Unexpected At-Bats
Posted by Raphy on February 25, 2008
Very often there are players who make an appearance in a game for a particular moment. Their manager decides that their skills are needed for a brief moment in the game, but no longer. They are asked to pinch hit and take a seat or to pinch run and then come out of the game. The pinch hitters' evenings are supposed to last for one plate appearance, the pinch runners' not even as long. This is the norm. Occasionally however, the bizarre happens. A deluge of offense allows the team to bat around and the pinch hitter gets to have an encore performance; the pinch runner a chance to actually pick up a bat. Let's take a closer look.
Using PI to find all the players who appeared in a game only as pinch runner and had at least one plate appearance or those who appeared as pinch hitters and had at least 2 PA's presents two problems.
1) The information for games in which there is no play by play data available doesn't seemed to be marked accurately regarding this stat. 2) Any player who pinch hits or runs for a DH is not assigned the DH position even in subsequent at bats.
The first point is minor and we'll work around it. The second is major and will prevent us from examining any AL games since 1973 .
1) Here are the instances from 1957-1972 in which a pinch runner has had a PA. As you can guess from their low scores, the McCarver and Murrell games are probably errors, and in fact following the game links shows you that Retrosheet does not have the play-by-play data for those games. I believe that the others are legitimate.
2) Since 1973 in NL only games, we have the following list. Eric Davis successfully did it once and then almost exactly 8 years later, did it even better. How about that Nino Espinosa game!
3) Pinch hitters who have multiple PAs are obviously more common. This is the complete list 1957-1972 and this is the NL only list 1973-2007.
I think the winner in this category has to be John Cangelosi. Cangelosi wasn't particularly successful in these games, but he played in an amazing amount of them. Even discounting his 3 seasons in the AL and all his interleague games, Cangelosi managed to do this an amazing 6 times in his career, including thrice in 1987!
EDIT: As is discussed in the comments, there is a way to find all the "unexpected plate appearances" in any season by using the "Team Batting Event Finder", searching for all Plate Appearances and setting the position to "??". Here are the results for 2007. Of course, once you are doing that, you can further limit your search. For example you can find the grand slams of Willie McGee and Bill Robinson in their second go around after pinch hitting.
February 26th, 2008 at 8:41 am
i was at a Toronto v Philadelphia game this past season (in philly) and saw this happen (http://www.baseball-reference.com/boxes/PHI/PHI200705190.shtml) ... the people in front and behind me were both having arguments about what the ph was the second time up. i was more than disappointed that i couldn't jump in with an answer... as i tend to be that guy at games who jumps in to neighboring conversations with the random fact or stat that proves one person right...
February 26th, 2008 at 10:12 am
yes good question. what is he the second time up. he does not get credited with another pinch hit does he. he probably gets credited with just an at bat. does he get credited with playing the field? how will it all be credited?
February 26th, 2008 at 10:58 am
I can't tell which guy you're referring to in that boxscore, but a player is only a pinch hitter once in a game. After that, he's playing whatever position he's put at. Following AB do not count as pinch hitting AB.
February 26th, 2008 at 1:33 pm
The stats for PH who come up a second time are listed as "other" when their stats are broken down by defensive position. For example, if you look at the Defensive Positions sections in the Cangelosi splits for 1987, Cangelosi has PA as a right, center and left fielder as well as a pinch hitter. He also has 3 PA listed as "other". Presumably these are the three plate appearances that I mentioned in the post.
February 26th, 2008 at 2:11 pm
I've actually been working on this very exact thing in preparation for the 2008 season, and what a pain it is.
It appears that the convention is that PH or PR for the DH results in a dual counting. It is both a PA or a Game played as a PH (or PR) and as a DH. Notice these are double-counted for Curtis Thigpen on ESPN's stats. I asked Tom Ruane at RetroSheet about it and they intend to double-count these, but appear to have a bug.
A non-dh ph coming up the second time would have position 'As Other'. Kind of a twilight zone of positional appearances.
I'm cleaning this up and the DH issue as well in the next release (with the 1956 data).
February 26th, 2008 at 2:11 pm
This seems to be the key to finding these at bats for games with DHs. I will modify the original post to reflect this.
February 26th, 2008 at 3:36 pm
I didn't comb in detail through these, but the 2007 results showed something interesting. It's not like these games are terribly common, although I suppose as offense has increased the incidence of batting around has increased making it more possible to have an inning like this. In 2007 there were 2 days (May 25 and August 16)where there were 3 of these innings.
And about all you can say is....huh.
February 26th, 2008 at 3:41 pm
Regarding Murrell and McCarver. These are errors in the retrosheet files, they are listed as PR and Dave Smith said they should be PR's.
Re: 7. I think the likely issue is that they are PR's for the DH and then stayed on as DH. They should have DH marked as a position played, but don't in the DB.
February 26th, 2008 at 4:02 pm
Re: 8. I'm confused to your comments regarding 7. All the games on those dates are NL games. Why would there be a DH?
Also, I'm confused as to why the NL games that are included in the link from the addendum are not returned in the searches from the original post. Was the record keeping for 2007 different in some way?
Thanks.
February 27th, 2008 at 2:10 pm
2007 is different because I was compiling the season as it went from a different set of files. Essentially, I created my own retrosheet files and then built them. There was a bug that caused this issue. I've cleaned it up and should have all of these fixed before the season starts.
February 27th, 2008 at 11:10 pm
I was thinking this might be about situations where a PH is announced and then pulled for another PH. Has it ever happened more than once in a single PA?
Also there is the bizarre instance of a PH appearing at the plate (with 2 outs) and maybe hitting a few long very foul balls that come close to being HRs and even popping up a foul ball and having it dropped for an error (to give the "PA" a legitimate 'event'), and then the runner at 1B is caught stealing to end the inning/game, making the appearance a statistical non-event.
February 28th, 2008 at 10:23 am
on comment 11. right that is wierd. an error obviously would have to go in the books but no official at bat is recorded. i would guess. but question now. Say that happened. A batter hit a foul ball that was dropped for an error then the baserunner got caught stealind during the at bat. Will that batter still be available for the rest of the game. I would think he is still available in the game. Ya I am pretty sure he is. But if someone can clarify.
February 28th, 2008 at 2:29 pm
[...] recent comment by fabio in my “Unexpected At-Bats” post got me thinking about pinch hitters who were [...]
May 29th, 2008 at 9:48 am
[...] article on BB Ref about this very thing. Not that it helps about what you call them necessarily. Stat of the Day Unexpected At-Bats [...]