Is Jon L and the rest of the HL brain trust working on the problem, or would that be stepping on toes?
Lecrone is too busy shopping for a legacy - which is nuts because he already checked out with a perfectly good legacy 15 years ago. He's been trying to return it to the store ever since.
I mean, there's nothing Lecrone and the conference can do during the season. The problem is that most of what he tries to do in the
offseason is counter-productive.
I like the in-season challenge with the Summit League. I've liked
almost every added team (barring YSU, IUPUI, but after IUPUI it sadly made sense to add Fort Wayne). I even kinda dig the attempt to create a double-elimination in the Horizon League Tournament to help make sure the best team gets to the dance.
This offseason, I'd like the conference to start working toward setting some standards for men's basketball. Plenty of HL fans like Commissioner at Detroit and Big D (WSU) have outlined ideas on how to elevate the conference.
If the conference adds teams to get to 12 members, I'd like to see a conference scheduling guideline play out to limit the amount of "bad" games a good team plays in the regular season. The way I would do this involves an 16-game conference schedule that basically takes the six "best" teams and puts them in a pod away from the six "worst" teams. You play a round robin within your own pod, then play a single game against each of the teams in the other pod.
I don't know how you'd predict who the "best" and "worst" teams would be. You can't just base it on last year's order of finish, because then you're screwed in a scenario in which some team finishes 5th but they lose 7 of their top 8 scorers.
This is where the Horizon League can benefit from some predictive metrics. They should have a set schedule every year (say, #1 plays #2 on the road first and at home last weekend of the season) but the teams in them should change. The conference's effort should be put into trying to predict the order of finish.
It wouldn't be perfect. Catastrophic injuries could turn a top 4 team into a bottom 4 team quick. But if you pair the predictive metrics with a 16-game schedule isolating the "bad" teams from the "good" teams, and keep something like the double-bye in the conference tournament to protect the best finishers, you can limit the amount of "bad" games for the top 2-3 teams and give them a better chance to have a better NET rating.
The other thing I'd like to see the conference do is enact a guideline for teams to schedule the remaining games. More challenges (I want to see the MAC, don't expect the MVC to agree but them too) for the conference to remove the need for teams to schedule more games. If the conference requires teams to finish with a top-200 SOS at least twice every three years, you limit the coaches who just take every low-major game they can get to pad their record and drop the conference's NET rating. Like the predictive metric, I don't know how you would do that for the upcoming season. But the people who are setting this are supposed to be professionals in running a basketball conference, so they
should be able to figure it out.
If this means the teams schedule more non-D-I games, fine. Those games don't affect the NET. If this means teams schedule more road games in the non-con, fine. Those games are more positively weighted for the road team, leading to a better NET. Getting blown out on the road against a top-100 NET team doesn't negatively affect a team unless they're
also a top-100 NET team.
Follow these guidelines and you could see the predictive #1 team in the Horizon League jump far higher in the NET. The bullet points:
- Playing fewer non-conference games against teams that drop NET rating regardless of result
- Challenge with other conferences adds "good" games b/c it matches predicted best teams
- Limiting conference games against teams that drop NET rating regardless of result (~6/16 by design)
- Higher NET rating from conference opponents due to scheduling requirements
- Stacking the deck in the conference tournament which limits possibility of another "bad" game right before Selection Sunday
For the 2017-18 season, Wright State finished with the best RPI in the conference at 99 (ugh). Their OOC SOS was 49, so they had a good non-con schedule that would have qualified for any scheduling guideline. Their full SOS, however, was 238.
With the guidelines: potentially eliminating bad non-con games like a couple of the ones in their bad in-season tournament (opponents 244, 286, 180). Eliminating half of their bad games (dropping 5-6 games against teams sub-250, in 2017-18 it was 5 games against for sure sub-230 teams).
What kind of RPI would WSU have had in 2017-18 if, say, the Summit Challenge had replaced one of the bad in-season tourney games, they kept the 180 and the other one was replaced with a MAC Challenge game? Funnily enough, the hypothetical MAC Challenge game wouldn't have helped much - WSU was predicted fifth, and the team predicted fifth in the MAC was Ohio. Ohio ended up 204 in the RPI. But that would have
still been better than the 244 or 286 games they had in the in-season tournament.
Wright State could have been top 40-50 in the RPI if scheduling was done right.