The more we learn about the inner workings of baseball, the more apparent and prominent the roles of players’ habits and humanity become. We are slowly finding that the game is about talent and strategy, but that talent is maximized by comfort, and strategy is optimized by the ability to account for the adjustments and anticipations and learning curves of all the parties involved.

If this statement rings empty or false for you, consider the pinch-hitting penalty (and the lesser, but still real, DH penalty), which shows that players have a hard time performing at their best when they come off the bench. Consider the times-through-the-order penalty, which has of late come under a bit more scrutiny, but which suggests that batters gain a progressive advantage from the ability to see what pitchers have in real time, from inside the batter’s box, more than via video. Consider the performance of relievers pitching on second and third consecutive days, and consider the fact that starting rotations and regular lineups exist at all. Players perform best when they are able to prepare according to a schedule, a routine, and teams strive to make sure they can do just that.

If baseball weren't so serious—if, for the occasional season, teams could really experiment without having to answer to a thousand scolds and shareholders—here's the one I'd want to see, to test which of two fairly important baseball routines is more fragile. I want to see what would happen if starting pitchers didn’t start games. Specifically, I want to see the following structure replace the traditional deployment of pitchers over the course of a game. We’ll use the Pirates as the example here, partly because they're a team that's sort of in need of something to save their season right now. Also because they have a deep (though struggling) pitching staff, and because they’re in as good a position as anyone to try this (given not only the aforementioned depth, but their need to improve quickly in order to stay in the playoff hunt, and their established ability to use information and creativity to get more out of their pitching staffs than people expect).

1. Identify the team’s four best relievers. In the Pirates’ case, these will be Mark Melancon, Tony Watson, Jared Hughes, and Neftali Feliz. They’re the only members of the pitching staff whose roles will be fundamentally unchanged.

2. The three other relievers in the bullpen (the Pirates are carrying eight relievers right now, in fact, so toss Rob Scahill into the bin with his betters, as a long guy/low-leverage sponge; only three pitchers are needed for this phase of the plan) form the new starting rotation. Each of Arquimedes Caminero, Cory Luebke and A.J. Schugel will start every third day, pitching to the first 3-6 opposing hitters. On some nights, they might pitch two full innings. More often, they’ll get three or four outs.

3. Once those relievers start and throw something like 20 pitches, the Pirates’ traditional starters—Francisco Liriano, Jon Niese, Jeff Locke, Jameson Taillon, and Juan Nicasio right now, with Gerrit Cole shelved and Tyler Glasnow not yet promoted to the parent club—relieve them and pitch roughly their normal amount. Clint Hurdle can manage the game more or less normally from there, removing the reliever-starter at whatever juncture he deems wisest, but presumably closer to the end of the game than under the old, boring way of doing things.

Obviously, this would be a bit of a shock to the system for Pirates starters. Most starting pitchers have long routines in the run-up to their starts, including stretching, throwing, running, and a bullpen mound session. They often time it very closely, so as to be as perfectly prepared as possible at game time. It’s a routine the benefits of which remain somewhat dubious, but I have no doubt that it’s one starters would object to having altered so significantly.

There's one potential problem. Another is that the team would be committing to at least one inning from one of its worst pitchers every day, without the benefit of knowing which games are going to be especially close in advance. Think about the potential disruptions for the offense, though. Depending on how the starter-relievers and reliever-starters line up, it might be harder than ever for opposing managers to gain the platoon advantage against Pirates pitchers. It would also change the dynamic of the in-game learning that fuels the times-through-the-order penalty, delaying the first, second, and third viewings of each day’s “starter” for about half the opposing batting order. It’s impossible to know for sure, without seeing the strategy in action, how opposing skippers would adjust their lineup in response to this kind of gambit, but if they stuck with a traditional lineup (which might still be the best option against this plan; I’m not sure), it would be the best hitters who saw the “starter” last, and got their third crack at him latest, if at all. On certain nights, the Pirates might also be able to send the “starter” to the plate one fewer time than they normally would, though it’s hard to predict how often.

Mostly, this is simply a way to pose the question: Who would be weirded out more? Would “starters” who no longer actually started fight command problems or have a hard time ramping up to their full velocity? Or would opposing offenses lose their rhythm, lose their ability to adjust and gain incremental advantages as the game went on, and end up producing less? And to whatever extent the “starters” did get hurt by this, would the relievers who gained the opportunity to prepare with absolute certainty and plenty of rest, who pitched in a more controlled environment than they used to, improve as a result? If an underlying battle in baseball is getting comfortable and keeping the other team from getting comfortable, who benefits when somebody changes all the phsyical laws of the game?

I don’t know the answers. I just want to know them. Like most potential baseball gambits, it’s unlikely to see the light of day, because clubs don't get to experiment just to sate their (or my) curiosity. Baseball's so serious. Maybe everybody's more comfortable that way.

You need to be logged in to comment. Login or Subscribe
I assumed that once you identified the teams best 4 relievers, you were planning on using a combination of 1 or 2 of them every game to start. I would really like to see a team like the Yankees try "closing" the first inning or two. If you can get an early lead before your starter even takes the hill that must be a very serious advantage. (Also, the other team is running out of innings to catch up). I believe the opposing team might be rattled knowing the first inning they are very not likely to score.
"If you can get an early lead before your starter even takes the hill that must be a very serious advantage." An early lead is always a serious advantage. I don't see how this could justify using your very best pitchers for innings where their use will often not be warranted. Each of Miller, Betances and Chapman could give you 80 to maybe 100 or possibly 120 innings every year. It's best to use these innings when you are much more certain that each out is more valuable. Therefore it is best to use them near the end of games. No, the Yankees (and every other team) are not using their best relievers optimally, but starting the game with these guys makes zero sense. Unless you're talking about days where they need work and the manager knows he wants to use them. That's a different story.
I agree with the comment above. I think it would only make sense if you start the game with the best bullpen pitchers. Ideally, you don't want to use the worst pitchers in your bullpen everyday.
I think Matt's idea is that by starting the lesser lights, maybe they get put in a situation where they won't be as bad as expected. Will it work? I dunno, but it's worth asking the question. Of course, any nonstandard pitcher usage has to pretty much show instant success or the press/fan/player pressure will kill it before any refinement can take place. Give somebody like Ben and Sam one year to gain the trust of their players and then *maybe* they could give something like this a fair shake in the second year...
Can you imagine any team trying this? Managers still save their best relievers for the 9th regardless of the opposing lineup.
Fantasy leagues with QS as a counting stat would be pissed
Oh well.
Well ideally this works for us but the starters have been hit & miss (except for Liriano who has miss) & the pen you mention has been pretty bad. Watson & Hughes have been below standard & Schugel has been better than expected. The rest have been switching apartments every other week in Indy (AAA)