D1 Team Race

[CLC FAN]

Odds for the Division 1 Team State Championship for 2016.  These are odds against, so 3:1 is a 25% chance.  No bets please.

Kaukauna         3:1
Hudson            9:1
WI Rapids        11:1
Stoughton        12:1
Milton             12:1
Burlington        16:1
Oak Creek       23:1
Bay Port          33:1
Stevens Point   72:1 
Sauk Prairie     53:1
Pewaukee        67:1 


Please note that these are based on projections of what will happen, not a ranking of who the best teams are.  The difference is that a ranking list does not take into account which regionals teams are in or which side of the team state bracket they fall on.  These projections do take those factors into account.  If you are looking for a fairly in-depth explanation of where these numbers come from, head over to the D2 Team Race thread.  Suffice it to say that historical data as well as returning starters were used to run simulations of all 16 regionals and the entire team sectional/ team state dual series.

For comparison sake, you can currently find the following Super Bowl Odds for 2016

Green Bay      6:1
Seattle           6:1
Indianapolis    8:1 
New England   9:1 
Dallas            14:1 
Chicago          60:1

[ElectricGuy]

First CLC -  I love your post and your analytical approach to prediction.  Always very enjoyable well written post.

But as for prediction, don't you feel a bit challegend to go back to the drawing board on your model if you at least look at D1 in a vacuum?  You have to toss Kaukauna out, because even my eleven year girl could of predicted them to win with that lineup, but for the final 8 down at the field house you had 37.5% correct.   You did ok with your field of 16 (Team sectionals)  Seems like were missing a piece to your puzzle - ability to shift lineups, depth, pinning ability, new freshman......  ??

1st Place - Cedarburg
1st Place - Kaukauna x
1st Place - Wisconsin Rapids Lincoln
1st Place - Pewaukee
1st Place - Sauk Prarie x
1st Place - Burlington x
1st Place - Stoughton
1st Place - Hudson

Well Coach, my research (moneyball analytics I believe you referred to them as) is usually more along the lines of forecasting rather than ranking.  So with that in mind, here is my team sectional and team state bracket for D1.  The percentages following the regional winners is the confidence I have in those teams winning their regional.  Those confidences come from at least 100 simulations of each regional using predictions for each team's number of wrestlers and each team's average points per wrestler, and then allowing for random variation.  Too wordy, I know, but that's just me.

D1   Kaukauna (99%)
                               Kaukauna
D2 Osh. North (42%)
                                                   Kaukauna
H1   Burlington (80%)
                               Burlington
H2 Oak Creek  (50%)
                                                                      Kaukauna
A1 Menomonie (77%)
                               Menomonie
A2    Merrill   (50%)
                                                   Menomonie
G1 Men. Falls  (76%)
                               Men. Falls
G2 Whtfsh Bay  (32%)
                                                                                        Kaukauna
B1   Bay Port (89%)
                                 Bay Port
B2 Wis. Rapids (47%)
                                                    Bay Port
F1 Pewaukee (83%)
                               Mukwonago
F2 Mukwonago (63%)
                                                                      Milton
C1 Sauk Prairie (82%)
                               Sauk Prairie
C2 De Forest (38%)
                                                   Milton
E1 Stoughton (66%)
                                  Milton
E2   Milton  (70%)

[CLC FAN]

First CLC -  I love your post and your analytical approach to prediction.  Always very enjoyable well written post.

But as for prediction, don't you feel a bit challegend to go back to the drawing board on your model if you at least look at D1 in a vacuum?  You have to toss Kaukauna out, because even my eleven year girl could of predicted them to win with that lineup, but for the final 8 down at the field house you had 37.5% correct.   You did ok with your field of 16 (Team sectionals)  Seems like were missing a piece to your puzzle - ability to shift lineups, depth, pinning ability, new freshman......  ??

