Bracketology

Status
Not open for further replies.
#176      
FeelYourPaign
On your final question, absolutely imo. Talent is widely distributed and great upper classmen don't stay, so volatility abounds.

On the conference questions, I put my faith in either Ken Pom or Sagarin. Here are best team / average team / worst team for
Big 12--- 4 / 31 / 78
Big 10 --- 11 / 54 / 135
ACC --- 2 / 47 / 174

Seems to me that the big 12 would be the hardest conference to go undefeated in. Not that anyone anywhere is likely to do it, just that it is harder in the Big 12
 
#177      
whovous
I am surprised at how much difference there is between KenPom and the top 25 rankings. The latter metric makes the ACC look far better than the Big 12.

I suspect that all of the volatility may carryover to tournament selections this year, though I confess that is entirely a gut reaction on my part that I rely upon so that I can continue to believe the Illini will make the dance.
 
#178      
WesterveltVictoryCigar
whovous said:
I am surprised at how much difference there is between KenPom and the top 25 rankings. The latter metric makes the ACC look far better than the Big 12.
Click to expand...

One criticism of KenPom is that people feel he doesn't put enough weight on winning, i.e. his rankings are about efficiency margins, so a 1 point loss isn't that different than a 1 point win. But obviously to voters in polls, there's a big difference.
 
#179      
jmilt7
On the one hand the Illini have moved down in the "Bracketology" (is that really a word?) per:

NCAA Tournament Bracketology: Illinois downgraded to Joe Lunardi’s ‘First Four Out’
<http://www.thechampaignroom.com/2017/1/4/14163356/ncaa-tournament-bracketology-illinois-fighting-illini-basketball>

I don't know who Joe Lundari is or whether he should be taken seriously. Just throwing it out there.

On the other hand:

Illinois Basketball: Illini Shooting Up the RPI Rankings
<http://writingillini.com/2017/01/03/illinois-basketball-illini-shooting-rpi-rankings/>

I know a lot of people here don't respect the RPI but I am just mentioning it because both these articles originally came from ESPN. And, again, just throwing it out there.
 
Last edited by a moderator:
#180      
FiveStar
whovous said:
Big 12 has three in the top 25, and two other teams receiving votes in either the AP or Coaches poll. B1G has three in the top 25 and five others receiving votes. The Big 12 is clearly stronger at the top, but overall?
Click to expand...

I kinda wanna expand on this. I think we all know who the top, top teams are in the power 6 (tourney "locks"). My question is, which conference has a better middle tier? Are the mid-tier B1G teams better than the big 12, sec, etc? ACC mid-tier the best? Where the B1G falls in that mid-tier pecking order will be UGE for us.
 
#181      
WesterveltVictoryCigar
FiveStar said:
I kinda wanna expand on this. I think we all know who the top, top teams are in the power 6 (tourney "locks"). My question is, which conference has a better middle tier? Are the mid-tier B1G teams better than the big 12, sec, etc? ACC mid-tier the best? Where the B1G falls in that mid-tier pecking order will be UGE for us.
Click to expand...

When you're looking at the at large teams, conference affiliation is irrelevant. All those teams are pitted against each other and their entire body of work is evaluated. If it comes down to it, we do have some wins over a few potential bubble teams - VCU, NCSU, BYU, tOSU.
 
#182      
ChiefIllini
FiveStar said:
I kinda wanna expand on this. I think we all know who the top, top teams are in the power 6 (tourney "locks"). My question is, which conference has a better middle tier? Are the mid-tier B1G teams better than the big 12, sec, etc? ACC mid-tier the best? Where the B1G falls in that mid-tier pecking order will be UGE for us.
Click to expand...

Hard to argue ACC wins by a large margin. They have 7 top 25 teams. We barely have 3. I think B1G is probably 3rd, just because the big 12 is smaller.

In no particular order:
B1G: MD, OSU, Mich St., Michigan St., Minny
big 12: ISU, Tech, Ok St., K St., TCU
 
#183      
Calvin
WesterveltVictoryCigar said:
One criticism of KenPom is that people feel he doesn't put enough weight on winning, i.e. his rankings are about efficiency margins, so a 1 point loss isn't that different than a 1 point win. But obviously to voters in polls, there's a big difference.
Click to expand...

Same for Sagarin, although he used to calculate a win/loss version, which he averaged with the points version. Now he uses a blend of point based models.

There's so much emphasis on predictions and betting during the season, that when you switch to actual performance in terms of wins/losses, it takes a shift in focus. That's why I've personally liked looking at KenPom and Sagarin for predictions, but the bracketmatrix for analyzing tourney invites. I think the point based models are also useful for looking at SOS.
 
