2006 Week 4 Beatpath Graphs

The official picks were a horrid 7-7 this week. These were the picks from 2005-2006 combined, and they were actually outperformed by the picks that came from 2006 only. Since we’re through the first four weeks of the season now, we will be continuing with 2006 only for the rest of the season – hooray! For the season, we are 35-25, or tied for 12th in King Kaufman’s Panel O’ Experts.

So without further ado, here are the two beatpath graphs. First, the beatpath graph for 2005-2006 combined:


But more relevantly and more interesting, here is the beatpath graph for 2006 only. The interesting news – we have two new beatloops: ATL=>CAR=>NO=>ATL and DAL=>WAS=>JAC=>DAL. So all those victories are canceled out of the graph. More interesting: there was originally a beatloop of ARI=>SF=>STL=>ARI, but St. Louis found an alternate beatpath to San Francisco since San Francisco is playing so poorly, so it’s our first “beatfluke” of the season. STL=>ARI=>SF is restored to the graph.


The upcoming power rankings might look funny to people, but that is what is cool about this system – it reflects the power rankings for the teams as they have actually played so far this season. In other words, the power rankings for 2006 only (posted later this week) are the most accurate power rankings possible for reflecting the game outcomes for this season so far – a combined record of 56-4!

4 Responses to 2006 Week 4 Beatpath Graphs

  1. JT says:

    It took a little searching to remind myself what constituted a beatfluke, but I finally found it at http://beatpaths.com/?p=89

    You might want to link this in with the “how this all works” section, or write up a new explaination, for new people, or those of us who can’t quite remember how it all works.

    Otherwise, I just want to say again that I rather enjoy this ranking method, and look forward to seeing the updated results each week.

  2. ThunderThumbs says:

    Yep, you’re right – I neglected to include the beatflukes definition – I’ll add that in. Thanks!

  3. PatsFan says:

    I am curious and don’t want to look around for it….What four games does the system get wrong (in the retro-predictions)? Do they have something to do with the beatfluke?

  4. ThunderThumbs says:

    Yes, they’re always games that have to do with beatloops and beatflukes.


    In each case, the loser is still ranked ahead of the victor.

Leave a Reply

Your email address will not be published. Required fields are marked *