In the 13th inning of tonight's Marlins-Braves game, right after Andruw Jones hit a slider about 750 feet to left for his 25th home run, Buck Martinez started talking about how before Chipper Jones was hurt, Andruw was hitting like .240, and had no power, and blah blah blah. Then, Chipper gets hurt, and Andruw suddenly turns it on and has a massive June and is going to win an MVP.
I've bitched about this in previous posts, but I am now ready to declare this my biggest beef with sportscasters: the inability to distinguish causality from coincidence.
For example.
Is there any actual proof that Chipper Jones getting hurt in any way helped Andruw Jones, physically or psychologically? Is there any explanation that makes any kind of sense? For that matter, can anyone even make one up? Martinez and Morgan (and others, I assume) just assert that the two are causally linked, but how? Why? If Andruw Jones is having a terrible April, and then one day he drinks a Diet 7-Up and watches the movie "Down Periscope," and then he hits fifty home runs, is it because he drank a Diet 7-Up and watched "Down Periscope?"
Or, for a better analogy, what if Andruw Jones is having a terrible spring, and then SOME OTHER GUY ON HIS TEAM, like, say Julio Franco, reads a "Bloom County" collection and punches his dog in the nose, and starting the next day Andruw Jones goes on a tear and hits 50 home runs, is it because anti-pooch feelings and pro-adorable penguin feelings have somehow passed trough Franco's body and unleashed something buried deep within Andruw Jones?
Why do people insist on finding links like this?
When the Mariners won 114 games a few years back, I heard one of their announcers (don't remember who) cite what he called a "telling" statistic: when Mark McLemore scores a run, the M's are [some kind of very good record]. I remember calculating the corresponding winning percentage, and finding it was within like .015 of the M's overall winning percentage. In other words, when McLemore DIDN'T score a run, the M's record was basically identical to when he did score a run. Thus, the stat was the exact opposite of "telling."
There are tons of these things. Like when a stat flashes on the screen that so and so's ERA in innings 4-6 is 4.50, but from innings 7-9 it's an incredible 1.35!!! Of course, as anyone could realize if (s)he thought about it for two seconds, the only times a pitcher goes 7+ are when he is performing well, hence his ERA will be lower. Or, this team must have a great bullpen, because it's an amazing 42-3 when leading after eight innings! Great. The league average winning % in that scenario is like 95%.
Starting now, I lead the crusade against misattributed causality! Who's with me?!!? No one?!?!? Awesome!!!!!!!!
Labels: buck martinez