#184      
Whitman2020
Calvin said:
Same for Sagarin, although he used to calculate a win/loss version, which he averaged with the points version. Now he uses a blend of point based models.

There's so much emphasis on predictions and betting during the season, that when you switch to actual performance in terms of wins/losses, it takes a shift in focus. That's why I've personally liked looking at KenPom and Sagarin for predictions, but the bracketmatrix for analyzing tourney invites. I think the point based models are also useful for looking at SOS.
Click to expand...

Exactly. KenPom is great for seeing if a team can win a single game. Not so much their overall season value, which the committee sees as W/L and SOS in relative W/L scenarios.
 
#185      
Illini1221
As good as some people think Kenpom is, I really dont like it much at all. A 6-7 Oklahoma team sits a spot ahead of us, at 11-4. Same thing with Syracuse, who has some bad losses. I get that it is more efficiency based, but it still seems like a terrible thing to go off of. Way too many numbers involved, and not enough concentration on the most important stat, which is to WIN. Not that it cant change throughout the conference season, but some of the teams in front of the Illini make zero sense to me.
 
#186      
danielb927
Illini1221 said:
As good as some people think Kenpom is, I really dont like it much at all. A 6-7 Oklahoma team sits a spot ahead of us, at 11-4. Same thing with Syracuse, who has some bad losses. I get that it is more efficiency based, but it still seems like a terrible thing to go off of. Way too many numbers involved, and not enough concentration on the most important stat, which is to WIN. Not that it cant change throughout the conference season, but some of the teams in front of the Illini make zero sense to me.
Click to expand...

There really aren't that many numbers involved in Pomeroy's rankings. How much did you score, how much did they score, how many posessions were there, and where'd you play. The other numbers like rebounding and shooting percentages are there to reveal more about how a team plays and what they do well, but aren't used as part of the rankings.

As others have pointed out, KenPom and Sagarin are good tools as predictors. For example, while one could think of several reasons (clutchness, matchups) why we should beat OSU in a hypothetical road game this year given our win on Sunday, the large-scale aggregate stats bear out that, knowing nothing other than "Illinois beat Ohio State by 5 in Champaign" you'd be wiser to expect a tOSU win.

On the flip side, winning IS the goal of the game, so you'll find no argument from me that wins should be heavily valued in awarding merit-based things like tourney berths.
 
#187      
UofIChE06
Illini1221 said:
As good as some people think Kenpom is, I really dont like it much at all. A 6-7 Oklahoma team sits a spot ahead of us, at 11-4. Same thing with Syracuse, who has some bad losses. I get that it is more efficiency based, but it still seems like a terrible thing to go off of. Way too many numbers involved, and not enough concentration on the most important stat, which is to WIN. Not that it cant change throughout the conference season, but some of the teams in front of the Illini make zero sense to me.
Click to expand...

The problem Kenpom and systems like his are having with this Illinois team is that the WVU and MD games can't be thrown out.... They are incredible drags on the efficiency numbers that take dozens of "normal" games to compensate for. Basically the issue is that this team has a penchant for vastly underperforming sporadically and only moderately overperforming very occasionally (ie VCU). If they were to come out and slaughter UW, PU, or even IU it would have a similar effect to the MD/WVU games but in the opposite direction.

That being said humans (ie the committee) doesn't work that way. A loss is a loss especially against good teams. Sure a bad loss like Winthrop could hurt but losing by 2 or by 50 to WVU or MD isn't going to make much difference come March. Same way with a win over someone like UW. It will look good no matter the margin. This is also in part why you see some of the upsets you do in the NCAA tournament. A lot of them are predictable using advanced metrics like Kenpom.
 
#188      
ChrisNMattoon
Looking ahead, I think we're going right to the wire at 9-9 in the B1G. 19-13 going into the BTT and maybe going 1-1 to finish at 20-14. Right square on the bubble and possibly in the play in game. Unfortunately, the way our program has gone, that would be a good thing for us.
 
#189      
danielb927
UofIChE06 said:
The problem Kenpom and systems like his are having with this Illinois team is that the WVU and MD games can't be thrown out.... They are incredible drags on the efficiency numbers that take dozens of "normal" games to compensate for. Basically the issue is that this team has a penchant for vastly underperforming sporadically and only moderately overperforming very occasionally (ie VCU). If they were to come out and slaughter UW, PU, or even IU it would have a similar effect to the MD/WVU games but in the opposite direction.
Click to expand...

