Devon Young sends this via e-mail:
1954-1995 (32 seasons) – 13 hit streaks of 29+ games.
1996-2009 (14 seasons) – 14 hit streaks of 29+ games, and counting.That’s a HUGE sudden increase. More than doubles. And the 14th season (2009) isn’t even half done yet. There could be another this year,
Two reasons come to mind off the top of my head. One is more teams. The majors added four teams between 1993 and 1998, providing more players in the pool to possibly put together a streak. The second is more offense. I like to use something I call Hit Average when I work with hitting streaks. That’s simple hits divided by total plate appearances, or the probability of getting a hit. (Note that OBA does not use all plate appearances, because OBA is a collective bargaining statistic.) From 1957 to 1995, the hit average was .228. From 1996 to 2009, the hit average is .236. So the probability of a player getting a hit in a four at bat game prior to 1996 was 0.646. In the later period, that’s up to 0.66. That doesn’t seem like much, but remember, we’re talking about long streaks here, so we’re multiplying. The chance of a player hitting in the next 29 games goes from .00000314 to .00000584, nearly double. Now, figure that over all possible players playing at least 29 batting games, and you can see how batting streaks would have increased. A player with 150 games played has 122 possible 29 game game streaks. If there are 5 players with 150 games played per team, there would be a .049 chance of a 29 game hit streak among those players pre-1996 in a 26 team league. There would by a .101 chance post-1996 in a 30 team league.
Now, not all hitters are the average hitters. Better average hitters means better elite hitters, so I’m not surprised we’re seeing one such streak a year instead of one every ten years. Since 1996, 14 players with at least 1500 plate appearances have hit averages above .280. From 1957 to 1995, only 11 fit that bill.
More teams, better hitters, more chances all lead to more long hit streaks.
(Note: the Day by Day Database currently goes back to 1957, so that’s why I used that as the date for my calculations, not 1954, as in Devon’s letter.)
Maybe all we need for someone to break DiMaggio’s streak is to increase the majors to 40 teams!
I went back to ’54, using the Play Index’s streak finder over at Baseball Reference.
One factor left out is diluted pitching. Since all the expansion was in the NL, I checked the league composite ERA for ’92 and ’93. In ’92 it was 3.50, and in expansion year one it was 4.03, over a half run higher.
In fact, in the 16 years beginning with ’93, the league composite ERA has been over 4.00 every year, peaking at 4.63 in 2000. In the 16 years ending in 1992, the NL composite ERA was over 4.00 only once, in ’87. The strike zone was enlarged in the off-season, and the composite ERA dropped to 3.45 in ’88.
I have to think that had an effect on hitting streaks, extra base hits and home runs. But then, it might undermine the “steroid era” meme, so let’s not talk about it.