cfb AP Poll as Accurate as a Coin Toss

[Gatorchip]

~62.5% accuracy guessing the top 20% is horrible. I UCForget that people don't understand Standard Deviation.

And that is particularly bad when the Polls basically ignore Wins and Losses and any objective measurement.

#UCFacts

Not only is that completely irrelevant, but you're showing 2 sigma.

New to statistics?

 

[UCFhonors]

Not only is that completely irrelevant, but you're showing 2 sigma.

New to statistics?

Just a generic illustration. But the 2nd std. deviation still shows a 62.5% accuracy on guessing the top 20% is well within probability with of coin toss results.

Polls are BS.

Everyone who knows math, understands this UCFact.

#UCFacts

 

[Gatorchip]

Just a generic illustration. But the 2nd std. deviation still shows a 62.5% accuracy on guessing the top 20% is well within probability with of coin toss results.

No it isn't. Blindly picking from 130 teams will land you averaging #65. If 62.5% of your picks end ranked in the top 25, that's what one would call and outlier.

Flipping coins won't get those results. Not even close.

 

[BamaDude]

During my Freshman, sophomore, Jr & Senior college years.

#UCFacts

He didn't ask when. He asked why.

 

[BamaDude]

Let me guess, your UCFhonors is not in a STEM field. English Lit? Gender Studies? Sociology?

Basket weaving.

 

[UCFhonors]

No it isn't. Blindly picking from 130 teams will land you averaging #65. If 62.5% of your picks end ranked in the top 25, that's what one would call and outlier.

Flipping coins won't get those results. Not even close.

Are you really....wait, did you think. Oh myyyy. You took 130 and divided it by 2, and you think that is probability theory.

It's little more complicated than that, Champ.

For starters, a team can move within that 25 nominal range and still be a positive selection. And that is just understanding the 62.5% accuracy stat. That's way before std. deviation and probability.

#UCFacts

 

[UCFhonors]

Some of you need to take a step back and a deep breath.

Then repeat after me:
UCFirst they deny.
Then they detract.
Then they come around to the UCFacts.

#UCFacts

 

[BamaDude]

Some of you need to take a step back and a deep breath.

Then repeat after me:
UCFirst I brag.
Then I deflect.
Then I make up some UCofFthe wall stuff.

#UCFacts

FIFY

 

[Bayou Tiger]

Y'all can bitch but you can't deny UCFacts!


Hahahahaha!

 

[Edisto_Tiger]

Dude, you need a new gig. This shit is old and tired.

 

[Gatorchip]

Are you really....wait, did you think. Oh myyyy. You took 130 and divided it by 2, and you think that is probability theory.

It's little more complicated than that, Champ.

For starters, a team can move within that 25 nominal range and still be a positive selection. And that is just understanding the 62.5% accuracy stat. That's way before std. deviation and probability.

#UCFacts

65 is average. Not probability. I was doing my best to simplify for you without busting out a calculator.

Try a combination calculation to get the 25 of 130 possibilities. Then improve those odds with your 62.5% that you mentioned.

And you say this is well within probability of a coin flip.... Come on dude.

I'll support your on your funny UCF stuff. Always have. I recommend sticking to that path.

 

[Edisto_Tiger]

At least entertain us like OD. That dude knew how to do this.

 

[Rolltide94]

Are you really....wait, did you think. Oh myyyy. You took 130 and divided it by 2, and you think that is probability theory.

It's little more complicated than that, Champ.

For starters, a team can move within that 25 nominal range and still be a positive selection. And that is just understanding the 62.5% accuracy stat. That's way before std. deviation and probability.

#UCFacts

How exactly do I coin flip my top 25. I mean, a coin only has two sides... Do I start with 130 teams and say maybe or definitely not...then take the 65 maybes and say probably or not...then what take the 33 or 32 probably and say top 16 or not. then take the losers and flip for the final 8 or 9 slots...so many questions, so little methodology, so much UCFoolishness. I'm starting to wonder about the quality of a UCF degree...

 

[naslundfan19]

Basket weaving.

Recess and lunch.

 

[Diego Roll Tide]

Yup it's a coin toss .... heads it's Clemson and tails it's Alabama

I wouldn’t go that far, but last season I took Alabama and Clemson against the field before the season. I would do the same this year.

Talent-wise only tOSU and UGA are comparable. Until UGA beats Bama, I have more confidence in Bama. tOSU has a killer roster, but somehow they manage to lose a game by a lot.

No team is a “sure thing” given injuries, but I feel reasonably safe that the CFP winner will be Clemson or Bama.

 

[7Samurai13]

If polls were 100% they wouldn’t need to play the games. Just recruit and the polls would just tell us who are the top teams.

 

[UCFhonors]

If polls were 100% they wouldn’t need to play the games. Just recruit and the polls would just tell us who are the top teams.

UCFinally someone sounds like a real UCFan of UCFootball.

#UCFacts

 

[7Samurai13]

Also every single ranking spot was significantly above a coin flip so the thread title is UCFake.

 

[TheReal_NU]

If polls were 100% they wouldn’t need to play the games. Just recruit and the polls would just tell us who are the top teams.

UCFinally someone sounds like a real UCFan of UCFootball.

#UCFacts

Honors and Samurai, sitting in a tree.
KISSI.........

Also every single ranking spot was significantly above a coin flip so the thread title is UCFake.

Never mind.

 

[7Samurai13]

Honors and Samurai, sitting in a tree.
KISSI.........



Never mind.

I was more making fun of him expecting them to be perfect with 0 data on hand. Preseason polls are very stupid though and most people here agree.