Aren't these 2008 Dodgers weird and streaky? While they seem quite worthy of their 0.500-ish record, it sure seems like they get there in a backwards fashion. A mere two weeks after sweeping the Phillies, they get swept by the Nationals. Not to mention that their current 3-game winning streak follows a 1-10 stretch, which itself followed a 6-1 stretch. What gives?
Well, maybe not as much as you think. I put this perceived weirdness to the test, taking a simple look at the Dodgers' record against six categories of opponents (Real Bad, Bad, Not Great, Decent, Good, and Really Good), based solely on their current winning percentage (as opposed to their winning percentage at the time of each game). Here's the result (through yesterday's games):
So while there's some minor backwards correlation (the Dodgers seem to fare slightly better as teams go from not great to decent to good, and do better against bad teams than real bad teams), in general it's not really far off from what you'd expect. They do the best against bad teams, and the worst against really good teams.
Now, regarding their perceived streakiness, I took a look at their winning % after a win vs after a loss, as well as after a 2-win/loss streak and after a 3-win/loss streak. Here's the result:
|after 1 W||0.507|
|after 1 L||0.471|
|after 2 W's||0.455|
|after 2 L's||0.486|
|after 3 W's||0.571|
|after 3 L's||0.421|
Acknowledging there's not enough data to reach a firm conclusion, I won't try to do so. But judging from the fact that their win % is lower than their overall win % after both single losses, 2 consecutive losses, and 3 consecutive losses, I think there may be some validity to their perceived streakiness, at least when it comes to losing*.
*note that in spite of any validity to perceived winning or losing streakiness, I stand by my assertion that any correlation between higher or lower win % and a preceding win or loss streak isn't caused by the preceding winning or losing, but is a result of either 1) how good the team is relative to its opponent at the time the games are played (including injuries, suspensions, or, as some have noted, who's pitching or site of game, etc), or 2) randomness.