MBB: College Conference of Illinois and Wisconsin

[WUPHF]

And how would you resolve the Wash U-IWU-Augustana triangle?

Washington University wins on points differential. 

[izzy stradlin]

Outside of preseason expectations, what would be the explanation for such a gap with these 2017-18 resumes?

http://www.d3hoops.com/teams/Augustana/men/2017-18/index

http://www.d3hoops.com/teams/Illinois_Wesleyan/men/2017-18/index

I don't understand why the teams are so far apart from each other either, but to try to answer your question...

- The most obvious data point is IWU pounding Augustana in Shirk. My guess is that voters are chalking that up to an anomaly and want to see what happens when they play in Rock Island.
- Let's consider the teams' losses to Carthage and Wheaton as a wash. Both were close home losses to quality conference opponents.
- Augustana's other loss is double-overtime to No. 6 Oshkosh at their place. Augie also has a win over No. 3 Wash U by a comfortable margin in their gym and a win over current WIAC leader Stevens Point.
- IWU's non-conference record includes a loss to No. 17 Emory and a non-competitive loss to No. 3 Wash U. I'm not sure what the Titans' best non-conference win is.

I think most pollsters would slot IWU after Emory (given the head-to-head result) and Emory isn't going to be terribly high because of its non-conference losses. Add in the Augustana's win and very competitive loss to teams in the Top 10 and the fact that IWU doesn't have anything similar on their schedule. The most obvious data point here is IWU bombing Augustana, but that's a conference game on the Titans' court so some people will discount that or at least wait for the rematch.

If I voted in the Top 25 for men, I'd probably want IWU higher but below Emory (assuming they are on my ballot). Based on watching the Vikings this year, I still think Augustana is one of the best 10 teams in the country, so I don't want to drop them farther than they are. And I end up with a spread pretty close to what we have here without referring to last season.

I agree this is probably the typical logic of the current top 25 voter.   The reasoning is so full of biases (selection, recall, confirmation, anchoring) though, it shows why the recent top 25 polls have gotten to be pretty bad. For example the reference to the rankings of Washu, UW Oshkosh etc, introduces bias. 

I would look at overall W/L% and some single metric to estimate strength of schedule using all games played. 

[Dave 'd-mac' McHugh]

And how would you resolve the Wash U-IWU-Augustana triangle?

Washington University wins on points differential. 

POINTS!

Bravo.

[gordonmann]

I agree this is probably the typical logic of the current top 25 voter.   The reasoning is so full of biases (selection, recall, confirmation, anchoring) though, it shows why the recent top 25 polls have gotten to be pretty bad. For example the reference to the rankings of Washu, UW Oshkosh etc, introduces bias. 

I would look at overall W/L% and some single metric to estimate strength of schedule using all games played.

Fair point.

So where would you rank the two teams then?

[Dave 'd-mac' McHugh]

I agree this is probably the typical logic of the current top 25 voter.   The reasoning is so full of biases (selection, recall, confirmation, anchoring) though, it shows why the recent top 25 polls have gotten to be pretty bad. For example the reference to the rankings of Washu, UW Oshkosh etc, introduces bias. 

I would look at overall W/L% and some single metric to estimate strength of schedule using all games played.

While Gordon thinks it is a fair point, I'll point out that the Top 25 usually stands up pretty well every season.

I will also add about your last point: Iknow a lot of voters who use all kinds of metrics along with other things like eye test and what-not. They don't just slap things on a wall or throw darts (as I have joked in the past).

You may consider there are biases, but that is also why there are three voters from each region in the poll. Of course, you have bias as well.

[izzy stradlin]

I agree this is probably the typical logic of the current top 25 voter.   The reasoning is so full of biases (selection, recall, confirmation, anchoring) though, it shows why the recent top 25 polls have gotten to be pretty bad. For example the reference to the rankings of Washu, UW Oshkosh etc, introduces bias. 

I would look at overall W/L% and some single metric to estimate strength of schedule using all games played.

While Gordon thinks it is a fair point, I'll point out that the Top 25 usually stands up pretty well every season.

I will also add about your last point: Iknow a lot of voters who use all kinds of metrics along with other things like eye test and what-not. They don't just slap things on a wall or throw darts (as I have joked in the past).

You may consider there are biases, but that is also why there are three voters from each region in the poll. Of course, you have bias as well.

That's not the kind of bias I'm talking about.

[Pat Coleman]

If this is an attempt to get back to the statistics course's definition of bias, I think it's been clear that we are generally using the vernacular. 

Not having a poll until January 1 is not really an option. Why would we let Sporting News be the only preseason poll? I don't see that as better.

Diligence of the voters is always the answer, no doubt. That's always something that can be adjusted for by changing voter panels. However, just because a few posters think the poll is "wrong" doesn't mean that's actually true.

