them to be at their most effective - and also how that would affect the bullpen pitchers.
What it would look like to have a piggyback rotation and how that could affect relievers.
And lastly, what it would look like to have a modified piggyback rotation and how that could affect the relievers as well.
This could get lengthy. For that, I apologize.
Prior to going into those four descriptions of optimizing the pitching staff that I lay out above, I put a question out to Twitter the other day about who the best starters are and in what order in the Cardinals' system. 20 results came back and they averaged out to say this:
Flaherty
Martinez (tied - my vote was the tie breaker)
Mikolas (tied)
Hudson
Wainwright
Helsley
Poncedeleon
Cabrera
Gomber
That is what I will go by to tell me who the best starters are. So when discuss the use of 5 starters, we will use Flaherty, Martinez, Mikolas, Hudson, and Wainwright. When we talk about a piggyback rotation we'll use the top 8. When we discuss the modified piggyback rotation, we'll discuss all 9 guys. Here we go.
The "5 Best Starters Throwing Every 5th Day" Optimization Plan
The St. Louis Cardinals could "optimize" their rotation by simply having the best 5 guys go every 5 games, like every other team in the league attempts to do (health presumed). That has been the go-to for years and years. It would give Flaherty and Martinez 33 starts and Mikolas, Hudson, and Wainwright 32 starts. Because of how the schedule works out, it would give:
Flaherty 14 home starts, 18 road starts, and 1 London start,
Martinez 17 home starts and 16 road starts,
Wainwright 17 home starts and 15 road starts,
Hudson 16 home starts and 16 road starts, and
Mikolas 15 home starts, 16 road starts, and 1 London start.
Since I knew that I would get Flaherty and Martinez 33 starts apiece, I made sure to give Flaherty the first start of the year, and thus allow him to have more road starts because I feel his stuff plays up anywhere he plays. I also took the liberty of figuring out how many home/road/London starts the #3-5 starters would each get and put Wainwright at home more often than any of the others receiving 32 starts in this plan. That's as "optimized" as this plan can get, however.
This plan would tax the bullpen slightly less than this past year (2019), most likely. You've got 5 starters. You're missing Wacha, who only went just over 5 innings per outing and replacing him with Martinez, who as a starter in his career, has gone just over 6 innings per outing.
This is not as optimal as I would like.
The "5 Best Starters But Better" Optimization Plan
In this plan, I intentionally do two things to make sure that the Cardinals get the best out of this starting pitching group as I can. The first thing I did was to look at how to get Wainwright as many home starts as possible in the season. The second thing I did was, surrounding those Wainwright starts, look at how many times I could feasibly throw Jack Flaherty while giving him at least four full days off between any games. The last thing I did was try to give Mikolas (rather than Martinez or Hudson) the remaining schedule with the most home games. I was left with 7 games (all just happened to be on the road) that needed to be spot started by someone else on the 13-man pitching staff. I believe that should not be a problem.
Here is what I was able to accomplish:
Flaherty - 34 starts - 17 at home and 17 on the road
Mikolas - 31 starts - 16 at home, 14 on the road, and 1 in London
Martinez - 30 starts - 13 at home, 17 on the road
Hudson - 30 starts - 12 at home, 18 on the road
Wainwright - 30 starts - 22 at home, 7 on the road, 1 in London
Spot Starters - 7 starts - 0 at home, 7 on the road
So this plan kinda sucks for Martinez, Hudson, and the spot starters. While I don't think a 13-17 or 12-18 split is excessive, it is out of the norm, I am sure. What the plan does is allow Wainwright, who has a 3.00 ERA in 123 innings at home in the last two years while allowing a 5.97 ERA in 89 innings on the road in the last two years, to throw 73% of his starts at Busch Stadium. That could seriously be the difference between a 4.50 ERA on the season and a 3.80 ERA on the season for Wainwright. I don't feel the other players have those sort of splits to worry about. Looking at the last two seasons, the other 4 Cardinals starters above have the following home/road splits by ERA:
Mikolas - 2.57 home ERA, 4.42 road ERA,
Hudson - 2.40 home ERA, 4.33 road ERA,
Martinez - 2.43 home ERA, 3.92 road ERA, and
Flaherty - 2.60 home ERA, 3.38 road ERA.
While those are all some home-favored splits, they're nothing like Adam Wainwright's. Waino's road ERA is nearly double his home ERA and is over a run and a half worse than anyone else's road ERA over the last two years (that is being considered as a starter in this space).
I personally think that this is the most likely way the Cardinals could attempt to optimize the lineup because the next two require outside the box thinking to the nth degree.
At about the 33 minute mark of this Chirps Podcast (normally hosted by Tara Wellman and Alex Crisafulli), Kyle Reis was filling in for Tara and opined about the lack of consistency and lack of guarantee in major league baseball bullpens and how they are nearly impossible for a manager (even a Manager of the Year like the Cardinals' Mike Shildt) to manage effectively. It was a common criticism of Mike Matheny when he was here and it is a common criticism of managers around the league. Hell, even Dave Roberts and AJ Hinch get flak on bullpen management and those teams won a combined 225 games this year (including playoffs).
