My Proposal For a New Pitching Regime

I love baseball. It is a game of tradition and history, and I love that, too. But sometimes I think maybe it relies too much on tradition, and managers make decisions informed more by what their habits are than by what the optimal approach is. Pitching arrangements are one of these things. So I’ve thought a little bit about how franchises might improve the way they handle pitching. Here’s the model, and then I’ll discuss it below.

[Note: I’m sure someone else has come up with ideas similar to this in the past; I’ve never read other research on this issue but I’m also not trying to claim this is the first time an idea like this is being presented.]

Current Model

12 total pitchers (five starters, seven relievers)

Starters: Expected to go ~6 innings/game (but as many innings as possible is ideal)

Relievers: Expected to go ~3 innings/game

Each starter pitches once every five games, and relievers pitch as needed (spreading out the workload).

My Proposed Model

12 total pitchers (four long-range, four mid-range, four short-range) 

Long-range: Pitch three to four innings

Mid-range: Pitch two to three innings

Short-range: Pitch one to two innings

A long-range guy will start each game, probably, but optimally pitch only three innings (i.e., once or so through the opposing lineup). Then other guys step in afterward. Let’s model a two-week span. Our pitchers are as follows:

LR1, LR2, LR3, LR4; MR1, MR2, MR3, MR4; SR1, SR2, SR3, SR4

Day 1:

            Inning:           1           2           3           4           5           6           7           8           9

            Pitcher:          LR1     LR1     LR1     LR2     LR2     LR2     MR1    MR1    SR1

            Comments: We’ll assume that to start everyone is on good rest.

Day 2:

            Inning:           1           2           3           4           5           6           7           8           9            10

            Pitcher:          LR3     LR3     LR3     MR2   MR2   MR3   MR3   MR4   MR4   SR2

Comments: Note the extra innings. Everyone’s pitching on full rest still. Also, let’s pretend MR4 gets injured at this point—not enough for a DL stint, which would allow the team to replace him on the roster, just enough that we don’t want to pitch him for a while.

Day 3:

            Inning:           1           2           3           4           5           6           7           8           9            10

            Pitcher:          LR4     LR4     LR4     LR4     MR1    MR1    SR3     SR3     SR4     SR4

Comments: Extra innings. LR4 goes four innings, MR1 comes back on 2 days’ rest, SRs 3 and 4 pitch two innings each.

Day 4: [Travel]

Day 5:

            Inning:           1           2           3           4           5           6           7           8           9

            Pitcher:          LR1     LR1     LR1     LR1     LR2     MR2   MR2   SR1     SR1

Comments: LR1 pitched on 4 days’ rest, LR2 had a bad outing (also after 4 days’ rest), MR2 came in on 3 days’ rest, and SR1 came in on 4 days’ rest.

Day 6:

            Inning:           1           2           3           4           5           6           7           8           9

            Pitcher:          LR3     LR3     LR3     MR3   MR3   MR3   SR2     SR2     SR3

Comments: LR3 pitches on 4 days’ rest, MR3 pitches on 4 days’ rest, SR2 pitches on 4 days’ rest, SR3 pitches on 3 days’ rest.

Day 7:

            Inning:           1           2           3           4           5           6           7           8           9            10

            Pitcher:          LR4     LR4     LR4     MR1    MR1    MR2   SR4     SR4     SR1     SR2

            Comments:   LR4 pitches on 4 days’ rest, MR1 pitches on 4 days’ rest, MR2 pitches on 2 days’ rest, SR4 pitches on 4 days’ rest, SR1 pitches on 2 days’ rest, SR2 pitches on 1 days’ rest.

Day 8:

            Inning:           1           2           3           4           5           6           7           8           9

            Pitcher:          LR1     LR1     LR2     LR2     LR2     LR2     MR3   SR3     SR3

            Comments:   LR1 pitches on 3 days’ rest, LR2 pitches on 3 days’ rest, MR3 pitches on 2 days’ rest, SR3 pitches on 2 days’ rest.

Day 9: [Travel]

Day 10:

            Inning:           1           2           3           4           5           6           7           8           9            10

            Pitcher:          LR3     LR3     LR3     LR3     MR4   MR4   MR4   SR2     SR2     SR1

Comments: LR3 pitches on 4 days’ rest, MR4 pitches on 8 days’ rest (back from his injury), SR2 pitches on 3 days’ rest, SR1 pitches on 3 days’ rest.

Day 11:

            Inning:           1           2           3           4           5           6           7           8           9            10

            Pitcher:          LR4     LR4     LR4     LR1     LR1     MR1    MR1    MR1    SR4     SR4

            Comments:   LR4 pitches on 4 days’ rest, LR1 pitches on 3 days’ rest, MR1 pitches on 4 days’ rest, SR4 pitches on 4 days’ rest.