[Gregory Sager]

Wednesday's games:
Augustana (12-3, 4-2) @ Elmhurst (10-5, 3-3)
Wheaton (11-4, 5-2) @ Carthage (9-6, 4-2)
Millikin (8-7, 1-5) @ Illinois Wesleyan (12-3, 5-1)
North Park (4-10, 1-5) @ Carroll (4-11, 0-6)
Finlandia (3-10) @ North Central (11-4)

Massey sez:

Augustana 78, Elmhurst 72	AC 72%, EC 28%
Carthage 78, Wheaton 76	CC 57%, WC 43%
Illinois Wesleyan 76, Millikin 63	IWU 90%, MU 10%
Carroll 69, North Park 65	CU 65%, NPU 35%
North Central 92,  Finlandia 57	NCC 100%, Finlandia 0%

The Finlandia @ NCC game is tonight, Tues., 1/16 @ 6:00, not tomorrow night.

Yep, I knew that and missed it, anyway. Sorry. Thanks, Mark.

I think conference games are sometimes looked at differently than non-conference games. I don't necessarily take that adage, but I know there is a rematch coming with IWU and Augustana.

Some voters weigh non-conference results differently than conference results because of familiarity between the teams.

A basketball game is a basketball game. Stop me if I'm wrong, but both the IWU vs. AC and the IWU vs. EU contests featured two baskets -- one on each end of a 94-foot wooden court -- and one basketball, plus a forty-minute game clock and five players per team, right?

Frankly, you can read the results the exact opposite way than the read you're attributing to "some voters". If two teams are familiar with each other, I'm apt to be more impressed by a sizable margin of victory than I would be in a game played between two teams that aren't familiar with each other. In other words, if this reasoning is going to be used to weigh one game against another, I'd put more stock in Illinois Wesleyan's thumping of Augustana than I would in Emory's thumping of Illinois Wesleyan. (I'll add that IWU was far more competitive against Emory than Augie was against IWU, but I don't expect pollsters to delve into the minutiae of how games actually played out.)

Surely you've seen instances where one team beats another in conference where the victor isn't considered clearly better than the vanquished, right?

FTFY.

Where would you slot the two teams? And how would you resolve the Wash U-IWU-Augustana triangle?

I wouldn't slot Augustana 16 rungs up the ladder and a whopping 368 points above Illinois Wesleyan, that's for darned sure. All that that gargantuan gap between the two teams tells an objective observer is that your poll has an irrefutable anchoring bias that's based upon a preseason poll that ought to have zero effect upon where teams stand on January 14. And, given that the WU/AC/IWU triad played three largely noncompetitive games between each other, I'd call it a wash (no pun intended) at this point. If anything, if I was measuring the three head-to-heads I'd give a slight edge to the Bears, who at least made a modest second-half comeback effort in their loss to Augie and who did their thumping of Illinois Wesleyan on the home court of the Titans -- which is where IWU, in turn, did its thumping of Augustana.

[Pat Coleman]

Lots of smug in here. Voting in this poll and balancing all that crap is damned difficult, and I'm glad I don't have to do it any longer.

[Gregory Sager]

If this is an attempt to get back to the statistics course's definition of bias, I think it's been clear that we are generally using the vernacular. 

I agree, which is why I only use the word in relation to the very specific statistics-course term "anchoring bias".

Not having a poll until January 1 is not really an option. Why would we let Sporting News be the only preseason poll? I don't see that as better.

I don't think that anybody's asking you to ditch the preseason poll, Pat. To be honest, I'd be very upset if you did, because the preseason poll put out by The Sporting News isn't worth the paper it's printed on (they still do put out a print edition, don't they?), and your preseason poll is about as scrupulously well-researched as any fan could reasonably ask. But why not put out the preseason poll and then begin the in-season poll after New Year's Day? Why not give the pollsters the latter part of November and all of December off and let them just watch basketball? It might not get rid of the anchoring bias altogether, but I'll bet that it gets rid of most of it, at least. It would remove the organic connection from one poll to the next prior to New Year's Day. That's my bone of contention: Pollsters get in the habit of using last week's poll as a template and then just re-slotting wherever there's a significant win or loss to be accounted for -- and they include the preseason poll as a part of this process when creating their first in-season ballots.

[Gregory Sager]

Lots of smug in here. Voting in this poll and balancing all that crap is damned difficult, and I'm glad I don't have to do it any longer.

Hey, you know me, Pat. I've been going after you about this issue for a decade and a half now. Now it's Gordon's turn in the barrel.

[Pat Coleman]

You can't really ask voters to not include the preseason poll as part of that first ballot, which is about four to five games into the season. But I agree that by the time the new year rolls around, at the very least, it's time to let last season go.

Not having regular-season polls for six weeks is no better than not having a preseason poll. We don't want someone referring to Claremont-Mudd-Scripps as No. 15 when they are 3-7 on Dec. 31. It's grating enough to see that for two weeks.

As regards the term bias, Gregory, I was referring to Izzy's post immediately before mine.

