Perhaps the BEST B1G Forum anywhere, here at College Football Fan Site, CFB51!!!

I suspect that at least one of those four teams will lose a game that they shouldn't.


I also suspect that at least one of the overlooked teams will be barking with the big dogs by season's end.

well, last season, PSU lost to Pitt, Michigan lost to Iowa, Ohio St. lost to PSU,

One way to rank teams is to use artificial lines from Vegas.  (I realize the actual lines have another influence in them.)


Who would beat who.  A "power poll" in effect, which might well have a 9-3 team favored over a 12-0 team.


You still have the issue with syllogism of course.


It's still great that these polls don't impact anything real, they are just for discussion.


Let's imagine FSU and Bama play a very even game decided in OT by a missed PAT.  How far should the loser drop?  If Bama lost, I might drop them to 3 or 4, at most, but they would really drop to 8 or lower, same with FSU.


You could even have a situation where the winner benefited from a 5-0 TO ratio and was handily outplayed by snuck out with a win.  You might drop the winner and keep the loser where it was.


But, that doesn't happen of course.  And it doesn't matter.

Individual games are decided on the field, but the winner is not always the better team.

the concept of pre-season polls is rejected by my over-active sense of fairness. 


this seasons seems more realistic than others, but it is an exception in that regard.  I get that Bama is Bama, and they are always an embarrassment of wealth, but #1?  I find that suspect.  Same with Clemson rounding the top 5.  USCw is a team I don't know enough about either this season or last.  I DO know they've amassed a nice roster of stars, so... gotta give them a 'pass-and-see'.


there always (this season not withstanding) seems to be a wild card at the top.  This season, I'm targeting FSU as that poser, but i don't feel as strongly about that as i have at times in the past while reviewing the pre-season polls.  If i were to wager, i'd wager the season opening showcase game of FSU and Bama is cause for FSU's ranking at that placement as much as any other reason.  that's because, in addition to having an over-active sense of fairness, i also harbor an over-active mechanism of cynicism.   


but... anyway..


the entire premise of polling and ranking a team requires what? 


this may be fodder for another thread....


I suggest that 'consistency' is a requirement for a team to be 'top five' as opposed to 'just' top ten/fifteen... I mean, we pretend these teams perform in a linear world, when they don't.  the team you see one day may be comprised of the same players playing the same play call tendencies, but appearing to play like a different team... they lack consistency.  We'll see possibly 'great ofensive' teams demonstrate flashes of astounding 'perfection in execution', and nod toward their greatness- but that's based on the resistance they encounter in their opponents defenses- which isn't linear in production either... so the flashing team breaks off a series of 20 yard hard fought yardage plays, and this impacts the pollsters interpretation- but... the defense they faced could have possibly rotated at the wrong time placing the wrong sets on the field, or the captain of the D was taking a breather and couldn't check back in, or the 'flashing O' found a component where they mismatched that puzzled the DC for more than that series lasted, and they capitalized... in such a case/condition, impressions are made and cemented in observers minds when it just isn't exactly what it appeared to be without consistency- and that consistency is rare in the dynamic environment of 'adjustments'.


All that said, this pre-season poll seems close to me with the exception of USCw and possibly FSU.  I think FSU is a top ten to be sure, but I just don't know about top five.  I don't know enough about USCw, but the impression that remains from last season (which i can't explain further than that) is they are a top15, but not a top 5 team.  Other than that- it seems legit.

Why would you dismiss a 5-0 turnover ratio?

Sure, the losing team would probably win if you take away all of the winning teams big plays. That's Football.

If you take away Georgia Tech's 222 points against Cumberland, then the biggest blowout in the history of CFB would have been a scoreless tie. So what?

The short answer is that it's a relative rarity. Basically, it took something unusual for one team to win a very close game. You wouldn't 100 percent dismiss it, but you would take it as part of a larger context.

The longer answer is that turnovers are quite hard to control. Fumble recoveries are mostly crap shoots. Interceptions are mildly more in a defense's control, but a lot are still tipped balls that rely on the magic of physics to be catchable. Returns off turnovers are likewise mildly random and of course can have big impacts. So a defense will try to do the things that lead to turnovers, go after the ball, hit QBs, cover well, but some days that'll lead to more takeaways than others. If you needed all five to hold on, or needed great placement for some, your fans will likely leave more relieved than confident.

Basically, football games are often weird scrambles of so many things. Taking only the final outcome of win/loss is fine, but it often leaves behind indicators of quality. Those indicators won't predict everything, as the game is beautifully random, but they can often shed light onto what's going on.

Turn overs are often random events.   A team that loses because they had 5 TOs to none could well have been the better team overall but the ball bounced against them.  The other team could have had 5 fumbles also that they managed to recover.


Other near random events are kicks for points that bounce off the goal post.  Yeah, a better kicker doesn't miss like that, but it can bounce through or to the side or back, a matter of inches.  Then there is where the ball is spotted on a key play where an inch makes a difference.  The spot is random within that inch.  Then there are holding calls made or not made.


In some games, all these things combine against you and the other team is just good enough to take advantage and beat you.  That is when major upsets happen.  The outcome doesn't necessarily mean the winning team is better, or even better that day.  That team with 5 fumbles lost may have very few fumbles in the rest of their games, but on one day, the "odds" fell against them.


Most major upsets, by which I mean where the losing team was favored by say 14 points or more, are due to turnovers, key penalties, an opposing QB who happens to get hot or has tipped passes that fall into the right hands, etc.


The losing team may go up and down the field and end up with 3 points due to some penalties or a missed FG or TO.  I've seen teams fumble at the goal line and the other team picks it up and runs 99 yards for a 14 point swing.  It's more random than anything.


The better team does not win every time out.  Most of the time, but not every time.

eh, football is a game of inches.

If one team doesn't take care of the ball while the other team does, then the outcome is predictable.

I don't generally punish winning teams for making more big plays than the other team.

Now if the refs butcher a bunch of calls, you might have a more sound argument. But that's less likely now that we have replay.