Day 12:

            Inning:           1           2           3           4           5           6           7           8           9

            Pitcher:          LR2     LR2     LR2     LR2     MR3   MR3   MR2   MR2   SR3

            Comments:   LR2 pitches on 4 days’ rest, MR3 pitches on 4 days’ rest, MR2 pitches on 5 days’ rest, SR3 pitches on 4 days’ rest.

Day 13: [Travel]

Day 14:

            Inning:           1           2           3           4           5           6           7           8           9

            Pitcher:          LR1     LR1     LR3     LR3     MR4   SR2     SR2     SR4     SR3

Comments: LR1 pitches on 3 days’ rest, LR3 pitches on 4 days’ rest, MR4 pitches on 4 days’ rest, SR2 pitches on 4 days’ rest, SR4 pitches on 3 days’ rest, SR3 pitches on 2 days’ rest.

*Note: I’ve liberally sprinkled the above 2-week span with extra innings and included some injuries, just to better replicate a high-stress stretch of baseball.

Upsides of this system:

1) Most lineups will see different pitchers every time around. Studies show that each time through the lineup, pitchers fare worse on average. This should reduce or eliminate that problem.

2) Roster flexibility: This approach allows teams some flexibility in who they pitch and when. They aren’t tied to a rotation, and are able to better exploit platoon advantages. This also means opposing teams will be worse-positioned to predict who they’ll face.

3) Better workload distribution: The rotation system tends to leave relief pitchers underused. At any given time, barring a lot of extra innings over a short time or a long stretch without any travel days, most teams probably have several relievers ready to go. Maximizing the number of fresh arms the opposing team sees is generally preferable, so cycling in more pitchers should produce better results long-term.

4) Cost effectiveness: The most expensive pitchers in the game are the ace-quality workhorses that throw complete games and sign massive contracts. Some of them are worth it; some of them prove not to be. My system puts less strain on pitchers, effectively lowering the bar for what’s considered above-replacement level quality. You don’t need Justin Verlander to go seven or eight innings if you anticipate getting only four innings out of your starting guy—someone like Joe Saunders, maybe. The premium on the difference between a guy who can reliably pitch three or four quality innings and a guy who can reliably pitch seven or eight quality innings is huge, and teams using my system won’t have to pay it.

5) Roster restructuring: Suite to point four, above. My system, as presented above, follows the 12-pitcher model most clubs use today. But as far as I know, there is no rule stipulating 12 as a magic number. My system is designed to be a lot more cost-effective, allowing teams to shift funds from their pitching staff to position players. Having a utility man around is obviously indispensable, but perhaps two extra outfielders AND two extra infielders would be unnecessary if more money could be invested in position players that wouldn’t be candidates for, say, replacement based on platoon matchups. That way, the number of pitchers could expand to 13 or even 14—creating an even stronger staff.

Downsides of this system:

1) Uneven rest. While pitchers will pitch less, on average, they will pitch in more uneven increments. Some people suggest that the rotation system creates a rhythm that’s essential to starting pitchers’ functioning. I anticipate that some pitchers would be unwilling to pitch under this system—specifically, those who depend on ritualistic procedures to prepare for appearances. This problem is not prohibitive, though. The pitchers in this system should be considered comparable to “relievers” used today, who don’t usually have the benefit of knowing whether they’ll be needed on a given day. Also, workloads are reduced, so extended conditioning rituals should be less necessary for preparation.

2) Egos: One of the most obvious issues with this system is that a lot of pitchers simply wouldn’t want to pitch under it. The system relies on systematic changes and short outings, so no archetypal starting pitcher would have a shot at a complete game. Further, traditionally relied-upon stats like wins, saves, and quality starts just wouldn’t work well with this system, and players seeking to rack them up would probably be dissatisfied under my regime. Note a few things: First, when I say “ego,” I don’t necessarily mean to criticize players as self-interested. They need to look out for what makes them marketable and what is best for their careers, and this system probably isn’t it. Second, wins and saves are really dumb metrics in the first place. Overall, I think there’s a strong argument that giving pitchers shorter pitching obligations would make them more consistent and ‘better’ overall—some adventurous guys, especially the sort of mid- and low-level starters that are perfect for this system, might have a lot to gain from a lower ERA (and xFIP, and a better K/9, etc.).

Moral of the story?

I think there’s been a lot of evidence lately that pitching development is so unreliable that the investments franchises make in pitching prospects might not be profitable in the long-run. There’s so much mediocre pitching out there available for cheap that adapting systems designed to maximize the benefit that market can offer could give teams an edge. We all know shorter outings create better average results; why not seek to maximize the number of short outings? I think there’s something to be explored here, at the very least.