In partnership with CBSSports.com
The largest and most active MSU Spartans board on the web
The place to ask questions to SpartanTailgate's recruiting experts
"The Duff" is dedicated to Michigan State football recruiting discussion
"The Bres" is dedicated to Michigan State basketball recruiting discussion
This is your pulpit to preach to the masses about everything from politics to religion
The place to buy, trade or sell Michigan State tickets
You have no favorite boards.
The most viewed topics.
The most replied to topics.
The most up-voted topics.
The most down-voted topics.
The most up-voted posters.
The most down-voted posters.
The most followed posters.
Assigning probabilities of wins for the remaining schedules of MSU, UM and IU I arrive at a maximum liklihood estimate of the finl standings:
@ Illini 0.5
@ OSU 0.5
@ MSU 0.25
@ Minn 0.5
I have attempted to assume these teams are equal in strength when assigning these probabilities: A 75% chance of beating the 3 top teams at home and a 100% chance of beating everyone else at home. A 50% chance of beating the next tier on the road (OSU, UW, PU, Illini, Minn) and a 100% chance of beating the lower tier on the road.
Yes, I am ignoring OSU as one of the heavyweights in this exercise.
This post was edited by Big Green Stick 17 months ago
Why do you give us a better chance @OSU than IU does?
I'm guessing you worked on this all of Friday night.
I think I caught that mistake and fixed it.
About 20 minutes in Excel just now.
The conclusion: This exercise was a waste of time to determine any favorite and points out that no team of the 3 really has much of advantage based on schedue alone. May the best team win.
What the fuck is this? MGoBlog?
Your a retard.
I've come to the same conclusion while reading this tripe.
Come on guys, how about a little something for the effort?
Thank you for this
so mattyfresh and so clean
Don't let the haters get you down. Shine on you crazy diamond.
I have us winning out.
"As far as the downvotes. It's a gnat biting an lion"
-- A member of tRCMB Justice League, taking the internet WAY WAY too seriously.
Can you edit the title to start with "Warning Science"?
Mods, can you pin this to the top?
13.75 wins? Sweet.
That extra .75 win should get us a 2 seed!
I like the idea, but the probabilities you assigned to wins seem quite arbitrary. Do you have any data to support your selections?
I have back fit this model to the current standings and would have predicted:
Edit: -0.25 correction on UM
So the actual vs. model has MSU and IU at -0.25 and UM at +.50 wins.
Have to give UM credit for going 2-1 @OSU, @Minn and @ Illini for a +.050.
IU blew the big one at home against UW for a -1.0 but made up 0.75 winning at PU and at home against MSU.
MSU is 1-1 in 50/50 games @UW and down 0.25 for the loss @ IU.
Good on you brother. Don't back down now. The tide is starting to turn.
So according to your model all three play an equally difficult schedule even though our one off games are PSU, Iowa, Illinois, and Northwestern. Yeah that makes sense.
Don't over analyze it, brah.
laying in the weeds
Soooo...are we gonna win the B10 or not?
My model actually predicts UM to have the toughest schedule. From the start of the season, IU and MSU would be predicted for 14 wins while UM 13.5.
MSU misses PSU, Iowa, @Illini, @ NW for 3.5 wins
IU misses @UW, NW, Illini, @ Neb for 3.5 wins
UM misses @Neb, @Iowa, Minn, UW for 4.0 wins
Each team would go 17.5-4.5 in a true round robin.
...can't really complain about unbalanced scheduling in this race, as the top teams are all getting two games against each other.
That settles it then. Cancel the games, we win.
Root for IU tonight. UM has a slight upperhand right now but will take a huge lead if they win this game.
...The biggest leap in these assumptions might be that MSU is equivalent in power to IU and UM. MSU needs to split against those teams, not have "bad" losses, and go into PU and OSU at full throttle. IU blew out Purdue at Mackey and UM was impressive at Illinois and at Minnesota.
Much too large of an assumption. You'd need to control for a host of covariates (e.g., days since last game, games in a row on the road, etc.) to generate valid expected probabilities. And the thought that p=1.00 for any game in the B1G this year is laughable.
+1 for the effort, though.
Disagree. All of that stuff would not add any material value and might actually have a negative effect of additional subjectivity. Obvously no win is "guaranteed" as IU found out against UW, Perhaps UW should get a 15% chance and PSU a 7% chance etc.. Perhaps I should do a study of playing a road game on two days rest vs. five days etc...
Consistency is what is important and you need a certain amount of simplicity to most importantly get a high level result that can be understood. You can tell by the reactions that any kind of mathematical reasoning is already far too overwhelming for many in the audience.
As we say in the profession - don't let the model manage you.
247Sports In partnership with CBS Sports