Sagarin ELO CHESS 9/22/2013

[charlie42s]

It doesn't matter if it predicted Georgia to beat Clemson. It will correct after either way. And anything above 70% is pretty good and as good as most humans. It's never going to be 100% accurate. Nevermind teams that should win and don't win, etc.

Ignoring the question. Does Sagarin post the results of his predictive system after game results?

I'll bet no.

 

[4down20]

Ignoring the question. Does Sagarin post the results of his predictive system after game results?

I'll bet no.

I dunno, but it doesn't really matter.

I post mine, it's not a big deal.

 

[4down20]

And btw, I had Georgia as 3 point favorites against Clemson. Which wasn't a bad pick and was only off 6 points total.

But I also has BYU and Ole Miss beating Texas.

Can't pick them all.

 

[TigerBait1971]

What's your computer say about the LSU UGA game?

 

[4down20]

What's your computer say about the LSU UGA game?

LSU Tigers football schedule and information

Georgia as 1 point favorites(33-32), but I wouldn't read into it much. The poster who said it was too early in the year for computer rankings to mean much pretty much nailed it.

Plus, it always hates LSU at the start of the year because you guys win ugly so much. Kind of drives me crazy.

 

[charlie42s]

I dunno, but it doesn't really matter.

You don't understand feedback and the positive influence it has when it is part of systems. I not surprised.

 

[4down20]

You don't understand feedback and the positive influence it has when it is part of systems. I not surprised.

 

[occupant]

-

its week 3 computer polls are garbage right now. Just not enough data to give accurate information

I know...but it's funny to see how many people care.

Like when I bitch about pre-season polls and early season polling and people tell me to simmer down..."OH IT'S EARLY YET...IT WILL PAN ITSELF OUT" they say.

I'm sorry, but wrong is wrong no matter when it happens.

Wouldn't you agree? I've never been sold on OU being ranked high to begin the season. Anything could happen...injuries, discipline issues, etc., and it has...trust me.

It would be very interesting to see what would happen if everyone started with a clean slate, and I'm attempting to do just that. Hope my shit will be ready by week 7.

 

[4down20]

I know...but it's funny to see how many people care.

Like when I bitch about pre-season polls and early season polling and people tell me to simmer down..."OH IT'S EARLY YET...IT WILL PAN ITSELF OUT" they say.

I'm sorry, but wrong is wrong no matter when it happens.

Wouldn't you agree? I've never been sold on OU being ranked high to begin the season. Anything could happen...injuries, discipline issues, etc., and it has...trust me.

It would be very interesting to see what would happen if everyone started with a clean slate, and I'm attempting to do just that. Hope my shit will be ready by week 7.

Most computer polls recalculate every week as if everyone is on a clean slate. Mine do. Previous weeks rank means nothing.

And the only one that really matters is the one before the bowl games, and it's #1 and #2 that really only matter then. The rest is entertainment.

 

[occupant]

Most computer polls recalculate every week as if everyone is on a clean slate. Mine do. Previous weeks rank means nothing.

So how then does Sagarin give "TOP 30" wins/losses at week 1? He's basing that on something.

And the only one that really matters is the one before the bowl games, and it's #1 and #2 that really only matter then. The rest is entertainment.

That's where I have a problem. Pre-loaded rankings that aren't based on anything but opinion, before a SET of performances can be inspected for the basis of an opinion, impact poll positions..EVEN 1 and 2 at the end of the season are liable to be skewed by rankings given.

It's like knowing that by measuring something, you are actually affecting the outcome JUST BY THE ACT OF MEASURING it.

And again it's easy to poopoo critique of the system when you benefit. My team has benefited from it plenty, still doesn't make it right.

 

[WhiteMamba]

when does first BCS poll come out?

After week 6 ya?

 

[occupant]

Yessir.

That is the first week the second human component (Harris) comes out.

 

[4down20]

So how then does Sagarin give "TOP 30" wins/losses at week 1? He's basing that on something.

Don't know about Sagarin personally, but I use the data from the previous year for preseason on current years schedule.

That's where I have a problem. Pre-loaded rankings that aren't based on anything but opinion, before a SET of performances can be inspected for the basis of an opinion, impact poll positions..EVEN 1 and 2 at the end of the season are liable to be skewed by rankings given.

It's like knowing that by measuring something, you are actually affecting the outcome JUST BY THE ACT OF MEASURING it.

And again it's easy to poopoo critique of the system when you benefit. My team has benefited from it plenty, still doesn't make it right.

I hear this argument about preseason rankings all the time, and yet nobody can really give an example of it actually happening. The only ones I hear that are close are Auburn and Oklahoma St. Yet Auburn just wasn't seen as in the best 2, and Oklahoma St being 3rd had nothing to do with preseason rankings since they rose and then dropped after losing.

Sorry, it's myth.

The same stuff would happen even without rankings. And it's not about my team benefiting, the entire reason I do computer rankings is to remove any and all bias. Yet I always come up with results towards the end of the year that are similar to that of the human polls.

