ADVERTISEMENT

Probability of Winning 2018 Edition

Unbiased_football_fan

Well-Known Member
Aug 18, 2006
77
741
1
Frederick, MD
UFF is back! (Note: referring to oneself in the third person is a sign of awesomeness). Welcome to the pre-tourney simulation discussion.

TLDR: Penn State is favored over Ohio State 54.2% to 46.8% with the field at <0.1%.

Background: I made a probabilistic simulation model a number of years ago (see my posts from previous years) to estimate the probability of winning for the top ten teams. The model uses past data from NCAA tournaments to assign the winning percentages of any seed against any other seed. IT ALSO INCLUDES BONUS POINTS. All brackets including wrestlebacks are simulated. The model was benchmarked (NOT CALIBRATED) against previous years tourneys. Calibration would make it better but it is good enough as is. Usually it gets all of the top five teams in order. Last year OkSt swapped 2 and 3 with Ohio State and VT swapped 5 and 6 with Mizzou. Sometimes teams swap a spot when their predicted scores are pretty close. The model has a nice GUI and is easy to use.

Based on the seeds/brackets that came out you can throw the numbers in a spreadsheet and assume everyone wrestles to seed and not assign bonus points, but why drive a Pinto when you can drive a Ferrari???? (Note: This is not to disparage Pinto’s that I actually think are cool, especially a Pinto with hydraulics that you could make bounce at a stoplight).

This Year: The race this year is tight if you believe the seeds are accurate. It looks like a dead heat between PSU (We Are!) and Ohio St. The probability of PSU winning is 54.2% vs. 46.8% for Ohio St. The field is essentially 0% (<0.1%). Missouri and Michigan have the lead on battling for 3rd and 4th with Iowa, NCSt and Lehigh in a dogfight for 5th through 7th. Predicted mean scores are:

PSU – 136.9
OSU – 134.8
Mizzou – 93.1
Michigan – 78.4
Iowa – 71.7
NCSt – 68.5
Lehigh – 67.8

Now the question becomes do you think the seeds are correct? It doesn’t really matter what the seeding committee assigns, if a guy is good HE WRESTLES LIKE A #1 even if he was given something else. Seeding isn’t an issue for him, seeding is an issue for the guys below him that shouldn’t be seeing him early in the tourney. Who it really screws is the very good guy that the underseeded guy beats that gets knocked into the wrestlebacks and in the blood round meets another guy like himself. Both of those guys would normally AA but one gets knocked out. It is a really bad look for the NCAA to have a coach of one of the top teams on the seeding committee. That would be like putting Saban on the CFB playoff committee. Anyone else think that is completely absurd? Why is this okay for wrestling? Wrestling fans should raise a stink about this, I digress….

I think Nolf will wrestle like a #1 seed. I think Joseph will wrestle like a #2 seed (though his quarter with Lewis could be tough). And I think Pletcher is going to really struggle to achieve 3rd. Put these things together and you get:

PSU – 91% and Ohio St – 9%

Should Cael have MFF’ed Nolf? Yes! See above, if Nolf is healthy it doesn’t matter what he is seeded. You are going to walk out on the mat and face 7 (or more likely 1 or 2) minutes of fury. Although Zain is more fury. Nolf is more like the pitcher that strikes you out on three pitches and you don’t even swing.

Ohio St won the Big Ten tournament therefore they are favored in the NCAA tournament – False! They are different beasts. The field is deeper and you face tougher guys starting in round 2 at NCAA’s. Take 141 for example. McKenna had an easy path to the BT title at 141. That weight is loaded nationally outside the Big Ten. All the credit to him if he can win the NCAA title at 141. Plus the weight management is a much bigger deal. Teams that are “big” and cutting a lot underperform relative to teams that are closer to their natural weights- until the finals. Then they can balloon up. Look for upsets in the quarters and semis more than the finals.

Team X got nine qualifiers therefore they are in great shape – False! Qualifiers are nice for the individual who has worked so hard, but you need thoroughbreds, not ponies, to win the NCAA tourney. Sure you can win with a bunch of scrappers, but it is mathematically MUCH more difficult. You need gatling guns (Retherford, Nolf, Nickal, Hall, Joseph) not calvary.

If you are a wrestler reading this, thank you for competing. You, even if you are 1-23 in the MAC, are an asskicker of an athlete.

If you have other scenarios you would like run post replies. I will check in every couple of days up until the tourney.

UFF
 
UFF is back! (Note: referring to oneself in the third person is a sign of awesomeness). Welcome to the pre-tourney simulation discussion.