I think you're overstating the effect of big results. Illinois' KenPom ranking was 71 before our first loss and did drop to 93 after the FSU game, BUT was right back up to 69 after the VCU win. And MD only dropped us 6 spots, not really that much of a drag. If a team has a tendency to lay an egg every so often, why shouldn't that be built into an assessment of that team's average ability for predictive purposes?
 
#194      
TriCityBurbs
Lunardi updated his bracket today:

West Virginia 2

Florida St 3

Wisconsin 5

Purdue 6

Minnesota 7

Michigan 9

Michigan St 9

Indiana 9

Maryland 10 (LAST 4 IN BYE)

NW 11 (LAST TEAM IN RND of 64)

ILLINOIS 11 (LAST 4 IN)

VCU 11 (LAST 4 IN)

NC St Next 4 Out

Ohio St Next 4 Out

The Webers Next 4 Out

Would be nice to have that Florida St game back, did they really look like a top 12 team when we played them? I don't think so...
 
#195      
ChiefIllini
TriCityBurbs said:
Lunardi updated his bracket today:

Would be nice to have that Florida St game back, did they really look like a top 12 team when we played them? I don't think so...
Click to expand...

9 B1G teams in? I can't forsee this happening. Even if we are one of the last 4 in.
 
#197      
ChiefIllini
Second and Chalmers said:
Why not? *sharpens knives* ;)
Click to expand...

Well.......:D

For one, if 10 teams are either in or on the bubble (OSU is one of the first four out in his bracket), and the highest seed a B1G team gets is a 5, how many quality wins can a team like Illinois have that would put them in over a team with a win or two less? Maybe I dont understand how the committee views quality over quantity but to me a team like NC St. has pretty much 10 more chances to get quality wins whereas the Illini have maybe 5. Of course we could delve into NC St. specifically since we beat them and then I would say even if they have better wins we should make it if we have more wins, but I think generalizing is a better exercise.

Same goes for NW, OSU, MD, and Michigan. I dont think any of them have more than maybe one quality win already with only a few chances to get another one just because of how diluted the B1G is.

Then again in the same vein you could say if 4 or 5 of the B1G teams have between 20 and 22 wins, I'd say some of those are top 50 teams, if not all, and therefore should be ahead of teams with weaker wins. in other words, maybe 3 or 4 of those teams only have 1 top 25 win, but they probably have 7 or 8 top 50 wins, because of how closely each of these B1G teams are wrt skill and ranking.

I see both sides of the argument. My question is whens the last time if ever the B1G got 9? I can picture it happening, but I dont think its likely.
 
#198      
Fighter of the Nightman
I was always under the impression that the committee uses a very, very abstract judgement process in comparison to those rankings, with the possible exception of letting a really high or low RPI make or break a borderline team's chances. They seem to take pride in having a sort of grandiose and nostalgic decision making process when it comes to what teams deserve to be in.

An Illinois team with 22 wins is in good shape, so let's make it happen. :)
 
#199      
ChiefIllini
Fighter of the Nightman said:
I was always under the impression that the committee uses a very, very abstract judgement process in comparison to those rankings, with the possible exception of letting a really high or low RPI make or break a borderline team's chances. They seem to take pride in having a sort of grandiose and nostalgic decision making process when it comes to what teams deserve to be in.

An Illinois team with 22 wins is in good shape, so let's make it happen. :)
Click to expand...

Sadly I was putting us at 20 or 21 wins. 22 gets us in, maybe not comfortable, but in nonetheless. Agreed, when they give reasoning behind certain teams over others they tend spit out some weird reasons that sound logical but also a little nitpicky. And yeah I've heard them reference SOS and RPI every so often to explain how a 19 or 20 win team makes it in over teams with 21 or so wins. So far, our RPI and SOS are awesome. Hopefully we remain right around 25 in both. I think that could be the difference in pushing us over the bubble. I also think it would be sweet to be a last four in and nw be a last four out :). Though I'd settle for a 6 seed as the kindergartner-designed website has us right now.
 
#200      
TriCityBurbs
Have to remember how weak the bubble has been in the last 4-5 years. We already have 4 (5 if Winthrop keeps winning) top 100 wins and 4 of those teams could finish the year in the top 50. Our worst loss is against RPI 104, and because of how strong the bottom of the big ten is this year (Rutgers could still finish in the top 100 RPI) we don't have another opportunity at a really bad loss. In other words, we have already played 7-8 top 100 teams, and could end the year with 23-26 games against the top 100. Even if we finish 9-9 in conference that's probably 12-13 top 100 wins,and 8-9 vs the top 50. Add in our very strong NonCon SOS, overall SOS and a top 50 RPI...I still think we can make the tourney with just 20 wins total.
 
Status
Not open for further replies.
Back