Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52

During the first four or five months of the season, I don’t care which teams are playing, as long as there is at least one day game I can watch from my location six time zones ahead of the East Coast. But when September arrives, I often find myself looking at the schedule in disgust when I learn that the only game played at 1 PM features two teams already out of contention.

September also brings a different kind of baseball, as rosters expand and teams pull out all the stops to make the playoffs. Given the altered nature of the game in the final month of the regular season, the men in charge of pushing the buttons should know the answers to a few questions that either do not arise or are not really relevant earlier in the season. Let’s have a look at a few of them.

The effect of DH-ing on production at the plate
The first question is this: How (and how much) does being employed as a DH affect one’s production at the plate? Some teams never have to face this issue. They play in the National League or feature a fixed DH, one who hits a ton and doesn’t even bother to bring his glove at the ballpark (someone like David Ortiz). For all other teams, though, it’s important to be aware of the offensive side effects of not playing a position.

You have a backup catcher who is a defensive wizard and can save you some runs with his glove, but your regular catcher is one of your best bats. (This example can apply to other positions as well.) Until September, you rarely risk using him as a DH, because an in-game injury to the catcher behind the plate might lead to consequences like having a pitcher bat or needing to go through a merry-go-round of double switches (which sometimes confuse AL teams). But now you have your bench crowded with September call-ups, so the worst-case scenario is ending up with your third-string catcher (who might also be competent at playing defense) behind the plate.

Penciling in your offensive catcher as the DH seems like an easy choice here, but would he hit equally well while spending more time on the bench? Sure, he won’t have to face the wear and tear of squatting and being hit by foul tips, but he may not be comfortable having nothing to do between at-bats.

It has been shown that coming in as a pinch-hitter does decrease the expected performance of a batter. Can the same be said of the DH role, especially for someone not accustomed to it? Research in The Book suggests that the “DH penalty” is roughly half the magnitude of the “PH penalty.” The task here is to quantify the drop-off (if it actually occurs) in the offensive stats of the player you are moving to the DH spot and decide if it’s significant enough that it’s better to preserve the lineup you used when the roster was composed of only 25 men.

Defensive replacement at the end of the game
Note: Ben Lindbergh brought this issue to my attention during a chat we had a few days ago.

As I have already mentioned, the player who comes to bat after having ridden the bench the whole game is expected to perform worse than if he had been playing the entire game.
Does the same go for defensive replacements?

Defensive replacements are used throughout the year, but bigger September benches afford teams more opportunities to employ the strategy. No analysis has been done (at least publicly) on this aspect of late-game substitutions.

While the pinch-hitter by definition has one chance to change the course of the game, a defensive replacement often won’t be a factor. However, when he does get a chance, it’s often an important one. You might remember DeWayne Wise entering the July 23, 2009 Rays-White Sox game in the ninth inning with Mark Buehrle working on a perfect game. The newly entered outfielder was immediately tested, and Wise responded with a spectacular catch to preserve the perfect game, the no-hitter, and the shutout.

One might argue that a defensive replacement might field even better than he usually does, because his legs are fresh. Wise’s amazing catch could be presented as anecdotal evidence in favor of this theory. But it’s equally plausible that he might get a worse jump or reach full speed more slowly because he hasn’t been active for hours. If that’s the case, defensive replacements might do more harm than good. The fact is that we don’t know whether the defensive substitute will field better or worse than he usually does, and the sample sizes involved might prevent us from finding out.

A note on framing
Jose Molina has entered some Rays games this season as a defensive sub in the ninth with Tampa Bay slightly ahead. I’m not sure if that’s because closer Fernando Rodney prefers having him behind the plate or because the Rays believe his framing wizardry can help maintain a small lead. This happened on Monday, and Molina stole a strike (the pitch was way outside) with Curtis Granderson pinch-hitting. This time, Molina’s effort didn’t make the news, since it was “only” strike two and didn’t involve any helmet-smashing, but one pitch later, the game was over on a swinging strikeout.

That particular play made me wonder what we should expect from a good framer entering the game in the ninth. Will he fool the umpire even more because the man in blue has not had the time to catch on to his tricks? Or does framing involve a long process of earning the umpire’s trust as the game goes on? For this one, sample size probably won’t be an obstacle, so I can come up with an answer in a future article.

Should pitcher workloads be handled differently?
You might expect me to write about Stephen Strasburg’s shutdown here, but I have something else in mind.

For most of the season, teams tend to live by the magical number of 100 pitches. Maybe the limit is extended a bit for known workhorses, while youngsters’ arms are treated with additional care, but there is usually a 90-110-pitch range in which the bell (and the bullpen phone) ring.

Should this be the case in September? There are a couple ways to consider this question.

  1. By the final month of the year, pitchers have accumulated a lot of innings, so they tire faster and should be removed at a lower pitch count (especially given the availability of more arms in the pen).
  2. The pitch limit imposed during the previous months has ensured that pitchers are not overworked and, given that they’ll have the winter months to rest, the limit can be raised.