TLDR: Penn State is favored over Ohio State 54.2% to 46.8% with the field at <0.1%.

Background: I made a probabilistic simulation model a number of years ago (see my posts from previous years) to estimate the probability of winning for the top ten teams. The model uses past data from NCAA tournaments to assign the winning percentages of any seed against any other seed. IT ALSO INCLUDES BONUS POINTS. All brackets including wrestlebacks are simulated. The model was benchmarked (NOT CALIBRATED) against previous years tourneys. Calibration would make it better but it is good enough as is. Usually it gets all of the top five teams in order. Last year OkSt swapped 2 and 3 with Ohio State and VT swapped 5 and 6 with Mizzou. Sometimes teams swap a spot when their predicted scores are pretty close. The model has a nice GUI and is easy to use.

Based on the seeds/brackets that came out you can throw the numbers in a spreadsheet and assume everyone wrestles to seed and not assign bonus points, but why drive a Pinto when you can drive a Ferrari???? (Note: This is not to disparage Pinto’s that I actually think are cool, especially a Pinto with hydraulics that you could make bounce at a stoplight).

This Year: The race this year is tight if you believe the seeds are accurate. It looks like a dead heat between PSU (We Are!) and Ohio St. The probability of PSU winning is 54.2% vs. 46.8% for Ohio St. The field is essentially 0% (<0.1%). Missouri and Michigan have the lead on battling for 3rd and 4th with Iowa, NCSt and Lehigh in a dogfight for 5th through 7th. Predicted mean scores are:

PSU – 136.9
OSU – 134.8
Mizzou – 93.1
Michigan – 78.4
Iowa – 71.7
NCSt – 68.5
Lehigh – 67.8

Now the question becomes do you think the seeds are correct? It doesn’t really matter what the seeding committee assigns, if a guy is good HE WRESTLES LIKE A #1 even if he was given something else. Seeding isn’t an issue for him, seeding is an issue for the guys below him that shouldn’t be seeing him early in the tourney. Who it really screws is the very good guy that the underseeded guy beats that gets knocked into the wrestlebacks and in the blood round meets another guy like himself. Both of those guys would normally AA but one gets knocked out. It is a really bad look for the NCAA to have a coach of one of the top teams on the seeding committee. That would be like putting Saban on the CFB playoff committee. Anyone else think that is completely absurd? Why is this okay for wrestling? Wrestling fans should raise a stink about this, I digress….

I think Nolf will wrestle like a #1 seed. I think Joseph will wrestle like a #2 seed (though his quarter with Lewis could be tough). And I think Pletcher is going to really struggle to achieve 3rd. Put these things together and you get:

PSU – 91% and Ohio St – 9%

Should Cael have MFF’ed Nolf? Yes! See above, if Nolf is healthy it doesn’t matter what he is seeded. You are going to walk out on the mat and face 7 (or more likely 1 or 2) minutes of fury. Although Zain is more fury. Nolf is more like the pitcher that strikes you out on three pitches and you don’t even swing.

Ohio St won the Big Ten tournament therefore they are favored in the NCAA tournament – False! They are different beasts. The field is deeper and you face tougher guys starting in round 2 at NCAA’s. Take 141 for example. McKenna had an easy path to the BT title at 141. That weight is loaded nationally outside the Big Ten. All the credit to him if he can win the NCAA title at 141. Plus the weight management is a much bigger deal. Teams that are “big” and cutting a lot underperform relative to teams that are closer to their natural weights- until the finals. Then they can balloon up. Look for upsets in the quarters and semis more than the finals.

Team X got nine qualifiers therefore they are in great shape – False! Qualifiers are nice for the individual who has worked so hard, but you need thoroughbreds, not ponies, to win the NCAA tourney. Sure you can win with a bunch of scrappers, but it is mathematically MUCH more difficult. You need gatling guns (Retherford, Nolf, Nickal, Hall, Joseph) not calvary.

If you are a wrestler reading this, thank you for competing. You, even if you are 1-23 in the MAC, are an asskicker of an athlete.

If you have other scenarios you would like run post replies. I will check in every couple of days up until the tourney.

UFF

This sounds awesome. Are you able to choose specific placements and reevaluate the model from there. For example, can you assume NaTo wins 125, then back into the probabilities of PSU/OSU winning the team race with everything else random?

Also, this may be a lot to ask but is the model available to the public to play around with?
 
A couple points you made that I think are key.

