- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Coaching Changes
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
0
Posted on 7/9/19 at 1:55 am to Salviati
Why would you average the % of all games and apply it to the entire schedule? That's just stupid. ANY % above 50% is a predicted LSU win. Wtf don't some of you understand?
Posted on 7/9/19 at 6:51 am to ImayGoLesMiles
quote:
Why would you average the % of all games and apply it to the entire schedule? That's just stupid. ANY % above 50% is a predicted LSU win. Wtf don't some of you understand?
He didn't. You just suck at statistics and so are hopelessly lost on what he did. Trust me, he did it right..... Or don't trust me. Either way, he did it right, and you embarrassed yourself.
Posted on 7/9/19 at 7:13 am to TheMuffinMan
It happens the the outcome is a binary choice.
Posted on 7/9/19 at 8:20 am to LSUTigersVCURams
quote:
Why do we only have a 65% chance of beating MSU? Are they supposed to be good?
No idea. I know they lost 4 starters on defense to the draft, three of which were first round picks. I don’t see them having anywhere near as good of a defense this year just on that alone. (#1 in Total D iirc) I know nothing of their offense.
This post was edited on 7/9/19 at 10:26 am
Posted on 7/9/19 at 8:35 am to ImayGoLesMiles
quote:
Why would you average the % of all games and apply it to the entire schedule? That's just stupid. ANY % above 50% is a predicted LSU win. Wtf don't some of you understand?
So if FPI gave LSU a 51% chance to win every single game (favored in every game), are you arguing that the most likely outcome in that scenario would be a 12-0 regular season? I mean, there are 12 predicted wins there, right??
Posted on 7/9/19 at 8:37 am to ImayGoLesMiles
quote:
Why would you average the % of all games and apply it to the entire schedule? That's just stupid. ANY % above 50% is a predicted LSU win. Wtf don't some of you understand?
Because if you play a 4 game stretch multiple times that you have a 75% chance of winning each game, statistically speaking you will go 3-1 most of the time. Sure, there will be the occasional 4-0 and 2-2 and even rarely the 1-3 stretch. You might even go 0-4 but that would be extremely rare.
The point is, it’s about the law of averages. Just because you are “favored” does not mean it’s an automatic win.
That being said, I still say anything under 10-2 this season is a failure.
Posted on 7/9/19 at 8:48 am to Salviati
Wrong answer will be better than 03 team
Posted on 7/9/19 at 8:48 am to RogerTheShrubber
quote:
What was your opinion of them?
The same as this one...preseason polls are meaningless. Whether they have LSU ranked #1 or #25.
The point is that it's funny that depending upon where we rank seems to determine of the poll is touted as worth giving credence to versus ignoring.
Posted on 7/9/19 at 8:48 am to Salviati
I’m pretty pumped that we have a 25% chance of beating Bama!
Posted on 7/9/19 at 9:30 am to Salviati
quote:
Wrong. It's called probability, and it's math. You could learn it. Maybe.
I have a graduate degree in a STEM discipline and 10 years of practice. I could go on about my qualifications to debate on this topic, but you'll just deflect and say I'm bragging, so I'll just say I understand "probability and math." 9.3 and 2.7 were completely pulled out of thin air. It is impossible to calculate those numbers because the FPI doesn't state 1) how many simulations were run, 2) how many of those simulations resulted in a SECCG, 3) what our record in SECCGs was. If we had that data, we could calculate the exact regular season w/l record. Since we don't, it's complete speculation.
It seems you're just a giant negatiger who enjoys slinging mud at the sunshine pumpers.
Posted on 7/9/19 at 9:44 am to Diesel88
Posted on 7/9/19 at 9:56 am to GeauxTigerTM
quote:
The point is that it's funny that depending upon where we rank seems to determine of the poll is touted as worth giving credence to versus ignoring.
