Couldn't it be that he plays better at home than away? Doesn't seem that crazy.
To me, it just sounds like more sour grapes from Purdue fans toward MSU.
*Official* 2017-2018 B1G In-Season Thread
- Thread starter Kevin Bryan
- Start date
To me, it just sounds like more sour grapes from Purdue fans toward MSU.
Of course it does. But is that evidence of my bias or your bias? Hmmm....
So they said they aren't using RPI any more, but they are basically just using it indirectly within this Quadrant BS that no one can figure out? What a bunch of hypocrites
Oh, absolutely it is crap. But after Sunday, it is readily apparent that the committee is using it. Hopefully next year they come up with a composite. Personally, I'd take the three predictive measures (BPI, KenPom and Sagarin), average them and then do the quadrant analysis off of them.
I don't think anyone said they aren't using it. They included better metrics on the nitty gritty sheets but RPI is still the primary driver. The shift to location based quadrants is a step in the right direction but until it's based on something other than RPI, it's still lipstick on a pig.
Oh, that's gay as hell
It is damning but Bilas is always looking for the most extreme example to support his position. Of course the committee is smart enough to not give equal weighting to those games. They have the list of actual results in each quadrant.
I used to like Bilas but now he is the king of hyperbole.
It is the most extreme example, but he’s not wrong, its a massive flaw. The committee is not smart enough because they are the dumbasses who created this way of measuring the teams.
Actually, the committee didn't create it. They are executing what the NCAA membership created. And they aren't dumbasses per se. They are typically made up of AD's and conference commissioners. I think they are pretty smart and while the basis for the report is wrong, I'm confident they are using more than just the quadrants when seeding.
Butler is absolutely going to get more credit for beating Villanova than a team would that won a game at Rider. To suggest otherwise is silly.
RollLaugh People bitch all the time that ADs and commissioners are dumb, christ, people still bitch about Delaney with the expansion. I understand why you like the quadrant system, but its bullshit because they use rpi to determine it. The selection process is like the catch rule in the NFL they keep adding to it hoping it gets better, but its getting worse.
Well the people who bitch are usually doing so without all the relevant information so I'm not sure that's who we should rely on for sanity.
And I'm not sure how you could possibly say that adjusting the groupings to reflect where the game is played is making it worse. That makes no sense.
Again, it's not good but it's definitely better and it is only one piece of what they look at.
Better than last year's where they grouped teams without adjusting for location....yeah, I'd say it's better. Now, it might be a D+ instead of a D- but it's definitely better.
To use Bilas' extreme examples:
Using last year's groupings, a home win over Louisiana would have been given the same credit as a road win at Villanova. This year, the first is in quadrant 2 and the second in quadrant 1.
Also, a home win over William and Mary would have been given the same credit as a road win at Nebraska. This year, the first is in quadrant 3 and the second in quadrant 2.
That is improvement, not what I'd like to see, but definitely improvement.
Yeah I still think some average of AP, Coaches, KenPom, BPI, Sagarin, and maybe RPI should be used to place teams 1 through 68 (or whatever), then standard rules regarding location, conference match ups, etc, be applied to a program to generate a bracket. The committee would then have x number of modifications they could make to help smooth out any kinks in the formula.
At the very least, this would add a level of transparency that the process needs.
At the very least, this would add a level of transparency that the process needs.
I don't know if I'd go that far. Do we really want a coach or biased media member to have a hand (however small) in the seeding process?
Hell, if I'm a coach looking at a 4 seed, I'm leaving everyone who might be competing with me for that protected seed completely off my ballot.
it's a fair point. not sure which polls you'd include, my point was that it would be beneficial to have some human polls in there with the data based metrics.
I'm good with just giving a committee tools that are not flawed (or at least a hell of a lot better than they currently have).
I must have because I thought you said that this year's sheet is worse than last years:
They are still using rpi to determine the quadrants. Just because they want to rephrase it to quadrant 1 or whatever doesn’t mean it still isn’t bad.
I'd also add that I'd tweak the committee. It's is currently administrators from member institutions. I'd go with a mix of those, some former coaches and a couple of national media members. As with the current committee, any member has to step outside the room when they have a conflict a la Mark Hollis and MSU in year's past.
Something I just don’t like at all in how the committee is emphasizing road wins and quality wins (that’s good) but ignoring everything else like all a teams bad losses or how a team is currently playing. (that’s really bad, IMO) A road win over Northwestern State is valued the same as a home win over Villanova. Both are Quadrant 1 wins. That makes no sense to me.
I was talking about you saying it got worse. Not sure what you were talking about.
interesting analysis done by Chris Forman on MSU's assists.
16 Home games: 368 assists / 500 FGs (23.0 APG; 73.6 assist rate)
11 "Away" games: 177 assists / 296 FGs (16.1 APG; 59.8 assist rate)
Maybe its not “worse” per se, but its just rearranging the deck chairs and calling it new and improved. At the root of it lays the problem-RPI.
So the Purdue SID is wasting time on this? Even if MSU is fudging assist numbers who the hell cares?
Engaging w/ fans on twitter is something he does. My guess is some guy on GBI saw my post and tweeted at him about it, so he engaged. He noted that he thought they gave us some generous assists as well.
If customer engagement is wasting time, you better start a seminar to preach to a lot of corporations.
It's customer engagement and those stats probably took him less than 5 minutes. He definitely has some API type feed to data or access to some advanced stuff that we don't.
Similar threads