The Big-10 is relatively weak at 3 weight classes ... 133, 141 and and 197. To their credit, Pletcher, McKenna and Moore took advantage by marching to the finals without facing a top-15 opponent. They will face tougher competition starting with the quarters.

The other point is weight management at a 3 day tourney. We've seen it before; Ok St the last 2 years, Minny in 2014 and 2013 come to mind. Teams that are cutting a lot of weight sometimes meet their Waterloo on Friday.
 
  • Like
Reactions: tikk10 and Lyons212
UFF is back! (Note: referring to oneself in the third person is a sign of awesomeness). Welcome to the pre-tourney simulation discussion.

TLDR: Penn State is favored over Ohio State 54.2% to 46.8% with the field at <0.1%.

Background: I made a probabilistic simulation model a number of years ago (see my posts from previous years) to estimate the probability of winning for the top ten teams. The model uses past data from NCAA tournaments to assign the winning percentages of any seed against any other seed. IT ALSO INCLUDES BONUS POINTS. All brackets including wrestlebacks are simulated. The model was benchmarked (NOT CALIBRATED) against previous years tourneys. Calibration would make it better but it is good enough as is. Usually it gets all of the top five teams in order. Last year OkSt swapped 2 and 3 with Ohio State and VT swapped 5 and 6 with Mizzou. Sometimes teams swap a spot when their predicted scores are pretty close. The model has a nice GUI and is easy to use.

Based on the seeds/brackets that came out you can throw the numbers in a spreadsheet and assume everyone wrestles to seed and not assign bonus points, but why drive a Pinto when you can drive a Ferrari???? (Note: This is not to disparage Pinto’s that I actually think are cool, especially a Pinto with hydraulics that you could make bounce at a stoplight).

This Year: The race this year is tight if you believe the seeds are accurate. It looks like a dead heat between PSU (We Are!) and Ohio St. The probability of PSU winning is 54.2% vs. 46.8% for Ohio St. The field is essentially 0% (<0.1%). Missouri and Michigan have the lead on battling for 3rd and 4th with Iowa, NCSt and Lehigh in a dogfight for 5th through 7th. Predicted mean scores are:

PSU – 136.9
OSU – 134.8
Mizzou – 93.1
Michigan – 78.4
Iowa – 71.7
NCSt – 68.5
Lehigh – 67.8

Now the question becomes do you think the seeds are correct? It doesn’t really matter what the seeding committee assigns, if a guy is good HE WRESTLES LIKE A #1 even if he was given something else. Seeding isn’t an issue for him, seeding is an issue for the guys below him that shouldn’t be seeing him early in the tourney. Who it really screws is the very good guy that the underseeded guy beats that gets knocked into the wrestlebacks and in the blood round meets another guy like himself. Both of those guys would normally AA but one gets knocked out. It is a really bad look for the NCAA to have a coach of one of the top teams on the seeding committee. That would be like putting Saban on the CFB playoff committee. Anyone else think that is completely absurd? Why is this okay for wrestling? Wrestling fans should raise a stink about this, I digress….

I think Nolf will wrestle like a #1 seed. I think Joseph will wrestle like a #2 seed (though his quarter with Lewis could be tough). And I think Pletcher is going to really struggle to achieve 3rd. Put these things together and you get:

PSU – 91% and Ohio St – 9%

Should Cael have MFF’ed Nolf? Yes! See above, if Nolf is healthy it doesn’t matter what he is seeded. You are going to walk out on the mat and face 7 (or more likely 1 or 2) minutes of fury. Although Zain is more fury. Nolf is more like the pitcher that strikes you out on three pitches and you don’t even swing.

Ohio St won the Big Ten tournament therefore they are favored in the NCAA tournament – False! They are different beasts. The field is deeper and you face tougher guys starting in round 2 at NCAA’s. Take 141 for example. McKenna had an easy path to the BT title at 141. That weight is loaded nationally outside the Big Ten. All the credit to him if he can win the NCAA title at 141. Plus the weight management is a much bigger deal. Teams that are “big” and cutting a lot underperform relative to teams that are closer to their natural weights- until the finals. Then they can balloon up. Look for upsets in the quarters and semis more than the finals.

Team X got nine qualifiers therefore they are in great shape – False! Qualifiers are nice for the individual who has worked so hard, but you need thoroughbreds, not ponies, to win the NCAA tourney. Sure you can win with a bunch of scrappers, but it is mathematically MUCH more difficult. You need gatling guns (Retherford, Nolf, Nickal, Hall, Joseph) not calvary.

If you are a wrestler reading this, thank you for competing. You, even if you are 1-23 in the MAC, are an asskicker of an athlete.