I don't see this at all. The only thing preseason polls are good for is message board fodder. It seems universally accepted
Posted on 7/9/19 at 12:51 pm to Diesel88
quote:
It is impossible to calculate those numbers because the FPI doesn't state 1) how many simulations were run, 2) how many of those simulations resulted in a SECCG, 3) what our record in SECCGs was. If we had that data, we could calculate the exact regular season w/l record. Since we don't, it's complete speculation.
It's impossible to calculate the mean regular season win total given all of the single-game win probabilities?
Here, I did it:
And given your previous statement on the matter:
quote:
Point is, this formula is predicting somewhere inbetween 9 and 10 wins for the season.
What are you arguing about?
Posted on 7/9/19 at 1:02 pm to TheCaterpillar
quote:
9-3 should lead to heads rolling.
But with a bowl win that would land O with 10 wins and that seems to be the comfort win total for most everyone here.
Posted on 7/9/19 at 1:33 pm to YouAre8Up
quote:
But with a bowl win that would land O with 10 wins and that seems to be the comfort win total for most everyone here.
Pretty sure most are saying 10 regular season wins is the minimum expectation. That’s what the consensus expectation was here under Les, and except for a few obvious “newcomers,” that’s pretty much still the minimum expectation.
As another poster mentioned, there aren’t 3 acceptable losses on this year’s schedule.
This post was edited on 7/9/19 at 1:35 pm
Posted on 7/9/19 at 1:46 pm to Salviati
Yep it ranks LSU 4th in the country. Yo guys can make anything negative
Posted on 7/9/19 at 3:29 pm to lostinbr
quote:
It's impossible to calculate the mean regular season win total given all of the single-game win probabilities? Here, I did it:
You did what? Calculated a number higher than 9.3 based on a randomly selected value of 5,000 simulations?
quote:
What are you arguing about?
What are YOU arguing about? I'm trying to point out that the OP truncated 9.5 to make a false claim that "statistics" predict that LSU will have a 9 win season. Let me help you - in any of the simulations in which LSU made it to the SECCG, they would have lost to Georgia since Georgia has a higher FPI rank. This inflated the 2.8 and deflated the 9.5. In reality, the statistics show that we are closer to a 10 win team.
Posted on 7/9/19 at 3:33 pm to Diesel88
quote:
Diesel88
You may have a degree, but you don't know shite about college football... Stick to chemistry.
Posted on 7/9/19 at 5:43 pm to Diesel88
quote:
What are YOU arguing about? I'm trying to point out that the OP truncated 9.5 to make a false claim that "statistics" predict that LSU will have a 9 win season. Let me help you - in any of the simulations in which LSU made it to the SECCG, they would have lost to Georgia since Georgia has a higher FPI rank. This inflated the 2.8 and deflated the 9.5. In reality, the statistics show that we are closer to a 10 win team.
Let me help YOU:
1. LSU would not have lost in all simulations vs. Georgia. They would have won some of these simulations and lost some of these simulations, which is the entire point of the statistical argument.
2. Simulations in which LSU made it to the SECCG may have increased the loss total but cannot “deflate the 9.5” as you suggested, because the SECCG is an extra game on the schedule. They don’t take away a regular season win because you lose the conference championship.
3. If anything, the simulations in which LSU reaches the SECCG inflated the win total, relative to the regular season, because LSU would have won in some of these simulations.
3. The SECCG doesn’t matter anyway because OP was talking about regular season record (as has been stated repeatedly in this thread) which doesn’t include the SECCG.
4. If you actually do the math with the single game probabilities listed in the OP you will find that the average regular season win total is less than 9.5, and that LSU has a ~52% likelihood to win 9 or less games.
5. The distribution IS truncated, because you can’t win more than 12 regular season games...
You can argue all kinds of stuff - that the FPI probabilities are BS, that the games are not discreet events, etc. - but you have picked an odd hill to die on over 0.2 wins.
Page 5 of 5
Popular
Back to top
Follow TigerDroppings for LSU Football News