MVC tourney seeding probabilities

[Stryker]

One other way to measure the difficulty of the remaining schedule for the top 5 leaders is the average NET number for their remaining 6 opponents-
Team____Average NET
Drake____157.7
Bradley___177.3
Belmont__.190.7
SIU______198.2
INSU.........213.0

By NET, Drake has the toughest schedule remaining. Indiana State and SIU have the easiest. Bradley and Belmont are in the middle.
This does not take into account whether the games are home or away.

According to KenPom.com, Bradley, SIU, Drake, and Indiana State are predicted to win 5 of their last 6 games.
That would result in a 3-way tie for first place between Bradley, SIU, and Drake. If that happens, Bradley would get the #1 seed for the MVC Tournament, based on their better NET rank.

I know Belmont is in the middle, but I could definitely see them winning 5 more games, which I think would give them the title outright. At least that is where I would put my money right now. I think BU can win 5 as well, but Belmont would have the tiebreaker in that case.

[64NIT]

Updated seeding probabilities. https://twitter.com/mjhackman/status/1622658883931054081

I'm a little surprised we jumped from 2 to 1 in projections after all the top 4 teams won this weekend.

SIU and Belmont won at home. Drake won at Valpo in double OT and Bradley won at UNI which is a stronger opponent than Valpo. This might be why Bradley moved to #1. This is how I figured it and, obviously, could be wrong. There could be another reason why Bradley moved to #1.

[Da Coach]

I mentioned this in an earlier thread... The MVC Tournament seeding tiebreaker system that will be used is flawed this year, in my opinion.
2023 MVC Tournament Tiebreaker- https://mvc-sports.com/documents/202...eaker_2023.pdf

Note that the 1st tiebreaker is head-to-head competition. If more than 2 teams tied, the tiebreaker uses a round-robin type calculation of the record against all the teams that are tied.
The next tiebreaker involves a calculation of the win-loss record by the tied teams against the highest other team in the standings, and if that doesn't break the tie, continues with the calculation on down the standings until the tie is broken.
Finally, if all that doesn't break the tie, then it goes to the NET rank for each tied team.

The flaw with this is obvious. With the 20 game, unbalanced schedule, not all teams played home and home against each opponent. There will be ties between teams who only played each other once. When that happens, it does not seem fair to break the tie on a single game played at one team's home court. That would be no different than holding a postseason tie-breaker game on one team's home court. Nobody would consider that to be fair.

For example, SIU only played Belmont once, in Carbondale. SIU won that game at home, and will not have to play Belmont in Nashville. If they tie in the final standings (which they currently are), SIU would win the tiebreaker without ever having to play at Belmont. This flaw would also affect 3-way and other multiple ties. Teams that tie with other teams that they never had to play on the road, will have an advantage in winning the tiebreaker.
I'll predict that when this happens, either this season or in a future season, there will be a coach, or coaches who publicly point out this flaw, especially if it means they fall out of the top 4 and have to play an extra game on a Thursday, or get a much more difficult matchup in the tournament. Maybe it will eventually get changed.

[Stryker]

I mentioned this in an earlier thread... The MVC Tournament seeding tiebreaker system that will be used is flawed this year, in my opinion.
2023 MVC Tournament Tiebreaker- https://mvc-sports.com/documents/202...eaker_2023.pdf

Note that the 1st tiebreaker is head-to-head competition. If more than 2 teams tied, the tiebreaker uses a round-robin type calculation of the record against all the teams that are tied.
The next tiebreaker involves a calculation of the win-loss record by the tied teams against the highest other team in the standings, and if that doesn't break the tie, continues with the calculation on down the standings until the tie is broken.
Finally, if all that doesn't break the tie, then it goes to the NET rank for each tied team.

The flaw with this is obvious. With the 20 game, unbalanced schedule, not all teams played home and home against each opponent. There will be ties between teams who only played each other once. When that happens, it does not seem fair to break the tie on a single game played at one team's home court. That would be no different than holding a postseason tie-breaker game on one team's home court. Nobody would consider that to be fair.

For example, SIU only played Belmont once, in Carbondale. SIU won that game at home, and will not have to play Belmont in Nashville. If they tie in the final standings (which they currently are), SIU would win the tiebreaker without ever having to play at Belmont. This flaw would also affect 3-way and other multiple ties. Teams that tie with other teams that they never had to play on the road, will have an advantage in winning the tiebreaker.
I'll predict that when this happens, either this season or in a future season, there will be a coach, or coaches who publicly point out this flaw, especially if it means they fall out of the top 4 and have to play an extra game on a Thursday, or get a much more difficult matchup in the tournament. Maybe it will eventually get changed.

Coach, other large BB conferences have unbalanced schedules as well. Do their tiebreaker system match this or is it constructed differently? If it is different I wonder why the MVC would take this approach, otherwise if the tiebreaker system is the same as other conferences with unbalanced schedules, there may not be a better solution to the situation.