1st Place - Cedarburg
1st Place - Kaukauna x
1st Place - Wisconsin Rapids Lincoln
1st Place - Pewaukee
1st Place - Sauk Prarie x
1st Place - Burlington x
1st Place - Stoughton
1st Place - Hudson

Funny you should mention that Guy.  This is a new system for predicting Team State Duals in the preseason.  D1 ate me up last year, no doubt about it.  Granted, it was November projections of what would happen in March, but I still would obviously like to do better.  In the past years I haven't used historical data to the extent that I am this year.  I also plan on updating not only these but also my regional predictions as the season unfolds.

In my defense specifically about D1 last year, I went 0-2 on two coin-flip scenarios that I think everyone in the state knew were going to be close.  If either the Stoughton/Milton matchup or the Rapids/Bay Port matchup goes the other way, I get not only another quarterfinalist correct, but likely also a semifinalist (and in Stoughton's case a finalist right too).  If both those coin-flip duals go my way, the projections end up looking pretty prescient.  

 The 4 team characteristics you specifically mention - ability to shift, depth, pinning ability, and freshmen - are not things I am able to quantify right now.  I suppose for depth you could introduce a variable for how many kids each team skinfolds, since that is public information.  Ability to shift, I guess I don't know how to gauge that in the preseason, since it would be a total guess as to where the returning kids' weights fall.  I do (now) give teams a boost who have been to Team State before and an even bigger boost to teams that have been there and done well - assuming they know how to play those chess games with the lineup better than other teams.  Pinning ability... right now for my own team I calculate each kid's PPW and PPL - his pins per win and pins per loss.  It ends up reading like a batting average.  If a kid has a .750 PPW, I know he can put people away, but can he win a close match?  If a kid has a .500 PPL, we need to talk about staying off his back and fighting off his back.  I wish such statistics were readily available for other teams, because you are right, it would be helpful.  One thing that I would just guess is that a team like Ellsworth has a ridiculously low PPL.  Even their lowest starters know how to keep a match close.  I remember being at Team State last year and Freedom had won the first 5 matches and was only up 15-0!  As for freshmen, sigh, I have no idea how to quantify who is coming, what school they are coming to, and what impact they will have.  The biggest frustration is the way Kids State looks on trackwrestling and the difficulty of tracking where kids are actually from or where they go to school.  It would be so nice to at least have the kids' grade in school listed on the bracket, rather than just knowing the birth years.
 I guess what I'm saying is if you have any good ideas for quantifying any of these variables that you mention - I am all ears.  

[CLC FAN]

Also, I am curious within your formula what tipped the scale (weighting factor) for Big K, I was surprised at their great odds, I know they have 4 returning state qualifiers but......??

For the past 5 years, an average Kaukauna starter has been worth 7.1 EDWIN points - this is a measure of expected contribution towards a win at Team State.  For context, a state champ is expected to contribute 10.3 EDWIN pts, a non-placing state qualifier from an average sectional: 7.4 EDWIN pts, and a wrestler who took 3rd at an average sectional: 6.8 EDWIN pts.  When Kaukauna graduated half their lineup last year, my system input 7 "average starters" for Kaukauna to build a prediction for this year. 
So, my dual meet prediction system doesn't take into account at all what you lost to graduation.  Maybe it should.  I like your "wins replacement" metric.  Right now my system expects two things: returners from last year will perform exactly as they did last year.  Incoming replacements will perform to a team's historical average.  Both of these are incorrect assumptions, but they offset each other to a large degree.  Returning starters should improve - in some programs more than others.  Also, we would expect incoming replacements to be slightly below an average starter, since it is their first year on varsity.  So returners are slightly underestimated while newbies are slightly overestimated.

[boowrestle]

FYI to everybody,Sparta is now Div1,jmo they are my pic behind kaukauna,

[CLC FAN]

FYI to everybody,Sparta is now Div1,jmo they are my pic behind kaukauna,

I have Sparta at 92:1, next behind Pewaukee.  They are a great team - the challenges for them are Holmen at regionals and then a stacked Sauk Prairie at Sectionals. 

[Hand Control]

Also, I am curious within your formula what tipped the scale (weighting factor) for Big K, I was surprised at their great odds, I know they have 4 returning state qualifiers but......??

For the past 5 years, an average Kaukauna starter has been worth 7.1 EDWIN points - this is a measure of expected contribution towards a win at Team State.  For context, a state champ is expected to contribute 10.3 EDWIN pts, a non-placing state qualifier from an average sectional: 7.4 EDWIN pts, and a wrestler who took 3rd at an average sectional: 6.8 EDWIN pts.  When Kaukauna graduated half their lineup last year, my system input 7 "average starters" for Kaukauna to build a prediction for this year.  
So, my dual meet prediction system doesn't take into account at all what you lost to graduation.  Maybe it should.  I like your "wins replacement" metric.  Right now my system expects two things: returners from last year will perform exactly as they did last year.  Incoming replacements will perform to a team's historical average.  Both of these are incorrect assumptions, but they offset each other to a large degree.  Returning starters should improve - in some programs more than others.  Also, we would expect incoming replacements to be slightly below an average starter, since it is their first year on varsity.  So returners are slightly underestimated while newbies are slightly overestimated.

FYI to everybody,Sparta is now Div1,jmo they are my pic behind kaukauna,

I have Sparta at 92:1, next behind Pewaukee.  They are a great team - the challenges for them are Holmen at regionals and then a stacked Sauk Prairie at Sectionals.  

Sparta and Sauk just finished the Luther team camp.Sauk won the team tournament on tie breaker criteria over Sparta. The big picture in this should take into account that Sauk had two filler kids(Licht-RV and Hanson-Deerfield).So in my opinion this makes Sparta a stronger TEAM,thus throwing the earlier odds out the window.Sparta is also returning a lot of good kids and throw in a national champ incoming freshman,a state champ and at least four other state qualifiers and that Krien kid ( who by the way I hope makes some HUGE noise at Fargo this week)= a much stronger team than Sauk.

[DarkKnight]

Any D1 teams drop to D2? just wondering, for the ranking staff's sake. I like Stoughton's incoming guys/returning guys as much as I like KK's. I'd have to compare Sparta and Stoughton's projected lineups to see who would be projected favorites, everything being just my projections, of course.

[Troy Grindle]

CLC,

Great post.  I always enjoy the numbers look at possibly where everybody will shake out.  It usually is pretty close and accurate.  Did you break down each regional like you did last year before regionals to figure out a regional winner and the percentage that they would win?  Those were really interesting to see who would have the close regionals and to look back on them to see where the regional was won and lost.

[CLC FAN]

As requested, here are my forecasts for regional results.  Please let me know if there are any problems viewing this link.

https://docs.google.com/spreadsheets/d/1J5HjP43tSvaWyWyDv3It00bXt4naEzuKOMnr1qf4XVo/pubhtml?gid=1314978627&single=true

[CLC FAN]

Also, for what it' worth... here is my Division 1 Team Series bracket. 

https://docs.google.com/spreadsheets/d/1J5HjP43tSvaWyWyDv3It00bXt4naEzuKOMnr1qf4XVo/pubhtml?gid=0&single=true

[Ghetto]

20% chance of us winning the regional? Thanks CLC. I think you fudged the numbers way up for us. Love you man!!!

[DarkKnight]

Pretty accurate estimates I must say. Hudson will give KK a real good scare, and Stoughton also has a good shot, but I agree KK is SLIGHT favorites, with projected lineups.

[EIUwrestler1]

In the Menomonee Falls regional, Germantown is way under rated.  They didnt lose a single starter and always have a full line up IMO.

Cedarburg appreciates the love tho. We reload some, but lost a lot of 100 win kids

[Hand Control]

I would not sleep that hard against Rhinelander. IMO They are bringing back some great kids and I will say that they turned some heads in Wuanakee last year.And I have to say this,Is it wrestling season yet?