The smoke has cleard from the battlefield that is Round 2 of March Madness. Here's how the crowd is performing:
The crowd picked 12 of 16 games (75% accurate), bringing the grand total to 37 of 48 (~77.1% accurate). That's 3 games ahead of the closest competition, and 10 games ahead of the last-place-performer (hey, having the West Bracket all but eliminated can do that to you).
In points, the crowd is out front but not in the lead. It stands in second place at 162; 20 points out of first. However, it's pretty much a lock for finishing at least second, as two of my competitors could not possibly score above 162 and a third will max out at 165. Moreover, the current leader has 3 dead games in the next round, and in two more games is predicting a 1-seed to fall. By contrast, the crowd has only a single dead game, and is predicting only a single upset (against a 2-seed at that).
The shallowness of losses continues to be the most amazing performance metric in the crowd's favor. Of 11 missed predictions, only one team was predicted to win in the subsequent round. As a result, the crowd's brackets have shown an impressive resilliency, since past losses almost never affect future accuracy.
More on the crowd's predictive genius at the end of the week (and the start of the next). Until then, we can all bask in the warm glow of James Surowicki's smugness over this latest validation of his theory.
Comments