surprisingly for a "Vapid" Unicorn, you speak hard things me no understand

i wont understand mathematical correlation. i was once a math prodigy, but the drugs i od'ed on changed that.

I think what Vapid is saying is that even where there is a strong correlation (and LSAT/GPA is very predictive of 1L success), it only takes a few outliers (i.e. the person that tested poorly on the LSAT but aces their law school classes, or the 180 scorer who developed a drug habit during 1L) to blow up any hypothesis you might have drawn.

So, a correlation number may seem low but can actually be extremely predictive. And even what seems like a major difference between two numbers of correlation may actually mean very little in terms of predictive ability.

Did that help?