(10-25-2014 09:23 AM)MannyCLE Wrote:  ...but in reality, those teams quite literally haven't accomplished anything significant enough to warrant Top 32. I have no idea how Jerry Rig Jeff concludes Michigan State is #11.

Neither has Marshall.

Jeff would have no idea how you conclude Marshall is #9 (or now 11).

Are the rankings objective if the creator adjusts (includes some data, omits other data) them to fit what he feels the outcome should be? I think that is what most CFB formulas are, and I think most computer rankings, like yours, are well-intentioned. But they are all subjective in that regard. There is no definitive formula. Each compiler has made a lot of subjective decisions along the way, as you have and as you continue to do. I don't see them as being better than well-intentioned polls with well-informed voters who make decisions about who is best by watching the games.

It's possible to game either one pretty easily. Ultimately, we are left with our own judgments about what we see either on the field or in anyone's rankings. I guess I fall on the side that thinks that stats and numbers tell a very limited story about what is a very complex game on the field.

(10-25-2014 10:47 AM)wleakr Wrote:  

(10-25-2014 09:23 AM)MannyCLE Wrote:  ...but in reality, those teams quite literally haven't accomplished anything significant enough to warrant Top 32. I have no idea how Jerry Rig Jeff concludes Michigan State is #11.

Neither has Marshall.

Jeff would have no idea how you conclude Marshall is #9 (or now 11).

I will not respond to you again. I've already explained and re-explained this you explicitly.

repost:

Marshall hasn't played any team of serious strength??

They've already got two (2) WVWTs. And one (1) win against a team with a 0.500 record (which counts towards OPPA++, because if that game wasn't played, the loser would have a winning record).

(To clarify:
- Wins against FCS teams with winning records DO NOT count as WVWTs)

As I stated earlier, Marshall has already achieved what over half of the FBS will not achieve all season --- secure two (2) WVWTs. (subject to change certainly as seasonal outcomes vary)

Michigan State has not accomplished what Marshall has relative to the rest of the FBS. And when the do, they will rise to the ranks of Marshall.

(10-25-2014 10:59 AM)axeme Wrote:  Are the rankings objective if the creator adjusts (includes some data, omits other data) them to fit what he feels the outcome should be?

Absolutely yes they can still be objective. I publish my criteria because criticism is a necessary part of development.

When MileHigh told me to look at Western Michigan. I looked deeper, and found a way to strengthen my criteria (that applies every FBS team). I don't care where WMU ended up.

And then again, when MileHigh told me to look at Michigan State, I looked deeper, and again found a way to strengthen my criteria (that applies every FBS team). I didn't care how Michigan State was affected.

But whenever the guy who keeps complaining about Marshall being ranked Top 16, I just hear noise. The criteria gets stronger, and Marshall stays put. (so does Ole Miss, Mississippi State, et al)

I could give every blue team 1 pt and it would be objective (would it not??). Then I would leave it to you guys as to whether its fair that I give every blue team a point. Is that reasonable criteria? How can I justify that? If I can't then we can agree giving every blue team 1pt is terrible criteria.

But when I award points based on wins vs winning teams, and I tell you that every MAC team who has had 3 WVWTs since 2005 has played in the MAC Championship game (except for one year when Kent and NIU had 4) can you really call that weak criteria??

Don't judge my objectivity based on the number of revisions I make to the rankings. That's improvement. Judge my objectivity on the method and the criteria.

Which specific criteria of mine are weak? What criteria am I missing, and why is that specific missing criteria strong?

So what happens to Marshall's standing when their 2 WVWTs decreases(i.e. Akron loses today, making them no longer a "winning" team at 4-4)? Dropping Marshall to 1 WVWTs...will that drop Marshall to Michigan State's level?

BTW, I'm not against Marshall...if they beat every single team that's in front of them, they should be at least #10 at the end of the entire season, because going undefeated is tough regardless of conference.

But not now.

My OPPA++ rankings correctly picked the winner of 36 of 46 FBS match-ups (78%) this week. Sagarin correctly picked 37 of 46 FBS match-ups (80%) this week. Seventy-four percent of our picks this week matched.

The statistical advantage of using Sagarin's overall ratings instead of my OPPA++ rankings this week to predict win/loss was +0.7%.

The results speak for themselves.

(10-26-2014 10:36 AM)MannyCLE Wrote:  My OPPA++ rankings correctly picked the winner of 36 of 46 FBS match-ups (78%) this week. Sagarin correctly picked 37 of 46 FBS match-ups (80%) this week. Seventy-four percent of our picks this week matched.

The statistical advantage of using Sagarin's overall ratings instead of my OPPA++ rankings this week to predict win/loss was +0.7%.

The results speak for themselves.

That's a decent week. But I think it will fall dramatically. I could make my own personal rankings based on no math every week and have a decent win % based on my own thinking power.

Basing the system off of the 2013 final poll where Toledo was 36th...come on. I know we got shafted for a bowl, but we weren't the best in the conference!!! Loss by 17 to NIU at home. Loss to Akron... decent show vs. Mizzou, and close loss to BSU weren't bad losses but if BSU is where you had them ranked (72)? It should be viewed as a bad loss! Same with (85) Florida, with or without Driskol!

If Toledo had 4 losses vs. teams behind them, and 7 wins vs. teams behind them. What makes Toledo deserve the spot ahead of 11 teams on the schedule???