[Da Coach]

Coach, other large BB conferences have unbalanced schedules as well. Do their tiebreaker system match this or is it constructed differently? If it is different I wonder why the MVC would take this approach, otherwise if the tiebreaker system is the same as other conferences with unbalanced schedules, there may not be a better solution to the situation.

I don't know what other conferences do. But there are several other options that don't penalize teams unfairly for something that is not under their control.

Why not just drop the head to head or round robin tiebreaker methods if the teams involved in any ties have not played each other twice, and then default in those cases straight to the NET rank? If using the NET as a later tiebreaker is acceptable, why not use it in cases like this?

Recall that a few years ago, MVC Commissioner Doug Elgin modified the tiebreaker, and made it dependent on teams' non-conference RPI. It resulted in teams scheduling better in their non-conference season, and it resulted in improved overall MVC RPI. But, after a couple years he dropped it, and non-conference scheduling (and the MVC RPI) has gotten worse again.

Or reconfigure the MVC into two 6 or preferably 7-team divisions. Then schedule each team to play their division rivals twice, and each team from the other division once. Then seed each division separately and structure the MVC Tournament with two divisional playoffs, and the 2 division champions playing each other for the championship and the autobid.

Or, if the MVC stays at 12 teams, one other option is to go to a 22 game conference schedule, and every team will play each opponent twice.

[OhRotanes]

Updated project conference final standings. https://twitter.com/mjhackman/status...716807/photo/1

[OhRotanes]

According to this, it appears to be a 93% chance at a top 4 seed for Bradley

[Stryker]

According to this, it appears to be a 93% chance at a top 4 seed for Bradley

I'll take those odds and the 1st round bye. Would be so great to win the regular season though. Was surprised Belmont lost yesterday, and now it is looking like Drake may be the toughest opponent left, and biggest challenger to the MVC crown for BU.

[OhRotanes]

I'll take those odds and the 1st round bye. Would be so great to win the regular season though. Was surprised Belmont lost yesterday, and now it is looking like Drake may be the toughest opponent left, and biggest challenger to the MVC crown for BU.

I would not be surprised if the 1 seed came down to the Bradley vs Drake game on the last day of the season

[braves16]

We better be ready for a slugfest at SIU.

[Tommy]

I would not be surprised if the 1 seed came down to the Bradley vs Drake game on the last day of the season

I wouldn't count out SIU. Everyone seems to be discounting them all of a sudden, but they just continue to win and are still tied for first, but agree with the sentiment. Drake does appear to be the team to beat right now.

[Stryker]

I wouldn't count out SIU. Everyone seems to be discounting them all of a sudden, but they just continue to win and are still tied for first, but agree with the sentiment. Drake does appear to be the team to beat right now.

Probably because Drake is looking really really strong and on a 6 game win streak, BU has a 5 win streak going, and SIU just lost to BU & ISU, and UIC took them to the wire last night at home. SIU just doesn't feel like they have been playing as well as the other teams at the top of the conference. SIU also has to play at Drake and host BU & UNI yet.

Don't get me wrong, they definitely could end up winning out or something crazy, it just doesn't feel like they will at this point in time based on how they are playing.

[64NIT]

Bradley is number 1 in the most recent seeding probabilities!!

MVC Tournament Seeding Probabilities #ArchMadness - February 9 Update:
https://twitter.com/mjhackman/status...41740736716807

Here is the breakdown of potential seed odds by conference record. Regarding the first round bye: 12 conference wins probably doesn't get it done (6% chance), 13 wins is the target if you are feeling lucky (58% chance), and 14 wins nearly guarantees it (98% chance).
https://twitter.com/mjhackman/status...41743785975814

And here is an team-by-team breakdown of first round bye odds at each win level. Certain team odds are really starting to diverge from the average. Also note the total team bye odds at the bottom of the graph.
https://twitter.com/mjhackman/status...41746210283521

[Stryker]

Very interesting race between ISUBlue and Belmont for that 4th bye spot (By probability). Going to be a really interesting finish this year.

[64NIT]

MVC Tournament Seeding Probabilities #ArchMadness - February 13 Update:

https://twitter.com/mjhackman/status/1625162286083063811

Here is the breakdown of potential seed odds by conference record. The bye target continues to shift up as the top teams stay grouped tightly: 12 wins is highly unlikely to get it done (4% chance), 13 wins is a coinflip (46% chance), 14 wins is nearly a lock (95% chance).

https://twitter.com/mjhackman/status/1625162290235428864

And here is an team-by-team breakdown of first round bye odds at each win level. Note the large variance by team at the 13-win level. Also the total team bye odds are at the bottom of the graph.

https://twitter.com/mjhackman/status/1625162293313937408

Bradley:

https://twitter.com/mjhackman/status/1625163333262315521

As of today, looks like the top four, for tourney byes, are
1/2) Drake/Bradley (virtual tie)
3) SIU
4) InSU