Analysis In Medias Res – Week 7

Hello everyone, Tom here with my post, getting things done at the last minute. Nothing like grading 40 atrocious midterms to suck the time right out of your week. Luckily, I found an extra hour from the switch away from daylight savings. I’ll keep it short and sweet.

Stability & Variance

The Week 7 rankings shifted around a good deal last week, but not quite as much as the week before. Given the number of upsets, the amount of change to the ranks is not that bad.

An interesting point of comparison is between Week 7 and Week 4. The raw number of ranking changes was just fewer in Week 7 than in Week 4 (113 vs. 112), but the variance in Week 7 was much higher (7.75 vs. 9.31). What this means is that in Week 4, most teams were moving several ranks, and there were only a very few big moves. In Week 7, most teams were not moving much, but a minority of teams had big moves—teams like Cincinnati’s rise from 17 to 6 and San Francisco’s fall from 8 to 20.

Pick Confidence

TT has already given this week’s official Beatpaths picks. Below are the unofficial rankings of how much ‘confidence’ the system has in making each of the picks:

(out of 100)
EdgePower comparison
(predicted winner – predicted loser)
Arizona-Carolina 46.59 73.86-27.27
Detroit-St. Louis 34.66 42.05-7.39
Indianapolis-San Francisco 31.82 77.84-46.02
NY Giants-Philadelphia 31.82 69.89-38.07
Dallas-Seattle 22.73 65.91-43.18
Denver-Baltimore 22.16 84.66-62.50
San Diego-Oakland 17.61 57.95-40.34
Jacksonville-Tennessee 17.04 49.43-32.39
Minnesota-Green Bay 15.34 70.45-55.11
New Orleans-Atlanta 13.07 75.00-61.93
Houston-Buffalo 7.96 44.32-36.36
Miami-NY Jets 3.41 50.00-46.59
Chicago-Cleveland 0.57 47.16-46.59

As usual, I’ll return to fill in the results of the picks as they come in.

2 Responses to Analysis In Medias Res – Week 7

  1. Tom says:

    Another bonfire of the vanities week for pick confidence. Lots of upsets and missed calls on the ‘safe’ picks.

  2. […] Last week the pick confidence didn’t do well. The top confidence picks were less reliable than the low confidence picks. We’ll see if the shuffle in the ranks gives us greater accuracy in making picks. […]

Leave a Reply

Your email address will not be published.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