If you have other scenarios you would like run post replies. I will check in every couple of days up until the tourney.

UFF
Totally awesome. Looking at the board yesterday, we had our yearly orange harvest occur. Oranges all over the place. Nice to bring a bit of math to the game. Would love to give you 1000 likes. Can you also mathematically verify that Tioga = 21guns? Certainly has that feel about it
 
Totally awesome. Looking at the board yesterday, we had our yearly orange harvest occur. Oranges all over the place. Nice to bring a bit of math to the game. Would love to give you 1000 likes. Can you also mathematically verify that Tioga = 21guns? Certainly has that feel about it
Not doubt re: Tioga - the same irrational absolutes.
 
UFF is back! (Note: referring to oneself in the third person is a sign of awesomeness). Welcome to the pre-tourney simulation discussion.

TLDR: Penn State is favored over Ohio State 54.2% to 46.8% with the field at <0.1%.

Background: I made a probabilistic simulation model a number of years ago (see my posts from previous years) to estimate the probability of winning for the top ten teams. The model uses past data from NCAA tournaments to assign the winning percentages of any seed against any other seed. IT ALSO INCLUDES BONUS POINTS. All brackets including wrestlebacks are simulated. The model was benchmarked (NOT CALIBRATED) against previous years tourneys. Calibration would make it better but it is good enough as is. Usually it gets all of the top five teams in order. Last year OkSt swapped 2 and 3 with Ohio State and VT swapped 5 and 6 with Mizzou. Sometimes teams swap a spot when their predicted scores are pretty close. The model has a nice GUI and is easy to use.

Based on the seeds/brackets that came out you can throw the numbers in a spreadsheet and assume everyone wrestles to seed and not assign bonus points, but why drive a Pinto when you can drive a Ferrari???? (Note: This is not to disparage Pinto’s that I actually think are cool, especially a Pinto with hydraulics that you could make bounce at a stoplight).

This Year: The race this year is tight if you believe the seeds are accurate. It looks like a dead heat between PSU (We Are!) and Ohio St. The probability of PSU winning is 54.2% vs. 46.8% for Ohio St. The field is essentially 0% (<0.1%). Missouri and Michigan have the lead on battling for 3rd and 4th with Iowa, NCSt and Lehigh in a dogfight for 5th through 7th. Predicted mean scores are:

PSU – 136.9
OSU – 134.8
Mizzou – 93.1
Michigan – 78.4
Iowa – 71.7
NCSt – 68.5
Lehigh – 67.8

Now the question becomes do you think the seeds are correct? It doesn’t really matter what the seeding committee assigns, if a guy is good HE WRESTLES LIKE A #1 even if he was given something else. Seeding isn’t an issue for him, seeding is an issue for the guys below him that shouldn’t be seeing him early in the tourney. Who it really screws is the very good guy that the underseeded guy beats that gets knocked into the wrestlebacks and in the blood round meets another guy like himself. Both of those guys would normally AA but one gets knocked out. It is a really bad look for the NCAA to have a coach of one of the top teams on the seeding committee. That would be like putting Saban on the CFB playoff committee. Anyone else think that is completely absurd? Why is this okay for wrestling? Wrestling fans should raise a stink about this, I digress….

I think Nolf will wrestle like a #1 seed. I think Joseph will wrestle like a #2 seed (though his quarter with Lewis could be tough). And I think Pletcher is going to really struggle to achieve 3rd. Put these things together and you get:

PSU – 91% and Ohio St – 9%

Should Cael have MFF’ed Nolf? Yes! See above, if Nolf is healthy it doesn’t matter what he is seeded. You are going to walk out on the mat and face 7 (or more likely 1 or 2) minutes of fury. Although Zain is more fury. Nolf is more like the pitcher that strikes you out on three pitches and you don’t even swing.

Ohio St won the Big Ten tournament therefore they are favored in the NCAA tournament – False! They are different beasts. The field is deeper and you face tougher guys starting in round 2 at NCAA’s. Take 141 for example. McKenna had an easy path to the BT title at 141. That weight is loaded nationally outside the Big Ten. All the credit to him if he can win the NCAA title at 141. Plus the weight management is a much bigger deal. Teams that are “big” and cutting a lot underperform relative to teams that are closer to their natural weights- until the finals. Then they can balloon up. Look for upsets in the quarters and semis more than the finals.

