- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- 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
re: The Big 10 and ACC are setting the stage for locking up 3/4 playoff spots
Posted on 12/13/20 at 8:25 pm to JayWhite
Posted on 12/13/20 at 8:25 pm to JayWhite
quote:
If we're going to take in data to fill in the gaps, let's do that, but we can't ignore what's inconvenient to our desired conclusions.
Again, no skin in this game and I completely get this attitude. I’m an engineer by trade and this is something that comes up often with numerical simulations. You should be able to demonstrate predictive power with a numerical method for a given experiment without having to alter the methods between runs.
However, I’m trying to address what I perceive to be a glaring weakness in using our quasi-objective metrics with an improvement. You don’t seem won over by my idea of an improvement, and frankly it won’t matter past this season as I doubt we have to cross this bridge again, but I do think we need a fair and quasi-objective way to normalize between dramatically different sample sizes and I’m not sure any algorithm used in Massey really does that at all
Posted on 12/13/20 at 9:01 pm to Ross
quote:
However, I’m trying to address what I perceive to be a glaring weakness in using our quasi-objective metrics with an improvement. You don’t seem won over by my idea of an improvement, and frankly it won’t matter past this season as I doubt we have to cross this bridge again, but I do think we need a fair and quasi-objective way to normalize between dramatically different sample sizes and I’m not sure any algorithm used in Massey really does that at all
Maybe, and I apologize if this doesn't apply to you, but it seems that this is something new that is focused around the outcome of this season. I haven't been here long enough to know one way or the other, but I'd wonder if the same fervor against the Big Ten and Pac 12 would exist of the roles were reversed.
I'm not opposed to improvement. No good data scientist would be opposed to it. I'm opposed to alterations for the sake of different outcomes.
I don't know what Massey and Sagarin do to handle a disparity in number of games played, but I'd suspect it isn't nothing.
Posted on 12/14/20 at 8:26 am to Ross
quote:
However, I’m trying to address what I perceive to be a glaring weakness in using our quasi-objective metrics with an improvement. You don’t seem won over by my idea of an improvement, and frankly it won’t matter past this season as I doubt we have to cross this bridge again, but I do think we need a fair and quasi-objective way to normalize between dramatically different sample sizes and I’m not sure any algorithm used in Massey really does that at all
What's a bit amusing is that the SEC has still been playing an 8 game conference season in hte past. It's a clear advantage vs conferences that are now playing 9 conference opponents in a normal season. There are inequities between conferences in scheduling and it would be nice to see a baseline established to even out the playing field.
Popular
Back to top
Follow TigerDroppings for LSU Football News