I think the worst part of the RPI is us saying @Rutgers is one of the biggest games left. They’re probably in winning their home games and beating Rutgers.
Now if they win the home games, lose @Rutgers, then win @Nebraska, they probably don’t make it. Same record, better road win but that loss kills it.
Don’t think the scenario you outline would really be that much different in the end.
Why would @Rutgers be better than @nebraska ?
Split and it’s the same. Hell, we should win both anyways.
Resume would show a bad loss, assuming Nebraska stays in the top 100. But I was thinking more resume and could be dead wrong on RPI.
@umhoops Really? Their RPI splits wouldn’t carry a big difference? Maybe I mis-spoke about the RPI. I definitely think it would be worse for their resume and getting in.
It’s such a crime that RPI determines the fate of teams.
Am I the only one who feels the bpi is much more accurate?
RPI is one of the most absurd oversimplifications of a sport to be valued by anyone who follows it. You’re most definitely not the only one who feels that way, although BPI does have its problems too.
KenPom does a great job of measuring which teams are better and who will win when facing each other.
On the other hand, RPI judges solely off of wins and losses. For example, Michigan’s loss to Virginia Tech doesn’t look so bad because it was close and the wins look a lot better because they were blowouts. That’s why Michigan’s KenPom rank is 20 spots higher than its RPI, but should that be how we determine who gets into the dance? Doesn’t really make sense to me.
You don’t think margin of victory should play a factor in whichever metric is being used to determine who gets in? I agree that kenpom itself would be a bad idea since it is solely “who is the better team” and not “who has the better resume” but I feel like there is a middle ground that could be formulated if they wanted to.
If Michigan gets to 19 wins ( any combination of regular season and B1G tournament ) they’ll be in the NCAAs.
Bubble is incredibly weak this year. Anything above 19-13 is guaranteed in.
I think the committee came out and said they they take into account a handful of computer models now (including Kenpom and maybe BPI?). That’s the way it should be. I think their biggest criteria is still RPI and SOS, but it’s nice that it’s not the be-all-end-all.
RPI and SOS is inherently flawed, too.
Team A:
Win to #1 overall team (20-0)
Win over #300 overall team (0-20)
Team B:
Win over #149 overall team (10-10)
Win over #150 overall team (10-10)
These two resumes would be waited equally in SOS and RPI, despite going 2-0 against Team A’s schedule being significantly more difficult.
I am not saying the RPI is perfect by any means… but I think there’s just some key flaws with some other metrics for the task at hand. Definitely would be possible to come up with something better, but other models aren’t designed to decide what team is most deserving to make the NCAA Tournament necessarily.
Yeah. It’s a tough one because most deserving is hard to determine when level of opponents are so different. Many people (namely mid-major honkers) were saying that Monmouth was “more deserving” than Michigan last year. Needless to say I argued against that.
I’m happy it’s a committee selection process and not just a single computer model.
U gotta think that a lot of the time the better team will have the better resume. Or you’d hope
You would… but there comes a point where a bunch of close losses to good teams doesn’t overcome a team who has good wins but bad losses. They could be equal on kenpom but no good wins shouldn’t get you in the tourney.
Yea I’m think getting to 21-11 would make us safe. Particularly if we do ok in the tourney. If they keep up this intensity on d and on the boards i believe we will sweep our away games, and make that ugly cant win on the road stat go away. 5 road wins will suffice
Not sure, you’d have to dig through old box scores in bulk and do the calculation. I would assume it has increased.
The shot clock was introduced in 1985 at 45 seconds, cut to 35 seconds in 1993, then 30 seconds in 2015. Obviously all of those adjustments would increase the pace.