Team X got nine qualifiers therefore they are in great shape – False! Qualifiers are nice for the individual who has worked so hard, but you need thoroughbreds, not ponies, to win the NCAA tourney. Sure you can win with a bunch of scrappers, but it is mathematically MUCH more difficult. You need gatling guns (Retherford, Nolf, Nickal, Hall, Joseph) not calvary.

If you are a wrestler reading this, thank you for competing. You, even if you are 1-23 in the MAC, are an asskicker of an athlete.

If you have other scenarios you would like run post replies. I will check in every couple of days up until the tourney.

UFF
my favorite post ever.
 
This sounds awesome. Are you able to choose specific placements and reevaluate the model from there. For example, can you assume NaTo wins 125, then back into the probabilities of PSU/OSU winning the team race with everything else random?

Also, this may be a lot to ask but is the model available to the public to play around with?

You could fix certain outcomes and simulate the rest, though it isn't currently set up that way. For instance you could fix Nato's finish at #1, Zain at #1, coin flip Snyder/Coon, and simulate the rest. That is a bit more complicated if you want to provide that flexibility but it is dooable. Let me know if there is something in particular you want to see and I will see if I have time tonight or this weekend.

With the software I use it could be made available for anyone to use for free, I just haven't done it because I sunk a fair number of PhD-hrs into it. Of course then I have to lecture myself rather than my 11 year old about sunk costs...but I think it could have some interesting applications:

- allocation of scholarship money (unless you have a giant slush fund)
- redshirt decisions
- entertainment
- odds making

This last one would be best for the sport I think. I see all the time on here where wrestling fans are so passionate they will put their money where their mouth is. I just don't know who to talk to about that one, so if any of you have ideas PM me.

As a general rule of thumb, listen to 21Guns about max bench and listen to 12Guns about probability.
 
That is because the Brands brothers get team points deducted and given to PSU or Ohio State :)

About one percent of the time there is a tie in the team score so they are both given the "win". Normally this doesn't show up but since in the base simulation they are pretty close it adds up to 1% extra in the win probabilities. I wasn't sure of the tiebreaker and if it was worth adding that in. While pretty good, we aren't dealing with predictions to three significant digits here even if I gave them.

Anyone know what the tiebreaker is if two teams tie in overall score? Do they play dodgeball as the tiebreaker?
 
UFF is back! (Note: referring to oneself in the third person is a sign of awesomeness). Welcome to the pre-tourney simulation discussion.

TLDR: Penn State is favored over Ohio State 54.2% to 46.8% with the field at <0.1%.

Background: I made a probabilistic simulation model a number of years ago (see my posts from previous years) to estimate the probability of winning for the top ten teams. The model uses past data from NCAA tournaments to assign the winning percentages of any seed against any other seed. IT ALSO INCLUDES BONUS POINTS. All brackets including wrestlebacks are simulated. The model was benchmarked (NOT CALIBRATED) against previous years tourneys. Calibration would make it better but it is good enough as is. Usually it gets all of the top five teams in order. Last year OkSt swapped 2 and 3 with Ohio State and VT swapped 5 and 6 with Mizzou. Sometimes teams swap a spot when their predicted scores are pretty close. The model has a nice GUI and is easy to use.

Based on the seeds/brackets that came out you can throw the numbers in a spreadsheet and assume everyone wrestles to seed and not assign bonus points, but why drive a Pinto when you can drive a Ferrari???? (Note: This is not to disparage Pinto’s that I actually think are cool, especially a Pinto with hydraulics that you could make bounce at a stoplight).

This Year: The race this year is tight if you believe the seeds are accurate. It looks like a dead heat between PSU (We Are!) and Ohio St. The probability of PSU winning is 54.2% vs. 46.8% for Ohio St. The field is essentially 0% (<0.1%). Missouri and Michigan have the lead on battling for 3rd and 4th with Iowa, NCSt and Lehigh in a dogfight for 5th through 7th. Predicted mean scores are:

PSU – 136.9
OSU – 134.8
Mizzou – 93.1
Michigan – 78.4
Iowa – 71.7
NCSt – 68.5
Lehigh – 67.8

Now the question becomes do you think the seeds are correct? It doesn’t really matter what the seeding committee assigns, if a guy is good HE WRESTLES LIKE A #1 even if he was given something else. Seeding isn’t an issue for him, seeding is an issue for the guys below him that shouldn’t be seeing him early in the tourney. Who it really screws is the very good guy that the underseeded guy beats that gets knocked into the wrestlebacks and in the blood round meets another guy like himself. Both of those guys would normally AA but one gets knocked out. It is a really bad look for the NCAA to have a coach of one of the top teams on the seeding committee. That would be like putting Saban on the CFB playoff committee. Anyone else think that is completely absurd? Why is this okay for wrestling? Wrestling fans should raise a stink about this, I digress….

