The Thing Fish Ratings, Week 11
Well, after this week’s games, I find myself having to admit the necessity of making some tweaks to the system. I think the established polls tend to rely too heavily on raw won-loss record and, partially because of that, to overrate mid-major teams. However, it’s now clear to me that my original system goes too far in the opposite direction on both counts. The last straw was finding that Stanford’s second loss was going to leave it STILL in first place, ahead of four P5 undefeateds, three of whom had played respectable nonconference schedules.
Hence, tweaks. I’ll post the tedious details in a spoiler box for brevity, [SPOILER]
The original rules are pasted below, with deleted parts in italic and new parts in bold
2 points for beating a P5 team.
2 point bonus for beating a top 25 team (per these ratings) – this is retroactive, so teams will be gaining and losing points all season as their former opponents move in and out of the top 25.
1 point bonus for winning on the road.
1 point bonus for winning by 15 or more.
No points for beating a non-P5 FBS team, and boy do we need a less clunky way of saying “FBS-but-not-really” – but 1 point bonuses for road and decisive wins still apply.
No points ever for beating an FCS team, and losing to them incurs a four-point penalty, plus one-point bonuses for home and/or decisive losses.
Exception: FBS minor league teams and FCS teams which win* a total of at least 10 regular season games * **their divisions (or conferences, if not divided) **shall be treated as if they were the next level up. Such teams shall also receive a four point bonus.
FBS teams which finish last in their divisions shall be counted as the next lower level team for purposes of calculating their opponent’s strength of schedule.
One point deduction for any loss to an FBS opponent.
The Big 12 regular season champion gets a four point bonus to compensate for the lack of a championship game. Any independent which finishes with one or fewer losses also gets a four point bonus.
The first and only tiebreaker is the whim of Thing Fish, 2 points for beating a P5 team.
2 point bonus for beating a top 25 team (per these ratings) – this is retroactive, so teams will be gaining and losing points all season as their former opponents move in and out of the top 25.
1 point bonus for winning on the road.
1 point bonus for winning by 15 or more.
No points for beating a non-P5 FBS team, and boy do we need a less clunky way of saying “FBS-but-not-really” – but 1 point bonuses for road and decisive wins still apply.
No points ever for beating an FCS team, and losing to them incurs a four-point penalty, plus one-point bonuses for home and/or decisive losses.
Exception: FBS minor league teams and FCS teams which win a total of at least 10 regular season games shall be treated as if they were the next level up.
The Big 12 regular season champion gets a four point bonus to compensate for the lack of a championship game.
The first and only tiebreaker is the whim of Thing Fish**, who will try to make decisions based on objective factors rather than on his personal preferences. Also, teams may receive a temporary one point bonus if needed to bring them into a tie with a team they beat (and thus eligible to be ranked ahead of them by tiebreaker.**
[/SPOILER] but basically I have added a small penalty for losing (where previously losses were bad only because they represented missed opportunity), and given some bonus points to the best mid-major teams.
I also figured, since I am giving extra credit for beating the* best* P5, mid-major, and FCS teams, I should logically give less credit for beating the worst teams in each category. This change should affect relatively few teams, but could be significant for a few (looking at you, Wisconsin and Bowling Green). Unfortunately, this makes the system too complicated to completely revise every week, so (except for those involving ranked teams) these adjustments will be made only towards the end of the season, once the best and worst teams at each level can be definitively identified; a few have already been made, but most remain to be determined. This could lead to some dramatic fluctuations in the ratings at the end of the season, which is inelegant but, at least, exciting. Here are the teams already covered by this rule:
P5 teams counted as mid-majors: Georgia Tech, South Carolina
mid-major counted as P5: Bowling Green
mid-majors counted as FCS: Charlotte, E Michigan, Wyoming, Hawaii
FCS counted as mid-major: Jacksonville St, Colgate (point for Navy!), Dayton, McNeese St (hurricane deprived LSU of a potential point)
I also figured that, since I am giving the Big 12 champion bonus points since I don’t want to punish it for not having a conference championship game, I should also do the same for independents that finish with one or fewer losses. The most obvious beneficiary of these changes is Notre Dame; where I previously thought that their chances of finishing in my Top 4 were nonexistent, they are now clearly in the group that could make it if they win out and get a little help (including strong finishes by Temple and Navy).
So, here’s this week’s new and, I hope, improved list!
- Iowa!
- Clemson
- Stanford
- OSU
- Oklahoma
- *The *OSU
- Alabama
- (gulp) Notre Dame – but big gap separating it from #7
- Michigan State
- Michigan
- UNC
- TCU
- Utah
- Baylor – will these people shut up now that they have lost? Probably not.
- FSU
- Oregon Holy shit! Holy shit! Holy shit! Holy shit!
- Bowling Green – all hail the MAC East champions!
- USC
- Florida
- Northwestern
- Wisconsin – big tilt with Northwestern next weekend
- UCLA
- Pitt – maybe back on track with big win at Duke
- Wazoo! – overcoming early loss to Portland State
- Houston – mid-major makes grade even without (yet)benefiting from new rules.
on the bubble: Georgia, Memphis, LSU, Ole Miss
coming up: Pitt, WSU, BGSU
going down: One Mississippi, Two Mississippi, TAMU, LSU – the trio which kept each other ranked last week goes down together this week, and LSU plummets after wins over Eastern Michigan and South Carolina are downgraded. SEC now down to two teams!
You might well say “This is change we can believe in? Stanford is* still* in third place, ahead of two unbeatens!” But I would argue that, if you list the teams that Stanford, OSU, and OSU have beaten thus far, Stanford’s list is clearly superior, even if shorter. Stanford has beaten three P5 teams which have winning conference records, the OSUs one each. Stanford’s nonconference schedule is clearly superior, even without counting the Notre Dame game they haven’t played yet. But more importantly, I now feel pretty confident that, unless a lot of weird things happen WRT margin of victory, both OSUs (and Oklahoma) will pass Stanford if all three teams win out; their schedules aren’t THAT terrible, they just play their toughest opponents in the last weeks of the season. Prior to these tweaks, I don’t think they would have.
Clemson is another story – right now they are ahead of Stanford only by virtue of tiebreaker. They also have three wins over winning P5 teams, but two of them were by a total of only five points, and their only chance for another quality win is the ACC title game, whereas Stanford has ND and probably the Pac-12 title game. So, even if Clemson wins out, I think they will finish behind a two-loss Stanford and quite possibly out of the top 4; their schedule really is THAT terrible. Of course, my fervent hope is that it won’t matter, because Stanford will lose to Cal this weekend, leaving the door open for Oregon to return to the Rose Bowl.
Here is a stupid article bewailing the Pac-12’s 9-game conference schedule, which makes it harder to get a one-loss champion and thus puts the Pac at a disadvantage in reaching the playoff. He suggests a lot of scheduling reforms the NCAA could demand to fix this problem, but wouldn’t it be simpler to just expect that the people choosing the playoff teams be smart enough to realize that a two-loss team might, under the right circumstance, have a better overall resume than a one-loss team? Sadly, I think he is probably right that in the real world, the Conference of Champions won’t be adding another one to the list this year. Don’t have a link, but Teddy Greenspan in the Chicago Tribune today was gloating over the Pac-12’s elimination, and also suggested that Notre Dame losing at Stanford would constitute a major upset!
