2023-24 Rankings

ADORAZ said:
Personally analytics are my favorite aspect of sports, going back to when I read the stats on the back of baseball cards as a kid, but I assume you mean in regards to using analytics to rank NCAA teams.

Just curious, how would you seed the Tournament without analytics? Award the top half Power teams? That'd make the OOC useless, and would penalize conferences that do well OOC. Also how to handle mid major conferences vs. low majors?

I don't think pro sports should use analytics to determine which teams make the playoffs (fortunately they don't), but college is so unique in that you have 350+ teams at all different levels.
Click to expand...
I am a big stats guy myself but more for historical purposes. You are correct as my comment was in regards to rankings.

There is no exact science how to seed the tournament. I’ve always been a big proponent of the eye test, but of course that can be objective as well. Double digit seeds have historically wrecked brackets throughout the years, but seems the NET and Kenpom analysis would have slowed down that trend. Of late, the St. Peter’s, FDU’s and FAU’s of the world have proven the analytics have zero bearing on the court.
 
espken said:
I am a big stats guy myself but more for historical purposes. You are correct as my comment was in regards to rankings.

There is no exact science how to seed the tournament. I’ve always been a big proponent of the eye test, but of course that can be objective as well. Double digit seeds have historically wrecked brackets throughout the years, but seems the NET and Kenpom analysis would have slowed down that trend. Of late, the St. Peter’s, FDU’s and FAU’s of the world have proven the analytics have zero bearing on the court.
Click to expand...

Fair enough. NET isn't perfect but I don't think there's a better solution out there since the eye test couldn't realistically be used (and if it was you'd have a lottttttt of angry fans claiming corruption).

I wouldn't say the rankings have zero bearing on the court, though. Teams like Saint Peter's are still outliers. They're more common now than they used to be (more parity amongst teams), but a bracket that chooses teams randomly each round would finish last place in a bracket challenge. By a huge margin.

Ultimately if analytics meant nothing you could make a killing in Vegas using the eye test. Saint Peter's were Vegas underdogs their entire run.
 
ADORAZ said:
Fair enough. NET isn't perfect but I don't think there's a better solution out there since the eye test couldn't realistically be used (and if it was you'd have a lottttttt of angry fans claiming corruption).

I wouldn't say the rankings have zero bearing on the court, though. Teams like Saint Peter's are still outliers. They're more common now than they used to be (more parity amongst teams), but a bracket that chooses teams randomly each round would finish last place in a bracket challenge. By a huge margin.

Ultimately if analytics meant nothing you could make a killing in Vegas using the eye test. Saint Peter's were Vegas underdogs their entire run.
Click to expand...
Winning by 10 points instead of 9 doesn’t make a team better. That’s my problem with analytics. That and an assumption a team has a 26.7% chance of winning. Like you said, nothing is perfect.
 
espken said:
Winning by 10 points instead of 9 doesn’t make a team better. That’s my problem with analytics. That and an assumption a team has a 26.7% chance of winning. Like you said, nothing is perfect.
Click to expand...

Gotcha that makes sense, so did you prefer the old RPI system which didn't take margins or efficiency metrics into account (only whether the team W/L)? I prefer NET to RPI as I believe RPI was more flawed, but there's definitely a solid argument for RPI. We agree there isn't a system that could 100% accurately rank each team, but hopefully one day they'll come up with a better solution than NET (or at least fix some of the big issues).
 
Last edited:
ADORAZ said:
Gotcha that makes sense, so did you prefer the old RPI system which didn't take margins or efficiency metrics into account (only whether the team W/L)? I prefer NET to RPI as I believe RPI was more flawed, but there's definitely a solid argument for RPI. We agree there isn't a system that could 100% accurately rank each team, but hopefully one day they'll come up with a better solution than NET (or at least fix some of the big issues).
Click to expand...
If I had to choose, I would go with the old RPI.
 
dk423 said:
www.espn.com

Bracketology: Make way for Marquette, Iowa State on the top line

Another week, another shake-up at the top. Iowa State and Marquette join Auburn and Tennessee as No. 1 seeds.
www.espn.com

We're not even listed. Utah is in "last four out". Dayton is 12 as a AQ.
Click to expand...

Since we're on a 3 game winning streak and he had us in the field last time, this must be about NET. #75 is not good for NCAAT purposes but it's very easy to bring that up this early in the season.

I mentioned prior to the NET reveal that I think our resume for these first 7 games would look good on Selection Sunday but that our NET for now wouldn't be good. I realize as fans we're biased and try to convince ourselves of things, but KenPom isn't biased towards St. John's opponents and he thinks teams like WVU and Michigan should finish much, much better than where they're at. Also has North Texas a lot higher than the NET does which would give us another Q2 win.