I think Nolf will wrestle like a #1 seed. I think Joseph will wrestle like a #2 seed (though his quarter with Lewis could be tough). And I think Pletcher is going to really struggle to achieve 3rd. Put these things together and you get:

PSU – 91% and Ohio St – 9%

Should Cael have MFF’ed Nolf? Yes! See above, if Nolf is healthy it doesn’t matter what he is seeded. You are going to walk out on the mat and face 7 (or more likely 1 or 2) minutes of fury. Although Zain is more fury. Nolf is more like the pitcher that strikes you out on three pitches and you don’t even swing.

Ohio St won the Big Ten tournament therefore they are favored in the NCAA tournament – False! They are different beasts. The field is deeper and you face tougher guys starting in round 2 at NCAA’s. Take 141 for example. McKenna had an easy path to the BT title at 141. That weight is loaded nationally outside the Big Ten. All the credit to him if he can win the NCAA title at 141. Plus the weight management is a much bigger deal. Teams that are “big” and cutting a lot underperform relative to teams that are closer to their natural weights- until the finals. Then they can balloon up. Look for upsets in the quarters and semis more than the finals.

Team X got nine qualifiers therefore they are in great shape – False! Qualifiers are nice for the individual who has worked so hard, but you need thoroughbreds, not ponies, to win the NCAA tourney. Sure you can win with a bunch of scrappers, but it is mathematically MUCH more difficult. You need gatling guns (Retherford, Nolf, Nickal, Hall, Joseph) not calvary.

If you are a wrestler reading this, thank you for competing. You, even if you are 1-23 in the MAC, are an asskicker of an athlete.

If you have other scenarios you would like run post replies. I will check in every couple of days up until the tourney.

UFF

Ummmmmm.. Who are you and why is this only your 8th post? Could you please run my life through this model and tell me exactly how much money I should contribute to my 401k so I can afford to go to NCAA's every year of retirement, assuming I die at age 75? Thanks
 
This guy has to be the love child of Roar and some MIT co-ed who was trying to analyze some weight cutting algorithms for the purposes of curing obesity in the US. She got tipsy while in the lab breaking down the nutritional make-up of red wine and confusing her red bull for Merlot. Roar (looking for a mens room between scouting New England high school district championships) opened the wrong door, was intrigued by the analysis, then swooped on in with his charm...
 
UFF is back! (Note: referring to oneself in the third person is a sign of awesomeness). Welcome to the pre-tourney simulation discussion.

TLDR: Penn State is favored over Ohio State 54.2% to 46.8% with the field at <0.1%.

Background: I made a probabilistic simulation model a number of years ago (see my posts from previous years) to estimate the probability of winning for the top ten teams. The model uses past data from NCAA tournaments to assign the winning percentages of any seed against any other seed. IT ALSO INCLUDES BONUS POINTS. All brackets including wrestlebacks are simulated. The model was benchmarked (NOT CALIBRATED) against previous years tourneys. Calibration would make it better but it is good enough as is. Usually it gets all of the top five teams in order. Last year OkSt swapped 2 and 3 with Ohio State and VT swapped 5 and 6 with Mizzou. Sometimes teams swap a spot when their predicted scores are pretty close. The model has a nice GUI and is easy to use.

Based on the seeds/brackets that came out you can throw the numbers in a spreadsheet and assume everyone wrestles to seed and not assign bonus points, but why drive a Pinto when you can drive a Ferrari???? (Note: This is not to disparage Pinto’s that I actually think are cool, especially a Pinto with hydraulics that you could make bounce at a stoplight).

This Year: The race this year is tight if you believe the seeds are accurate. It looks like a dead heat between PSU (We Are!) and Ohio St. The probability of PSU winning is 54.2% vs. 46.8% for Ohio St. The field is essentially 0% (<0.1%). Missouri and Michigan have the lead on battling for 3rd and 4th with Iowa, NCSt and Lehigh in a dogfight for 5th through 7th. Predicted mean scores are:

PSU – 136.9
OSU – 134.8
Mizzou – 93.1
Michigan – 78.4
Iowa – 71.7
NCSt – 68.5
Lehigh – 67.8

Now the question becomes do you think the seeds are correct? It doesn’t really matter what the seeding committee assigns, if a guy is good HE WRESTLES LIKE A #1 even if he was given something else. Seeding isn’t an issue for him, seeding is an issue for the guys below him that shouldn’t be seeing him early in the tourney. Who it really screws is the very good guy that the underseeded guy beats that gets knocked into the wrestlebacks and in the blood round meets another guy like himself. Both of those guys would normally AA but one gets knocked out. It is a really bad look for the NCAA to have a coach of one of the top teams on the seeding committee. That would be like putting Saban on the CFB playoff committee. Anyone else think that is completely absurd? Why is this okay for wrestling? Wrestling fans should raise a stink about this, I digress….

