Let's clear something up. There seems to be a disconnect over what makes for a good Giant Killer.

As colleague Peter Keating and I have explained previously, we're looking at a specific type of upset in our analysis. We're not seeking to evaluate which mid-majors are the best teams; instead, we're looking at the little guys most likely to pull off tourney upsets. Our model spots trends among non-power-conference teams that knock off a favorite seeded at least five spots better, and history has shown that these teams share certain similarities in playing style.

These Davids aren't always the ones you'd expect based on national rankings or the RPI, though. Which brings us to the case of the Harvard Crimson.