Big 12 Champion Computer Projection

[erikbj]

I still think KSU is going to be a tough team. The schedule is in our favor - but would be nice if Wade had a set back for a week or two.

Winning the Big12 would be a boost to lock up DM.

 

[ChickenNuggetMan]

If we’re ever gonna win this thing, the time is NOW.

 

[CascadeClone]

With all that said, it seems like how teams do against the lower half that makes the difference

This. The teams that stumble at OU, or Texas, or even TCU, will drop out. KU has always been the king of not stumbling.

 

[CySmurf]

I think the champion will come from the state of Kansas. I think K State is the best team in the conference. I think either them or KU will win the league.

It's not Kansas. K-State looks like the best team at this point. However, so much can change. Just keep winning ISU. Everything else will take care of itself.

 

[cyclone101]

 

[Prone2Clone]

Baylor and K-State must be destroyed.

Man, that sounds weird.

 

[VeloClone]

It's not Kansas. K-State looks like the best team at this point. However, so much can change. Just keep winning ISU. Everything else will take care of itself.

Literally - since ISU still has game(s) against each of the teams one game ahead of them in the loss column ISU controls its own destiny again.

 

[BillBrasky4Cy]

K-State who struggles to score still has KU twice, Baylor twice, @TCU, and @WVU. I can see purple Kansas falling back to 4th.

 

[surly]

K-State who struggles to score still has KU twice, Baylor twice, @TCU, and @WVU. I can see purple Kansas falling back to 4th.

Schedule does them no favors, for sure.

 

[TedKumsher]

Basically, this is 100,000 Monte Carlo simulations of the rest of the Big 12 basketball season in Python, using Ken Massey’s rating system for the odds for every remaining game.

Nerd question.
Why run 100,000 simulations? Shouldn't it just be statistical math based on odds?

 

[RonBurgundy]

Nerd question.
Why run 100,000 simulations? Shouldn't it just be statistical math based on odds?

While there are probabilities, there are also standard deviations around those probabilities, which differ for every team. So a Monte Carlo simulation can be used to evaluate the overall expected results. And with today's computers, would take only a minute to run once all the inputs are in.

 

[Dale]

Nerd question.
Why run 100,000 simulations? Shouldn't it just be statistical math based on odds?

Great question. Yes, the ideal solution would be to work out the odds of every possible W/L combination scenario in the remaining (at this point) 53 conference games. But that means you'd have have to figure out the odds of 2^53 ≈ 9 quadrillion different scenarios, which—back of the envelope estimate—would take my computer a few centuries to finish. So the Monte Carlo simulations give you a reasonable estimate (within a couple tenths of a percent) of the real answer in two minutes or so.

When it gets closer to the end of the season, it becomes a lot more reasonable to do it the "right way"...but still, the extra accuracy it would give over 100,000 Monte Carlo simulations is very minimal.

 

[fsanford]

ISU probably has the easiest road because they only have 3 games against the other contending teams Baylor, Tech, KU, KSU and 2 of them are at home.
Tech has 4 and the other 3 have 5 games.

KSU, KU and ISU have beating another contender on their home court.
All contenders but KU have lost at home.

Whoever finish 5-4 on the road probably wins this thing. KU might be able to slip by with 4-5

KU is what 1-3 with games at Tech, KSU, and TCU still to go.. This might be the year they go down

 

[huntt26]

I think the Cyclones have a real good shot of at least a share of the conference title IF they stay healthy and no suspensions.

 

[Cyclonepride]

Schedule does them no favors, for sure.

Everyone plays the same schedule.

 

[Cyclonepride]

Great question. Yes, the ideal solution would be to work out the odds of every possible W/L combination scenario in the remaining (at this point) 53 conference games. But that means you'd have have to figure out the odds of 2^53 ≈ 9 quadrillion different scenarios, which—back of the envelope estimate—would take my computer a few centuries to finish. So the Monte Carlo simulations give you a reasonable estimate (within a couple tenths of a percent) of the real answer in two minutes or so.

When it gets closer to the end of the season, it becomes a lot more reasonable to do it the "right way"...but still, the extra accuracy it would give over 100,000 Monte Carlo simulations is very minimal.

If that makes sure it's right, then do it. Get back to us when it's complete.

 

[NoCreativity]

Everyone plays the same schedule.

I assume he's smart enough to realize that and I assume he meant remaining schedule. K-State actually has the 6th easiest schedule left so they don't have a gauntlet by any means so Im not buying that excuse.

Of course the teams at the top are usually going to have an easier schedule left because they don't have to play themselves twice. Iowa State has the 8th easiest left.

 

[Cyclonepride]

I assume he's smart enough to realize that and I assume he meant remaining schedule. K-State actually has the 6th easiest schedule left so they don't have a gauntlet by any means so Im not buying that excuse.

Of course the teams at the top are usually going to have an easier schedule left because they don't have to play themselves twice. Iowa State has the 8th easiest left.

I'm sure he did, though it's silly to complain about in a round robin.

 

[ClonesFTW]

Nerd question.
Why run 100,000 simulations? Shouldn't it just be statistical math based on odds?

My buddy flipped a quarter 1,200 times in a dorm room and recorded the results on a pizza box to prove a point in stat class once; sometimes statistical math is based on Admiral Nelson.

 

[VeloClone]

ISU probably has the easiest road because they only have 3 games against the other contending teams Baylor, Tech, KU, KSU and 2 of them are at home.
Tech has 4 and the other 3 have 5 games.

KSU, KU and ISU have beating another contender on their home court.
All contenders but KU have lost at home.

Whoever finish 5-4 on the road probably wins this thing. KU might be able to slip by with 4-5

KU is what 1-3 with games at Tech, KSU, and TCU still to go.. This might be the year they go down

To take it to the next level here is what each team within 2 games of the lead have left:

BU 5-2 (.500 remaining games teams' combined winning percentage)
5 road - 4 winning record, 1 @ .500 (.605 remaining road game teams' combined record)
6 home - 1 winning, 1 @ .500, 4 losing record (.409)

KSU 5-2 (.481)
6 road - 2 winning, 1 @ .500, 3 losing (.455)
5 home - 3 winning, 2 losing (.514)

TTU 5-3 (.459)
5 road - 2 winning, 3 losing (.459)
5 home - 2 winning, 1 @ .500, 2 losing (.459)

KU 5-3 (.451)
5 road - 2 winning, 3 losing (.486)
5 home - 3 winning, 2 losing (.500)

ISU 4-3 (.451)
5 road - 1 winning, 1 @ .500, 3 losing (.432)
6 home - 2 winning, 1 @ .500, 3 losing (.467)

UT 4-4 (.514)
5 road - 3 winning, 2 losing (.486)
5 home - 3 winning, 2 losing (.543)

TCU 3-4 (.494)
5 road - 2 winning, 1 @ .500, 2 losing (.444)
6 home - 4 winning, 2 losing (.533)