I think Nolf will wrestle like a #1 seed. I think Joseph will wrestle like a #2 seed (though his quarter with Lewis could be tough). And I think Pletcher is going to really struggle to achieve 3rd. Put these things together and you get:

PSU – 91% and Ohio St – 9%

Should Cael have MFF’ed Nolf? Yes! See above, if Nolf is healthy it doesn’t matter what he is seeded. You are going to walk out on the mat and face 7 (or more likely 1 or 2) minutes of fury. Although Zain is more fury. Nolf is more like the pitcher that strikes you out on three pitches and you don’t even swing.

Ohio St won the Big Ten tournament therefore they are favored in the NCAA tournament – False! They are different beasts. The field is deeper and you face tougher guys starting in round 2 at NCAA’s. Take 141 for example. McKenna had an easy path to the BT title at 141. That weight is loaded nationally outside the Big Ten. All the credit to him if he can win the NCAA title at 141. Plus the weight management is a much bigger deal. Teams that are “big” and cutting a lot underperform relative to teams that are closer to their natural weights- until the finals. Then they can balloon up. Look for upsets in the quarters and semis more than the finals.

Team X got nine qualifiers therefore they are in great shape – False! Qualifiers are nice for the individual who has worked so hard, but you need thoroughbreds, not ponies, to win the NCAA tourney. Sure you can win with a bunch of scrappers, but it is mathematically MUCH more difficult. You need gatling guns (Retherford, Nolf, Nickal, Hall, Joseph) not calvary.

If you are a wrestler reading this, thank you for competing. You, even if you are 1-23 in the MAC, are an asskicker of an athlete.

If you have other scenarios you would like run post replies. I will check in every couple of days up until the tourney.

UFF
So....turns out that I’m not only a loser at life but also on this forum...


‘Preciate the reminder brah...o_O
 
  • Like
Reactions: Dogwelder
I'm blocking someone. Not sure who, yet, but someone. The reason being that posts like UFF's highlight that reading some of the other drivel on here is an utter waste of time, and might even kill brain sells. Unfortunately, UFF's posting frequency isn't high enough to make up for the loses.

New, fake Spyker, you'd better wach out.
 
UFF is back! (Note: referring to oneself in the third person is a sign of awesomeness). Welcome to the pre-tourney simulation discussion.

TLDR: Penn State is favored over Ohio State 54.2% to 46.8% with the field at <0.1%.

Background: I made a probabilistic simulation model a number of years ago (see my posts from previous years) to estimate the probability of winning for the top ten teams. The model uses past data from NCAA tournaments to assign the winning percentages of any seed against any other seed. IT ALSO INCLUDES BONUS POINTS. All brackets including wrestlebacks are simulated. The model was benchmarked (NOT CALIBRATED) against previous years tourneys. Calibration would make it better but it is good enough as is. Usually it gets all of the top five teams in order. Last year OkSt swapped 2 and 3 with Ohio State and VT swapped 5 and 6 with Mizzou. Sometimes teams swap a spot when their predicted scores are pretty close. The model has a nice GUI and is easy to use.

Based on the seeds/brackets that came out you can throw the numbers in a spreadsheet and assume everyone wrestles to seed and not assign bonus points, but why drive a Pinto when you can drive a Ferrari???? (Note: This is not to disparage Pinto’s that I actually think are cool, especially a Pinto with hydraulics that you could make bounce at a stoplight).

