Letters to the Editor
anon anon
Published Letters: 47 Editor's Choice: 3
-
Did Scooter break the law?
[Read the article: What comes next for Karl Rove?]
[Read more letters about this article: Here]Does Judy Miller have a security clearance? If she does, giving her information isn't prima facie evidence of a security violation.
-
The explanations aren't very complicated
[Read the article: Is Karl Rove still on the hook?]
[Read more letters about this article: Here]"Mr Rove, did you disclose Mrs Wilson's CIA affiliation to the press?"
"No, I didn't."
"Mr Rove, are you aware of the penalty for perjury?"
"No, but I'm sure it isn't more than the penalty for treason."
-
Sorry, misspelled Capt Brinson's name.
[Read the article: Abu Ghraib officer fights reprimand]
[Read more letters about this article: Here]Sorry, misspelled Capt Brinson's name.
-
I got yer luck right here
[Read the article: King Kaufman's Sports Daily]
[Read more letters about this article: Here]Kaufman: "I think luck affects a season in a lot of ways. There's injuries, sure. But what about your opponents' injuries? You get to play the champs six times when their slugger's on the DL, and your division rival has to play them six times three weeks later at full strength. That's dumb luck. How about rainouts screwing up your pitching rotation? Some team (the Dodgers was it?) had to play double-headers on three straight days in a pennant race one recent year. Bad luck for them, good for whoever they were in the race with."
King,
The biggest factor of true luck in baseball is the sequence in which good things happen for the team at the plate (whether hits, walks, sacrifices, or defensive errors).
A player can be expected to perform at a given level over the course of a season, but can't be expected to produce a good offensive play (hit, walk, or error) in any specific situation. The odds are against it. If a good thing happens in any specific situation where it will especially help the team, it's probably luck. (There's a lot of discussion of the "myth" of situational hitters.)
For example, if one good thing happens in each inning, the team may score no runs at all (except for home runs). If the same nine good things happen in one inning, that team will score at least six runs.
Except for home runs, about three good things must happen before the first run can be scored in an inning; after that, each additional good thing in the inning may produce another run. That's the source of the traditional emphasis on the big inning. (That's also why a failed steal has such a high price. A stolen base doesn't much increase the chances of scoring a run, but failure simultaneously negates the unlikely good thing that already was in the bag, and makes an out.)
Here's a simple example of the math.
Let's assume that a good thing (hit, walk, sacrifice, or error) randomly happens half the time for each player at bat, and three good things are required for the first run:
* 12.5 percent of innings: RBI 1 before Out 1.
* 12.5 percent of innings: RBI 1 between Out 1/2.
* less than 12.5 percent of innings: RBI between Out 2/3 (with three outs, run probably must be scored on hit or error, so call the percentage 8.333.)
With the preceding example, the first run is scored in every third inning played. That still wouldn't produce much random variation in game results. Teams would hover pretty close to their median level of performance, like football; you don't often see 56-14 games between football teams that are well matched on paper.
The very random (lucky) score variations in baseball occur after the first run scores in the inning, because two random factors are multiplied:
* The occurence of a good offensive play.
* The number of outs remaining in the inning.
After the first run, every good play probably scores another run. For my example, after the first RBI, you basically have a fifty-fifty chance of scoring a run on every batter or making an out (probably one-in-three before the third out). However, there's basically nothing much that can be done to increase the likelihood of a successful play on a specific at-bat when needed. For practical purposes, the likelihood of a good play after the first RBI is as random as in any other game situation.
How many good plays will be made after the first RBI in the inning? That depends on *two* factors that are essentially random:
* The batter's result on any individual at bat
* The number of outs that can be made in the rest of the inning
Multiplying these two random factors produces wildly random results for specific games between equally matched teams.
For my example, here are the approximate average scores after the first RBI. A prime scoring opportunity with no outs is more valuable than the other two cases put together:
* 0 outs: 2.5 more runs (3.5 total avg for each such inning)
* 1 outs: 1.5 more runs (2.5 total avg for each such inning)
* 2 outs: .5 more runs (1.5 total avg for each such inning)
The preceding example produces an average of about seven runs per game for each team. However, the occurence or absence of the big 0-out inning makes a profound impact on the the score of any particular game between these teams, and each such inning is so scarce that you can't count on it in any specific game.
If the big 0-out inning occurs randomly about once per game for each team, then the following pattern will occur for each team in my example:
50 percent of games: same number of big innings; no advantage to either team. (Time for small ball.)
25 percent of games: Team has more big innings, and has a huge advantage (50 percent of the other team's average score) that is unlikely to be made up by other kinds of innings.
25 percent of games: Team has fewer big innings, and has a huge disadvantage (50 percent of the other team's average score) that is unlikely to be made up by other kinds of innings.
For the preceding example, each of these evenly matched teams is essentially uncompetitive in 25 percent of its games and can expect to lose such games by an average of at least three runs.
Sometimes, it just isn't your day.
-
Tell my statistics to shut up
[Read the article: King Kaufman's Sports Daily]
[Read more letters about this article: Here]Sorry, need to correct this paragraph in my preceding letter:
For my example, here are the approximate average scores after the first RBI. A prime scoring opportunity with no outs is ALMOST AS VALUABLE AS the other two cases put together: