Sunday, December 30, 2012

Girl's Basketball Sectional Forecast #1

Ladies and gentlemen, the first girls forecast is listed below followed by comments.  The boys will follow either tonight or tomorrow.  We are about a third of the way through the season with 287 games logged and a lot will change between now and when the Sectional Seedings come out.  So don't yell I'll do my best to explain.

I do want to preface this by pointing out that the model gives each team a percentage value.  This is the order of those values.  It's based on what round in the tournament they should get to, so a team that gets a .900 in the model would be projected to make the final, regardless of seeding position.  

Class AA

(1) Bethlehem  6-0
(16) Bye
(8) Catholic Central  4-4
(9) Columbia  4-3
(5) Bishop Maginn  6-3
(12) Niskayuna  1-4
(4) Colonie  5-2
(13) Schenectady  0-6
(3) Shaker  5-0
(14) Bye
(6) Guilderland  3-2
(11) Ballston Spa  1-6
(7) Shenendehowa  3-1
(10) Saratoga Springs  3-3
(2) Albany  6-0
(15) Bye

First things first, the model gives the Suburban the top spot even though both the top teams in the Big 10 and Suburban have the same record.  That seems logical to me.  Bethlehem is performing greater than the historical averages would suggest against their schedule than Albany is against theirs.  Bethlehem also has a greater MOV.  The same two points can be made for Albany over Shaker.  At least the model is consistent.  Shen might be the only one that is slightly out of line if you put them in order by winning percentage and that has to do with their really low MOV.  History only has them as a .500 club at their current rate.  Finally, Bethlehem is the only team that is close to the assigned value where the model puts them in the final outright.  Bethlehem and Albany are both over the threshold for the final four and Shaker and Colonie are close.

Class A

(1) Troy  4-1
(16) Bye
(8) Mohonasen  2-5
(9) Amsterdam  2-4
(5) Averill Park  4-3
(12) Bye
(4) Queensbury  5-1
(13) Bye
(3) Glens Falls  4-2
(14) Bye
(6) South Glens Falls  6-2
(11) Lansingburgh  0-6
(7) Scotia-Glenville  4-3
(10) Burnt Hills  1-6
(2) Holy Names  6-0
(15) Bye

This one gets a little more complicated.  Troy (you can put an asterisk next to this one because I haven't heard if the girls are playing in AA as the boys are) the only one here who even crosses the final four mark, though Holy Names, Glens Falls and Queensbury are close.  So why does Troy outscore Holy Names even though they have a loss and two less wins?  The answer has mostly to do with who each team has played.  Troy's opponents not only have larger enrollments, they are also performing better than expected by 2 full games over an 18 game schedule.  Holy Names opponents are performing 2.5 games below average in an 18 game schedule, so Troy's opponents are 4.5 games better than average than Holy Names' and have larger enrollments.  It gets substantially worse from there as I now have to explain how Glens Falls is ahead of Queensbury even though Queensbury has now beaten them twice.  Oh yeah, and South Glens Falls is two spots below Queensbury even though they beat them once.  The answer is a mishmash of unbalanced schedules, MOV's and expected winning percentages all falling just right and would take half my day to explain.  This one will sort itself out as they complete their league schedules and their common opponents come into line.  These are the inherent dangers of running the model so early in the season.

Class B

(1) Watervliet  6-1
(16) Chatham  3-5
(8) Broadalbin-Perth  3-3
(9) Corinth  4-2
(5) Greenville  5-2
(12) Catskill  3-5
(4) Ravena  4-2
(13) Johnstown  2-4
(3) Tamarac  7-1
(14) Hudson  4-4
(6) Emma Willard  5-2
(11) Cobleskill  4-3
(7) Fonda-Fultonville  3-4
(10) Schalmont  5-2
(2) Taconic Hills  6-2
(15) Cohoes  5-3

Unranked:  Granville, Albany Leadership, Bishop Gibbons, Coxsackie-Athens, Schuylerville, Hudson Falls, Ichabod Crane, Cairo-Durham

On the surface I'm ok with this one.  The top three are right regardless of order and the order will straighten itself out.  There are some problems within the Colonial Council teams much like with the Foothills in A, but again, they'll work out themselves in the next few weeks.  The interesting thing about this one is that none of the teams are above the final four grading.  I'm curious if that changes as the year goes on and Watervliet hits their league opponents. 

Class C

(1) Fort Plain  6-0
(16) Schoharie  2-3
(8) Berne-Knox  4-1
(9) Duanesburg  4-2
(5) Middleburgh  4-1
(12) Mekeel Academy 2-4
(4) Lake George  6-1
(13) Whitehall  2-3
(3) Waterford  6-2
(14) Canajoharie  4-4
(6) Hoosic Valley  5-2
(11) Greenwich  4-3
(7) Galway  4-2
(10) Hoosick Falls  4-2
(2) Maple Hill  9-1
(15) Mechanicville  4-3

Unranked:  Albany Academy, Mayfield, Stillwater, Saratoga Catholic, Berlin, Hadley-Luzerne, Rensselaer, Cambridge, Voorheesville

What I believe you'll see in the C bracket is a ton of movement from week to week.  Hoosic Valley has already moved up one from my initial run two days ago and the separation between 2 and 8 is pretty slim.  I think the biggest shock here is how low Berne-Knox is, but any bracket that can put a team of that caliber 8th under any circumstances is going to be a tough bracket to win for any of the teams.  Fort Plain is further ahead of Maple Hill than Maple Hill is of Galway at number 7 and that is almost solely due to their undefeated record.  Any way you slice it this bracket is packed and I think the model is doing a fairly good job so far of sorting it out.  Remember, it's made for a full season of information not a third.

Class D

(1) Fort Edward  5-0
(16) St. Johnsville  0-3
(8) Hartford  1-4
(9) Northville  3-5
(5) Loudonville Christian  3-2
(12) Salem  1-4
(4) Fort Ann  6-0
(13) Doane Stuart  1-5
(3) Heatly  5-2
(14) Sharon Springs  0-2
(6) Germantown  2-3
(11) New Lebanon  2-4
(7) Warrensburg  3-3
(10) North Warren  2-4
(2) Argyle  4-2
(15) Bolton  0-6

Unranked:   Hawthorne Valley

Let me start by saying that Fort Edward is the only team to have a forecast over 1.000 which means the model thinks they should win the whole thing, not just get to the final.  I tend to agree with that sentiment.  The most notable surprise here is that Fort Ann is only 4th even though they are undefeated and beat Argyle by 20.  The reason for this is that other than Argyle not one team they have played is doing anywhere near as well as can be expected from historical trends.  In fact, they are a combined 4-19 excluding their losses to Fort Ann.  This will work itself out when each team plays everyone in the Adirondack League and their opponents expected percentages even out. 

This wasn't bad for a first run especially when most teams haven't even made it once through their in conference schedules.  The accuracy and reliability will improve with each game that's played.  I'll be working on the boys forecast tonight and hopefully I can get it done, if not, tomorrow sometime. 

No comments:

Post a Comment