The workload issue is not limited to starting pitchers. It also applies to bullpen aces: you can use more arms in a September game, but it makes sense to send your bona fide stoppers to the mound as often as possible in high-leverage games. How will they handle longer stretches of consecutive days of work? And the same can be said of catchers, who are usually given a rest on day games following night games.

A final question: What do you do with your 24-year-old phenom who has reached the innings limit devised for him at the start of the season when you’re in the middle of your franchise’s first playoff race? Oh, I said I wasn’t going to talk about that.

Starter matchups
Nowadays, managers don’t get too creative with rotations. They may skip some fifth starters when off-days allow, but otherwise, they more or less go through the season with regular rotations.
That hasn’t always been the case. As Chris Jaffe showed in Evaluating Baseball Managers, skippers used to leverage the starts of their best arms. For example, Mordecai Brown started half of his games of the 1909 season against either Pittsburgh or New York (the top two opponents of the Cubs) and half against the remaining five National League teams.

It’s not clear which is better over the course of a season: maximizing the number of innings thrown by your best pitchers, given fatigue/rest constraints (the current method), or not wasting some of your ace’s starts against teams you can beat with your back-of-the-rotation guys, or whose defeats don’t matter as much in the standings (the “elastic” rotations approach).

Let’s assume evolution has taken care of this, and the current way of managing the rotation is the optimal one over a full season. Nevertheless, the approach should be reconsidered for the final month of the season, and even in the season’s closing days.

Here’s a hypothetical scenario. You need to win two of the last three games of the year to clinch a postseason berth, and the natural rotation would be the following.

Your Starter

Opposing Starter

Chance of Victory (Given Pitching Matchup)

Albert Ace

Andy Amazing


Tommy Tough

Omar Ordinary


Michael Mediocre

Pete Passable


You can do the math: the chances of winning at least two games out of three are 60 percent. But what if you play Ace against Ordinary and Tough against Passable, giving both of them a higher likelihood of winning a game for you? That would mean finding a starter for the first game (assuming Mediocre would not be available so early) and having the matchups look like this:

Your Starter

Opposing Starter

Chance of Victory (Given Pitching Matchup)

Clyde Callup

Andy Amazing


Albert Ace

Omar Ordinary


Tommy Tough

Pete Passable


While your odds are almost hopeless in Game One (unless Clyde Callup is 2011 Matt Moore), your chances for winning two games out of three are now close to 75 percent.

These numbers are made up, and a real team would hardly be able to come up with figures this precise given all the factors in play. One would also need to know whether pushing Ace’s and Tough’s starts ahead by one day would disrupt their routines (and make them less effective) or if the extra rest would give them a boost.

Bonus questions
How do you manage the rehabbing of a pitcher without the possibility of appearances in the minors? The Yankees are currently considering this one. Andy Pettitte was his usual self (if not better) for a bunch of starts after his comeback earlier this year. He was sidelined by a hard liner to the ankle in June but is nearly ready to return to the team.

The Yankees’ big lead in the standings has evaporated, and they need their veteran for the final stretch. Had he returned earlier, they would have let him throw some innings at various minor-league levels before adding him back to the big-league roster. But in September, the options are either having him throw simulated games (and tossing him into the pennant race without having faced a real opponent in a real game) or letting him get comfortable by throwing a few innings here and there in games that matter.

Finally, there’s a question that Mitchel Lichtman raised a few weeks ago. It’s relevant throughout the year, but I believe it’s even more important with more arms in the bullpen and speed specialists available to pinch run.

Runner on first, lefty on the mound, righty at the plate, close game. The obvious move is calling the bullpen for a right-handed pitcher. Assuming that you properly compared the value of your lefty with the unfavorable platoon to the value of your righty with the matchup on his side (and that’s not always a given), does the platoon advantage to choosing the right-handed pitcher makes up for the southpaw being better at dissuading the baserunner to take off on the pitch?

Some strategic questions have different answers in September. The teams that answer them correctly have a better chance of reaching October—when it becomes a different kind of ballgame again.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
Great questions; I'd love to see them all studied.
I think that small sample size would make any of these studies very questionable though... However, it would seem that contenders facing the situation in your first scenario SHOULD DH their hard hitting / poor fielding catcher.

Remember, of course, that, while the season doesn't matter for teams like the Cubs or the Astros, it does matter for the individual players. A player will not be happy "experimenting" if it could mean worse stats (leading to a smaller future salary)
A lot of good questions.

Please let Joe Girardi know about the 3-catcher thing... he hasn't updated his binder to reflect the fact that he could PH/PR for his catchers now that he has a 3rd guy on the bench.
I've wondered recently if Buck Showalter is handling starter matchups with more than the usual thought. Joe Saunders is pitching in his regular rotation spot against Sabathia this week, but it would have been a good choice anyway. When the O's faced Vernlander on August 17, they sent Tommy Hunter to the mound a day early. Of course, the Orioles "rotation" has been anything but ordinary for a number of reasons lately, and it's hard to predict from day to day which of them is going pitch a great game.
Michael Mediocre was a stud in the PCL, what Happened?
Clyde Callup is my new fave player
Glad to see there is another sufferer in CEST. I love it when my A's play a Sunday 1pm game against someone on the East Coast.