While the Week 15 Power Rankings remained mostly stable, the EOY Projections changed quite a bit, with two completely different teams in the Super Bowl. In addition, we’re figuring in tonight’s Green Bay upset win over Minnesota – the main result there is that in addition to GB being ranked ahead of MIN, we now have NYJ ranked ahead of MIA.
While if the playoffs ended today, beatpaths seedings would be:
AFC: IND, SD, BAL, NE, DEN, CIN
NFC: NO, CHI, PHI, SF, DAL, SEA
If we project out the season so that the matchups play out in line with the power rankings, the seedings as of the end of the regular season would be:
AFC: IND, SD, BAL, NE, DEN, CIN
NFC: NO, CHI, PHI, SF, DAL, SEA
Yep, identical. What’s interesting, though, is that before tonight’s GB=>MIN upset, the end-of-season NFC seedings would have been different, with MIN at #5 and SEA out of the playoffs.
Denver is still ranked ahead of New England, but with the #5 wildcard seed.
Wildcard round:
Baltimore hosts Cincinnati and wins (BAL then leapfrogs SD)
New England hosts Denver and loses
Philadelphia hosts Seattle and wins
San Francisco hosts Dallas and wins
Division round:
Indianapolis hosts Denver and wins
San Diego hosts Baltimore and loses (BAL then leapfrogs IND)
New Orleans hosts San Francisco and wins
Chicago hosts Philadelphia and wins
Conference round:
Indianapolis hosts Baltimore and loses
New Orleans hosts Chicago and wins
Super Bowl:
Baltimore (#1) defeats New Orleans (#10)
Final rankings:
BAL,IND,SD,DEN,NE,CIN,BUF,JAC,PIT,NO,CHI,PHI,CLE,NYJ,MIA,KC,SF,TEN,DAL,SEA, NYG,ATL,GB,MIN,WAS,CAR,HOU,STL,TB,OAK,ARI,DET
I’ve been trying to come up with a response to why, while this is an interesting calculation, I don’t think it should be used to seed playoffs. I’ve also been trying to convince myself that I would feel otherwise even if it did predict my Bears to be #1 in the NFC
I was trying to make the point that it penalizes teams for something they have no control over (the quality of opponents they play), but I think you could make the same case for the current system. I also was coming up with something about how your standing (especially this season) can be greatly affected by how you do outside your conference, which doesn’t seem fair; but again, how is that different from the current system?
I think I’ve come down to two things:
1.) Beatflukes. If you didn’t do beatflukes, or required more evidence for a beatfluke, it might be different. But right now, not only can evidence from the field be thrown out, but it can actually be overruled and claimed that the evidence goes a different way (if I’m understanding the system correctly). That seems wrong to me. We might want to say that the Lions beating, well, anyone, is a fluke, and we might be right. But it did happen, and they deserve to have that result stand up.
2.) I can’t quite articulate this one in one word or phrase. If I tried to guess, I’d probably go with “the NFL may not be well ordered”. The point I’m trying to make is that the graphs try to enforce a strict, one-way, top down view of ability in the NFL. Therefore, they view a loop as anomalous, and throw it out as bad data. However, the NFL is not just a measure of how powerful each team is–different teams match up differently against each team. Suppose A => B => C => A. It looks to beatpaths like that is ambiguous, but it may in fact be the exact truth. Maybe A matches up well with B, B matches up well with C, etc. This is data that is useful, but not to beatpaths.
That last one isn’t quite clear in my head, but that’s the best I can explain it. In any case, I really enjoy the site, but I don’t think I would argue that it should replace W-L records for the playoffs.
That’s all I have to say. Merry Christmas for those who celebrate it, Happy Holidays to those who don’t, and GO BEARS!
I don’t have time for a full reply here, but I thought it might be interesting to see this:
Before this weekend’s games, the playoff seedings according to the NON-beatfluke variant were:
AFC: IND, SD, NE, BAL, BUF*, DEN (Buffalo ranked ahead of Baltimore)
NFC: NO, CHI, PHI, SF, DAL*, SEA (Dallas ranked ahead of San Francisco)
Through today’s Christmas Eve games (but before Christmas Day games), the NON-beatfluke seedings are:
AFC: IND, SD, BAL, NE, BUF, NYJ
NFC: SEA, CHI, NO, PHI, DAL, STL
While the beatfluke seedings are:
AFC: IND, SD, NE, BAL, DEN*, BUF* (DEN ahead of NE, BUF ahead of BAL)
NFC: NO, CHI, PHI, SEA, DAL*, CAR* (DAL and CAR both ahead of SEA)