- FanGraphs Baseball - http://www.fangraphs.com/blogs -

Rule Change Friday: Batting Order On the Fly

As you may have heard, the current rules of baseball were not set in stone and brought down the mountain by Abner Doubleday, never to be changed. There have been many changes over the years. Whether those changes made the game better or worse is debatable, but to believe the game exactly as currently played is somehow “sacred” such that any changes would make it something other than baseball is silly.

The idea of Rule Change Friday, as I first tried to implement in a discussion of pickoff throws, is to (hopefull) generate fun discussions about potential rule changes the game better more enjoyable (from a variety of perspectives). In that first post, I discussed a suggestion by Bill James that (as far as I know) has never actually been practiced in professional baseball. For this installment, I would like to explore the idea of moving forward by moving backward by looking at the possibility of in-game establishment of the batting order.

My knowledge of the history of early baseball is even shabbier than my knowledge of baseball history in general. But when I do read about early baseball, I almost always have a good time learning new things. Today’s idea comes from something upon which I stumbled a while back in… The Bill James New Historical Abstract (that guy again?). It was not a suggestion of his this time, however, but simply a short comment he appends in his summary of professional baseball in 1880s. After noting that Cap Anson led that decade in RBI with 967, James notes:

“Early rules did not require the batting order to be announced prior to game time. Before this was changed in the early eighties, Anson would sometimes wait and see if the first two men got on. If they did not, he would bat; if not, he would wait and hit in the next inning.”

The focus of this post in on batting order and rules, so let’s leave aside some of Cap Anson’s less praiseworthy contributions to baseball history (not to mention his role as Potsie on Happy Days) and focus on the tactical aspect of the situation. It makes good sense, doesn’t it? With runners on and less than two outs in the first inning, you probably want your best hitter to hit. However, with none on and two outs, run expectancy for the inning is very low. This is true in any era, but particularly so in a time during which a player could lead the league in home runs with six.

The paragraph above is not clear about this, but it seems that if he did not hit third in the first ining, he would lead off the next inning, thus hitting fourth for the rest of the game. That is more good baseball logic. If he hit fourth, he did not wait any longer to get men on, he was not worried about “wasting” his bat by leading off an inning, since he still had a good chance to get on base, thus increasing the run expectancy for his team. Moreover, it acknowledges that the batting order is going to “come around” again, so he still might have the chance to hit with runners on.

One could go on along these lines, but you get the general idea. This was a sound tactic given the rules at the time, but this piece is not meant to be a celebration of Anson (and without knowing one way or the other, it is reasonable to guess that other teams employed similar tactics), but rather to take a look at the rule. My brief internet research (not much, although I also did a bit of asking around) did not reveal exactly when (all I found was the “early eighties” in the James quote above) and why the rule was changed. Perhaps what Anson was doing was taken to be against the “spirit” of the game. Maybe it was just one more thing for umpires to track. Baseball was still in development at this stage. To name a just one interesting example of differences from today: foul bunts were not yet classified as strikes.

Without knowing all of the details of how this worked and why it was changed, I wonder if it might not be a fun “back to the future” rule change for contemporary baseball. The rule would simply be that the batting order is not “set” prior to the game. I assume the defensive lineup (and the DH for games in which that would apply) would have to be announced before the game, but the batting order would only be set as each spot in the order came up for the first time. On a practical level, there should be no problem. The umpires in conjunction (perhaps with the aid of the official score keeper or someone like that) could keep track of it in the same way they keep track of which bench players have been used.

But for the sake of what I hope will be a discussion in the comments, I think this could lead to a more strategically and aesthetically interesting game. I would guess that the lead-off hitters would remain mostly the same, but managers would have to display more awareness of what sorts of hitters and players are better in particular base/out states. At the moment, there are ways in which most teams could be using more optimal batting orders, and while the difference is not completely insignificant, for the most part “plug-and-play” lineups work pretty well. But if managers could make their batting orders according to the proposed rule change, I think we would have a chance to see some pretty big differences — or at least who is paying attention. If a manager opts to put in high-average/high-on-base hitter without exceptional power up third in the first inning with none on and two outs, we will know that either the manager is not paying attention or really is not familiar with what is going on.

Of course, things are not always that cut-and-dried, and the differences perhaps would not always be that glaring. After all, managers still would not know how things would play out the rest of the game after the batting order is set for the first time in terms of base/out situations confronting each hitter. Without doing any modeling (as if I could, anyway), I would guess that most of the time, the best hitters would still be hitting in the first four or five spots. What would mostly come out is to what extent managers were distinguishing between the particular skills (power, getting on-base, contact, walks, speed) of those players, and how those would likely work out in particular base/out situations. Maybe an analogy with bullpen usage might help. Currently teams can leverage their relievers to match certain situations at the end of the Perhaps one way of looking at the proposed change is that it would give teams a chances to leverage their hitters in accordance with the situations that would come up at the beginning of the game.

This is all for the sake of speculative discussion. The game as it is now is fine, but this might make it more fun in a different way. This is an aesthetic issue rather than a mathematical one. The more I think about is though, the less crazy and radical is seems. What do you think? Let’s use the most scientific of all possible methods: the internet poll. It would be great to hear different viewpoints in the comments as well.

My thanks to a certain undead pitcher for an attempt to answer my request for aid.