Analyzing the Rankings – Week 9

Tom here with commentary on the rankings changes from Week 9. The number of rank changes is actually slightly up from last week. This is a bit surprising really. There were a number of significant upsets leading up to the Week 8 rankings (all of which contradicted a Beatpick), whereas there were only three games that Beatpaths might consider an upset this past week. Yet those three games—Tampa’s victory over Green Bay, Tennessee’s win over San Francisco, and Pittsburgh’s defeat of Denver—ended up causing more total changes to the rankings.


As you can see, the variance is up. This means that there is even greater disparity this week between the majority of teams that moved only a very few ranks if at all and a few teams that moved significantly. I’ve added a second line to the variance chart, displaying the average ranking change alongside the variance. This gives you an easy visual way of comparing how far the ‘big movers’ are deviating from the norm.

The two teams enjoying the largest reassessments by the Beatpaths system this week are Buffalo and Carolina. Both were plucked from the dregs of the rankings (26th and 30th) and re-ranked as solidly middle-of-the-pack teams (14th and 17th). Kenneth explained how Dallas’ victory over Philly caused the Beatpaths system to reassess Carolina’s strength (and thus Buffalo’s strength, which is based on Carolina’s).

Pick Confidence – Week 10

Last week, pick confidence worked pretty well, getting only one high confidence pick wrong (Green Bay losing to Tampa Bay), while getting two lower confidence picks incorrect. As the season progresses, and the rankings of the teams become more accurate according to Beatpaths, we should see this pattern continue: about 2/3rds of the incorrect picks consigned to the low confidence games.

Here is the pick confidence ranking of Week 10’s matchups, from highest to lowest:

Matchup
(Winner-Loser)
“Confidence”
(out of 100)
EdgePower comparison
(predicted winner – predicted loser)
Result
Denver-Washington 52.91 75.58-22.67
New Orleans-St. Louis 47.09 74.42-27.33
Minnesota-Detroit 44.77 73.84-29.07
Miami-Tampa Bay 37.8 56.40-18.60
Baltimore-Cleveland 35.46 67.44-31.98
Dallas-Green Bay 19.77 69.77-50.00
Atlanta-Carolina 16.86 65.12-48.26
Buffalo-Tennessee 13.38 52.33-38.95
Oakland-Kansas City 13.37 39.53-26.16
San Francisco-Chicago 9.88 46.51-36.63
Arizona-Seattle 9.3 45.35-36.05
NY Jets-Jacksonville 6.4 44.19-37.79
Cincinnati-Pittsburgh 2.32 71.51-69.19
San Diego-Philadelphia 1.74 63.95-62.21
Indianapolis-New England 0.58 69.77-69.19

In the biggest match of the week, Beatpaths seems to give Indianapolis only the slightest edge over New England. Less hyped, but just as big a game according to Beatpaths rankings, is the Cincinnati-Pittsburgh division showdown, in which Beatpaths gives a tentative preference to a Cincinnati repeat of their Week 3 victory. As usual, I’ll update the results on game day so we can assess how well pick confidence captures the predictive power of ranking teams by ‘edges.’

3 Responses to Analyzing the Rankings – Week 9

  1. Alternator says:

    I believe you mean Pittsburg’s defeat of Denver, not Cincinatti, right?

  2. Tom says:

    Alternator: quite right. That’ll teach me to post at 1 AM…

  3. […] the system’s confidence in making picks is lower this week than in previous weeks. Oddly, last week’s picks suffered the most in the mid-range, rather than in the low confidence range. We’ll see how […]

Leave a Reply

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