I think that is as good a lead in as any for why my next two ideas could actually work. Hopefully that, along with the evidence I've compiled over the last full season of baseball, can convince you it's worth the look at what otherwise would likely be looked at as a bat-shit-crazy idea.
The "Straight Piggyback Rotation" Plan
Here is the premise of the next two ideas. (I like the second more than the first, which is why we're talking about this one first.) I came up with this idea prior to last year (at least for this iteration of the Cardinals team). I had talked years ago, as I said earlier, about doing this with a pitching staff as a possibility. My idea came from trying to get the most you could out of Michael Wacha and his terrible 3rd time through the order splits and Alex Reyes and his right arm that was in the process of falling off at the time. It didn't work out for Reyes staying healthy enough to try that this year. *sigh* (Poor guy. Hope he gets healthy soon.)
The premise is that you have four pairs of starting pitcher partners that "piggyback" to give you 18 batters apiece every 4 games - for a total of the first 36 batters of the game. That should work out to about 70-75 meaningful pitches every 4 games. I would most likely hold them to the 70-75 pitch limit since they would be pitching every 4 games - not a full rest compared to the normal starter around the league.
I looked at every single start in the 2019 season and ended everyone's start after 18 batters (if it went that long, some would have been pulled early) and here is what the Cardinals would have done in 2019.
It took Cardinals starters an average of 4.13 innings to get through 18 batters this year. That means that each piggybacked start would get you through 8.27 innings on average. That means that the bullpen (which Kyle's point above was that they are completely unreliable year to year) only would be needed to - on average - get you 2-3 outs per game. With the new rules allowing for 13 pitchers on the 26-man roster, that gives you 5 guys to get 10-15 outs TOTAL every 5 games. With the new rule in place that pitchers cannot throw to less than 3 batters unless they finish an inning, you would not really have the chance to completely burn your bullpen under this plan.
What would this look like in terms of quality of starts? Well, in those 8.27 innings the piggyback starters would allow 3.77 runs on average (3.42 earned). They gave up 7.95 hits and walked or hit 3.16 additional batters per 8.27 innings. So that's about 10 base runners every 8 innings. Not bad, right? If you were told that you could have 5 guys in your bullpen and everyone else would go 8-8 1/3 innings every night giving up 3-4 runs, you would take that every night of the week and twice on Sundays, right?
The pitchers actually averaged 69.79 pitches per 4.13 innings and 139.58 pitches per 8.27 innings in the season long sample - lower than my suggested 70-75 pitches per 4 games.
With the top 8 from the way above Twitter crowd sourced polling, I would pair:
Flaherty and Cabrera and give them 20 home and 20 road starts,
Martinez and Poncedeleon and give them 19 home, 20 road, and 1 London start,
Hudson and Wainwright and give them 20 home, 19 road, and 1 London start, and
Mikolas and Helsley and give them 20 home and 20 road starts.
That would leave 2 road starts in the last two games of the season to go to whomever the Cardinals wanted to.
The obvious cons of this program would be 1) Jack Flaherty likely leaves when his contract is over due to not pitching as deep into games and not getting as many innings pitched, most likely - which means less money for a guy that is going to get paid and 2) it is unconventional and convincing everyone involved would be terribly difficult because of baseball norms.
The obvious pros of the system are that you can 1) avoid the ups and downs of a bullpen season to season and 2) you can attempt to keep arms healthier by throwing less pitches slightly more often.
The "Flaherty + Piggyback" Plan
In this plan, I use the same theory as the prior plan. However, I have Flaherty pitching every 6 days (every 5 or 6 games), getting him 31 starts (slightly more rested than a normal season). With the extra rest, he could potentially go slightly deeper into games - this year averaging 5.95 IP/start - hopefully pushing to closer to 6.5 innings a start to get him just over 200 innings in 31 games. That would leave 2.5 innings for just 4 relievers every 6 days. Then, I would piggyback 128 (32 each) of the 131 remaining games with these four groups:
Wainwright + Helsley get 17 home, 14 road, and 1 London start,
Mikolas + Cabrera get 15 home and 17 road starts,
Hudson + Poncedeleon get 15 home and 17 road starts,
Martinez + Gomber get 15 home and 17 road starts
The 3 remaining starts would be spot starts (or go to the first three piggyback groups) in game numbers 159, 160, and 161 on the season with game 162 going to Flaherty unless the team is ready to simply use him in game 1 of the post-season. Then you give the last 4 games to the piggyback groups one last time and wait on Flaherty until then.
The cons of this plan are that there are only 4 relievers, but if you only need one per game to get 2-3 outs, it should not be a problem in 128 of the games. The problem arises if Flaherty goes really short in a game. It'll happen at some point, I"m sure. Then again, maybe he goes all Adam Wainwright or Chris Carpenter of the mid-2000s and pulls of a 215-230 inning season somehow. Then he gets paid even more.
In any case, I believe that you can see that there would be many ways to maximize the roster by using the depth of the St. Louis Cardinals starting pitching. I prefer the fourth method, but the second is easily the most doable with the optimization of the 5-man rotation that prevails across Major League Baseball.