 

[WhiteMamba]

Yessir.

That is the first week the second human component (Harris) comes out.

Ya, thought so. I knew the Harris doesn't release til later and sure it was week 6 therefore us getting the BCS polls.

Only two more weeks and we will be spilling over the BCS.. yay

 

[occupant]

Don't know about Sagarin personally, but I use the data from the previous year for preseason on current years schedule.

And that would be part of the problem.

I hear this argument about preseason rankings all the time, and yet nobody can really give an example of it actually happening. The only ones I hear that are close are Auburn and Oklahoma St. Yet Auburn just wasn't seen as in the best 2, and Oklahoma St being 3rd had nothing to do with preseason rankings since they rose and then dropped after losing.

It doesn't just affect the 1 vs. 2 matchup. *See Northern Illinois BCS Game 2012.

Sorry, it's myth.

Your acknowledgement of the two best examples kinda refutes your statement.

The same stuff would happen even without rankings.

Err...that doesn't make any sense.

And it's not about my team benefiting, the entire reason I do computer rankings is to remove any and all bias. Yet I always come up with results towards the end of the year that are similar to that of the human polls.

Similar. Care to provide some examples of the similarities?

 

[4down20]

And that would be part of the problem.

No it's not, once data is available for the current year it's not used anymore.

It doesn't just affect the 1 vs. 2 matchup. *See Northern Illinois BCS Game 2012.

Ok? Sorry, I don't really give a crap.

Your acknowledgement of the two best examples kinda refutes your statement.

I debunked them both. Especially Oklahoma St. You are just making the assumption that because a team was 3rd, previous rankings were at fault. Which is just bullshit. If people thought they deserved it, they would have went. They didn't. That has nothing at all to do with the went rankings start or anything else.

Auburn was a similar story. If they made it to 3rd, then it's a matter that voters didn't think they were a top2 team and it had nothing to do with previous rankings.

And at any rate, it won't even be an issue since playoffs start next year.

Err...that doesn't make any sense.

People are still going to think some teams are better than others even without rankings.

Similar. Care to provide some examples of the similarities?

I post all the results on my website for the past 5 years.

 

[michaeljordan_fan]

So how then does Sagarin give "TOP 30" wins/losses at week 1? He's basing that on something.

.

Its based off of the current top30 in his rankings, which recalculates each week.

As for the Clemson v UGA argument, its unlikely he had computer rankings, since he had no data to input.

 

[occupant]

No it's not, once data is available for the current year it's not used anymore.

But you've used junk data (2012) to arrive at a solution. Error.

Ok? Sorry, I don't really give a crap.

Thanks for reaffirming one of my statements. It didn't affect your team so it doesn't matter.

I debunked them both. Especially Oklahoma St. You are just making the assumption that because a team was 3rd, previous rankings were at fault. Which is just bullshit. If people thought they deserved it, they would have went. They didn't. That has nothing at all to do with the went rankings start or anything else.

You debunked nothing. If either team had been:
1. Weighted correctly at the beginning of the season (impossible without current relevant data)
2. Enjoyed the benefit of the nonsensical weighting that other teams benefited from at week 6.
The resultant BCS formulae would have been different.

Auburn was a similar story. If they made it to 3rd, then it's a matter that voters didn't think they were a top2 team and it had nothing to do with previous rankings.

See above.

And at any rate, it won't even be an issue since playoffs start next year.

LOL. So, since it's scrapped next year it doesn't matter. Classic.

People are still going to think some teams are better than others even without rankings.

I post all the results on my website for the past 5 years.

Knowing how you calc...I don't think your website is for me. Thanks anyway.

 

[4down20]

But you've used junk data (2012) to arrive at a solution. Error.

Until the team has 5 FBS games played on teh current year, then it no longer uses the data. And since the results are recalculated each week, the previous weeks rankings has absolutely NO effect on the current weeks.

Not sure what is so hard for you to grasp. But as usual, you are just being your asshole self.

Thanks for reaffirming one of my statements. It didn't affect your team so it doesn't matter.

Yeah, it's because it doesn't affect my team and not because it's the last spot for a non AQ team.

You debunked nothing. If either team had been:
1. Weighted correctly at the beginning of the season (impossible without current relevant data)
2. Enjoyed the benefit of the nonsensical weighting that other teams benefited from at week 6.
The resultant BCS formulae would have been different.

Oklahoma St went to #2 before they lost. Try again. The #1 team was LSU, and you are plain stupid if you don't think that ranking was deserved. It has nothing to do with preseason rankings, you just refuse to accept it.

LOL. So, since it's scrapped next year it doesn't matter. Classic.

Unless you just want to continue being a butthurt asshole, it doesn't.

Knowing how you calc...I don't think your website is for me. Thanks anyway.

You live in a fantasy land, nothing is for you.

 

[occupant]

...you are just being your asshole self...unless you just want to continue being a butthurt asshole...

Ahhhh....the slither into the refuge of the vanquished.

I love that sound!

Thanks for playing!