This Year: The race this year is tight if you believe the seeds are accurate. It looks like a dead heat between PSU (We Are!) and Ohio St. The probability of PSU winning is 54.2% vs. 46.8% for Ohio St. The field is essentially 0% (<0.1%). Missouri and Michigan have the lead on battling for 3rd and 4th with Iowa, NCSt and Lehigh in a dogfight for 5th through 7th. Predicted mean scores are:

PSU – 136.9
OSU – 134.8
Mizzou – 93.1
Michigan – 78.4
Iowa – 71.7
NCSt – 68.5
Lehigh – 67.8

Now the question becomes do you think the seeds are correct? It doesn’t really matter what the seeding committee assigns, if a guy is good HE WRESTLES LIKE A #1 even if he was given something else. Seeding isn’t an issue for him, seeding is an issue for the guys below him that shouldn’t be seeing him early in the tourney. Who it really screws is the very good guy that the underseeded guy beats that gets knocked into the wrestlebacks and in the blood round meets another guy like himself. Both of those guys would normally AA but one gets knocked out. It is a really bad look for the NCAA to have a coach of one of the top teams on the seeding committee. That would be like putting Saban on the CFB playoff committee. Anyone else think that is completely absurd? Why is this okay for wrestling? Wrestling fans should raise a stink about this, I digress….

I think Nolf will wrestle like a #1 seed. I think Joseph will wrestle like a #2 seed (though his quarter with Lewis could be tough). And I think Pletcher is going to really struggle to achieve 3rd. Put these things together and you get:

PSU – 91% and Ohio St – 9%

Should Cael have MFF’ed Nolf? Yes! See above, if Nolf is healthy it doesn’t matter what he is seeded. You are going to walk out on the mat and face 7 (or more likely 1 or 2) minutes of fury. Although Zain is more fury. Nolf is more like the pitcher that strikes you out on three pitches and you don’t even swing.

Ohio St won the Big Ten tournament therefore they are favored in the NCAA tournament – False! They are different beasts. The field is deeper and you face tougher guys starting in round 2 at NCAA’s. Take 141 for example. McKenna had an easy path to the BT title at 141. That weight is loaded nationally outside the Big Ten. All the credit to him if he can win the NCAA title at 141. Plus the weight management is a much bigger deal. Teams that are “big” and cutting a lot underperform relative to teams that are closer to their natural weights- until the finals. Then they can balloon up. Look for upsets in the quarters and semis more than the finals.

Team X got nine qualifiers therefore they are in great shape – False! Qualifiers are nice for the individual who has worked so hard, but you need thoroughbreds, not ponies, to win the NCAA tourney. Sure you can win with a bunch of scrappers, but it is mathematically MUCH more difficult. You need gatling guns (Retherford, Nolf, Nickal, Hall, Joseph) not calvary.

If you are a wrestler reading this, thank you for competing. You, even if you are 1-23 in the MAC, are an asskicker of an athlete.

If you have other scenarios you would like run post replies. I will check in every couple of days up until the tourney.

UFF


Can i take you to Vegas??
 
Best Vegas story for a dork: Of course you all are probably aware they play Keno in Vegas. Keno is lottery type game with numbers generated randomly. Well back in the early days the random number generation algorithms weren't as complicated as today. Some smart guys got the idea to use the string of past results to determine the seed for the algorithm. Placed a big bet on the next numbers and of course won. Plan fell through when the casino manager came to the room to congratulate them and found the room full of computer equipment.

I forgot this one from my initial post:

Cael can't coach, all he does it get great recruits that is why Penn State does so well. False! Here are the point differentials from predicted vs. actual using WrestleSim from 2010-2017 for select schools at the NCAA tournament :

PSU +7.2
Ohio St +3.4
Minnesota +2.7
Cornell -0.2
OkSt -4.8
Iowa -5.4

And, it isn't because Penn State has had a couple fantastic years (like last year). Penn State is the only school to be higher all eight of those years. There are some people that haven't done a great job preparing their teams but it isn't Cael.
 
I'm blocking someone. Not sure who, yet, but someone. The reason being that posts like UFF's highlight that reading some of the other drivel on here is an utter waste of time, and might even kill brain sells. Unfortunately, UFF's posting frequency isn't high enough to make up for the loses.

New, fake Spyker, you'd better wach out.
Might? Looks like it's already happening.
 
That was fun! We are so blessed to be Penn State wrestling fans. For those of you that are interested, here is the final comparison between predicted and reality:

Team Base Model Actual
PSU 136.9 141.5
OSU 134.8 133.5
Mizz 93.1 66.5 (Woof! Holy overseeded Batman)
Mich 78.4 80
Iowa 71.7 97 (Spencer Lee is awesome, loved the judo story)
NCSt 68.5 80
Lehigh 67.5 36.5 (Ouch! Wrestled like their uniform colors)

Adjusted model (with Nolf winning) was even better than the base model. I will send out some early 2019 predictions in the upcoming weeks once I get some time to look at redshirts and incoming freshmen.

Peace,

UFF
 
ADVERTISEMENT
ADVERTISEMENT