[Dave 'd-mac' McHugh]

My last thought: how does any of us know exactly how the voters are voting? I am the only one I know who puts their ballot out there for the public and even explains what I am thinking. If we are assuming all are doing the same as me, then there is a problem. I don't even know how more than five of the voters are voting or how they are thinking.

To assume preseason thinking is still in play is just that - an assumption. It isn't in my head. I've seen Augustana games and I feel they are better than IWU right now in the way they play outside of the head-to-head. Does anyone really know if the other 24 voters (maybe 23, Ryan weighed in) are thinking the same or differently?

While my blog will be out later, I have IWU 25 and nervous and admittedly low. I have liked them since early in the year, but there are items I just can't grasp if they are as good as they seem or not. I wanted them higher, but I need to keep them below Emory (16) and I didn't like them above the others as of right now. Doesn't mean I'm right; doesn't mean I am wrong.

I have Augustana now 10th. Feels high. I would want them lower to be sure, but they are there because I think they are better than the other 15 teams I ranked. That doesn't mean in a normal year they would be 10th, but they are there because I need to put someone tenth. The last two weeks I held on to them at 2 because, again, everyone was losing around them as well and I didn't see a place where I could put them where I felt there were other teams better. I did weigh in their win over WashU last week when I moved WashU up. I rather have Augie maybe 15, but...

There were nearly 50 teams I was considering this week (and last). It just isn't as easy as many think it is. Not when you know a national poll (and others) is counting on your vote to be well thought out, researched, and such. The biggest challenge... when every think you are using doesn't produce an answer. It only produces more questions.

[AndOne]

I'd vote for Francis, Lake and Brady Rose for first team CCIW. 

Likely Sorenson and ? to round out the first team -- perhaps Raridon? 

Surely depends to some degree now on which teams rise to the top, who wins the CCIW championship.

Raridon on the first team at this point in the season would provide for an interesting discussion of anchoring bias.

There is no discussion. If anyone thinks that Connor Raridon isn't, minimally, one of the best players in the CCIW then I'm sorry, but they don't know a basketball from their rear end.
Connor has been a wing/forward his whole basketball life. In that role, he has excelled. You can look it up. Gong into this season, he expected to be anything but a point guard. However, due to a season ending injury, NCC was put in a position of having to find a replacement at the point. Much of the choice involved the question of inserting someone with little or no college level experience or switching an experienced player to the spot generally responsible for the offensive direction of the team.
Because of his overall basketball skills, Connor became the choice to switch to a position at which he had next to no previous experience. This is something that only a shooter, only a rebounder, or any player only superior in any one or even two aspects of the game would not be able to successfully accomplish. Additionally, doing so would require a player able to primarily concentrate on
a part of the game that previously was a secondary skill—a switch from primarily being a scorer and rebounder to being a ball handler and passer. This is something that many players wouldn't be able to handle too well. However, because of his overall skill level, Connor has been able to make a pretty darn successful transition. Ask yourself what are the things you want your PG to do. Many people are going to answer that question with something like "I want him to make good decisions, run the offense smoothly, avoid turnovers, pass more often to a teammate than someone with a different color jersey. And, it wouldn't hurt if he can be enough of a scoring threat that defenders couldn't sag off him and double on another player." Well, that is exactly how Connor is operating. Perhaps the most telling of stats as to how a PG is running things is A/TO Ratio. Checking the stats for both all games and conference games only reveals Connor is the leader in this area by a pretty wide margin. He has more assists, and fewer turnovers than players that have played the point their whole life. Additionally, while essentially learning a new position on the fly, he is in the top 10 players in the conference in 5 statistical categories, and in the top 15 in 3 more.
And, should any doubt remain, Connor almost always guards the opponent's top scoring forward. Unlike a lot of guys who can only score, he is a complete player.
He has sacrificed some of his scoring and rebounding abilities—the things that usually garner the most attention when it comes to regional and national recognition—for the good of his team.
Lastly, find me a coach who wouldn't want Connor on his team. Does that sound like someone that there should be any "anchoring bias" discussion about? Or, for that matter, any discussion at all. 

[gordonmann]

I wouldn't slot Augustana 16 rungs up the ladder and a whopping 368 points above Illinois Wesleyan, that's for darned sure. All that that gargantuan gap between the two teams tells an objective observer is that your poll has an irrefutable anchoring bias that's based upon a preseason poll that ought to have zero effect upon where teams stand on January 14. And, given that the WU/AC/IWU triad played three largely noncompetitive games between each other, I'd call it a wash (no pun intended) at this point. If anything, if I was measuring the three head-to-heads I'd give a slight edge to the Bears, who at least made a modest second-half comeback effort in their loss to Augie and who did their thumping of Illinois Wesleyan on the home court of the Titans -- which is where IWU, in turn, did its thumping of Augustana.

You didn't answer the question so I'll try again.

Where would you have them ranked? Voters don't have the option of casting a ballot that says, "Here's where they aren't ranked."