clock menu more-arrow no yes mobile

Filed under:

Better rumor-mongering through automation

Now that the NHL Trade Deadline is in our collective rear-view mirror, it's time get to down a serious review of performance. No, not of the GM's who swapped assets like eBay addicts, but instead of the "Anonymous Hockey Blogger" who plies his trade over at HockeyBuzz and purports to dish out inside information.

As I noted a few weeks ago, the boys at have been tracking the success rate of the rumors that get floated over at the site. My challenge was a simple one; that a few rudimentary rules could kick out a list of trade possibilities that were likely to be more correct than those spun by the old fashioned way; presumably involving late-night hours perusing message boards, a Ouija board, and an assortment of mind-bending drugs.

The rules were simple. First, I took the pending UFA's on teams that were outside the playoffs. Next, I created 17 scenarios for each. The first one had them remaining with their existing team, while the other 16 involved them going to one of the teams that was, at the time, in playoff position.

There were 104 such players, thus yielding 1,768 scenarios (which I published for all to review in a Google Spreadsheet). In the end, 94 of those turned out correct, yielding a 5.5% success rate.

Yeesh, that sounds kinda bad, doesn't it?

Maybe not compared to Eklund's 3.2!