If his projections hold up then I think we'll have a very solid resume for these first 7 games. Worst case if they don't then no big deal given the big majority of our resume still needs to be filled out and we're trending up.
 
ADORAZ said:
Since we're on a 3 game winning streak and he had us in the field last time, this must be about NET. #75 is not good for NCAAT purposes but it's very easy to bring that up this early in the season.

I mentioned prior to the NET reveal that I think our resume for these first 7 games would look good on Selection Sunday but that our NET for now wouldn't be good. I realize as fans we're biased and try to convince ourselves of things, but KenPom isn't biased towards St. John's opponents and he thinks teams like WVU and Michigan should finish much, much better than where they're at. Also has North Texas a lot higher than the NET does which would give us another Q2 win.

If his projections hold up then I think we'll have a very solid resume for these first 7 games. Worst case if they don't then no big deal given the big majority of our resume still needs to be filled out and we're trending up.
Click to expand...
Bracketology on December 5th should never be a big deal.
 
Relevant to St. John's Metrics Movement:
KenPom dropped from 60 to 62
NET dropped from 75 to 82 without playing a game

Mainly Because:

Michigan lost (NET dropped from 89 to 105)
North Texas lost (NET dropped from 104 to 108)
Utah won (NET dropped from 39 to 46)
Butler won yet (NET dropped from 38 to 61)
Xavier lost (NET dropped from 68 to 89)
Villanova lost (NET dropped from 50 to 55)
Seton Hall lost (NET rose from 81 to 77)
Boston College won (NET rose from 96 to 88)
 
Utah barely won a Quad 4 game last night, that would have hurt bad.
michigan lost at home vs Indiana that is much better than their NET.
North Texas lost away at Boise st, that also is better than its NET.
I think/hope both Michigan & north Texas NET improve during year.
 
SJUDave said:
Relevant to St. John's Metrics Movement:
KenPom dropped from 60 to 62
NET dropped from 75 to 82 without playing a game

Mainly Because:

Michigan lost (NET dropped from 89 to 105)
North Texas lost (NET dropped from 104 to 108)
Utah won (NET dropped from 39 to 46)
Butler won yet (NET dropped from 38 to 61)
Xavier lost (NET dropped from 68 to 89)
Villanova lost (NET dropped from 50 to 55)
Seton Hall lost (NET rose from 81 to 77)
Boston College won (NET rose from 96 to 88)
Click to expand...
Thanks for the recap!
At this point these ratings don't make a lot of sense to me but they are what they are.
That said all we can do is win and do it as well as possible. Hofstra is now 62 in the NET so for now that and the BC games are quad 2. Then the Butler and Xavier games at home are quad 2 also. I think even if we could stay within 10 against UCONN on the road would be a help.

I am really not sure how good we can be but we have chances ahead of us.
 
SJUDave said:
Relevant to St. John's Metrics Movement:
KenPom dropped from 60 to 62
NET dropped from 75 to 82 without playing a game

Mainly Because:

Michigan lost (NET dropped from 89 to 105)
North Texas lost (NET dropped from 104 to 108)
Utah won (NET dropped from 39 to 46)
Butler won yet (NET dropped from 38 to 61)
Xavier lost (NET dropped from 68 to 89)
Villanova lost (NET dropped from 50 to 55)
Seton Hall lost (NET rose from 81 to 77)
Boston College won (NET rose from 96 to 88)
Click to expand...
There should be no impact by the teams that we have not yet played for the day to day change. Those games/teams will be incorporated into the ranking as the games are played. It does hurt potential future value.
 
Yeah NET is 100% about the teams we've played already and doesn't make any projections for our future schedule. Having 3/7 of the teams drop one day can create that kind of movement, though. Expect a lot more daily movement until our resume fills out some.
 
Stevep502 said:
Utah barely won a Quad 4 game last night, that would have hurt bad.
michigan lost at home vs Indiana that is much better than their NET.
North Texas lost away at Boise st, that also is better than its NET.
I think/hope both Michigan & north Texas NET improve during year.
Click to expand...
Michigan is 4-5. Yikes.

I stand by my post that they beat anyone when they are on and lose to anyone when they aren’t.
 
Marillac2 said:
Michigan is 4-5. Yikes.

I stand by my post that they beat anyone when they are on and lose to anyone when they aren’t.
Click to expand...
There seem to be a fewteams like that in the Big East(Villanova and Xavier are two). I hope we aren't one of them also.
 
You must log in or register to reply here.
Back
Top