Toggle sidebar

Expected Team Points as of 02/23/26 - The "Everything Everywhere All At Once" Edition

W

Wrestleknownothing

All-Conference
Oct 30, 2021
1,689
4,485
113
Based on Flo rankings here is my attempt to squeeze too much information into a single place.

I am not sure how I feel about it. Feel free to tell me how you feel about it.



Let's start with the obvious.
  • PSU is sitting on 159 points. But as @Psalm 1 guy and I discussed over on HR, PSU has an annoying habit of outperforming the model.
  • Therefore, I have created a PSU-specific adjustment (call it a fudge factor, call it bullcrap). This year it is 25.5 points.
  • That would put PSU at 184.5 points. A record even I could recognize.

The less obvious.
  • At 159 points there is only between a 0.3% - 5% chance anyone catches them. The lower number is the chance anOSU scores over 159. The higher number is the chance PSU scores under 93.7. There are obviously a lot of other scenarios for the ambitious in the crowd to work out.
  • At 184.5 the chances drop to 0.005% - 2%.
  • Given that there is only a 10% spread among Ohio State, Iowa State, Oklahoma State, and Nebraska second through fourth is a jump ball.
  • Iowa - sigh. I fear the rankings have not caught up to their current state. I fear this estimate may be on the high side. There are just so many unknowns about this team. Who are the starters at four (?) weights? How injured are Ferrari and Kueter and what will they look like after being out so long?
If you are a betting man I think you have to take the over on PSU and the under on Iowa. Not sure who you would bet that with, but....
 
  • Like
Reactions: kingstown, namenoonechose, BriantheLion and 8 others
J

Jason1743

All-American
Jan 23, 2006
23,935
6,651
113
Your spreadsheet is impressive, but with my limited wrestling knowledge I think incorrect. For starters PSU has 6 #1 ranked wrestlers and you give them 3.1 champions. Not saying it couldn't happen, but unlikely. Pen State has 9 wrestlers ranked #5 or higher, but you give them 8.4 AAs. Based on recent history Penn State tends to over perform their rankings at nationals, not under perform.
 
  • Like
Reactions: Nitlion1986, BriantheLion and Wrestleknownothing
M

Misalorales

All-Conference
Jun 3, 2025
1,163
3,572
113
Wrestleknownothing said:
Based on Flo rankings here is my attempt to squeeze too much information into a single place.

I am not sure how I feel about it. Feel free to tell me how you feel about it.



Let's start with the obvious.
  • PSU is sitting on 159 points. But as @Psalm 1 guy and I discussed over on HR, PSU has an annoying habit of outperforming the model.
  • Therefore, I have created a PSU-specific adjustment (call it a fudge factor, call it bullcrap). This year it is 25.5 points.
  • That would put PSU at 184.5 points. A record even I could recognize.

The less obvious.
  • At 159 points there is only between a 0.3% - 5% chance anyone catches them. The lower number is the chance anOSU scores over 159. The higher number is the chance PSU scores under 93.7. There are obviously a lot of other scenarios for the ambitious in the crowd to work out.
  • At 184.5 the chances drop to 0.005% - 2%.
  • Given that there is only a 10% spread among Ohio State, Iowa State, Oklahoma State, and Nebraska second through fourth is a jump ball.
  • Iowa - sigh. I fear the rankings have not caught up to their current state. I fear this estimate may be on the high side. There are just so many unknowns about this team. Who are the starters at four (?) weights? How injured are Ferrari and Kueter and what will they look like after being out so long?
If you are a betting man I think you have to take the over on PSU and the under on Iowa. Not sure who you would bet that with, but....
Click to expand...
I know I've talked to you about my new fangled idea on projecting points. I decided I need another season of pissing with the numbers and formula so I'm not putting myself out there yet BUT I have the range somewhere between 176 and 210. Which is precisely why I need another year. That range is absurd.🤣 That said, you know what I'm using, I do think they're going to break that record again because of the in season dual dominance. And I have played with different standard deviations like you advised and other things and gotten it to align with past results much more so than it doesn't.
 
  • Like
  • Love
Reactions: BriantheLion, JoeBagobagels, Col and 1 other person
M

Misalorales

All-Conference
Jun 3, 2025
1,163
3,572
113
Jason1743 said:
Your spreadsheet is impressive, but with my limited wrestling knowledge I think incorrect. For starters PSU has 6 #1 ranked wrestlers and you give them 3.1 champions. Not saying it couldn't happen, but unlikely. Pen State has 9 wrestlers ranked #5 or higher, but you give them 8.4 AAs. Based on recent history Penn State tends to over perform their rankings at nationals, not under perform.
Click to expand...
I greatly appreciate this comment because it's going to elicit a response that I am going to enjoy from an analytics nerd standpoint. 😎
 
  • Like
Reactions: kingstown and Pointingdogsrule
W

Wrestleknownothing

All-Conference
Oct 30, 2021
1,689
4,485
113
Jason1743 said:
Your spreadsheet is impressive, but with my limited wrestling knowledge I think incorrect. For starters PSU has 6 #1 ranked wrestlers and you give them 3.1 champions. Not saying it couldn't happen, but unlikely. Pen State has 9 wrestlers ranked #5 or higher, but you give them 8.4 AAs. Based on recent history Penn State tends to over perform their rankings at nationals, not under perform.
Click to expand...
Let me start by quoting two people - one much more brilliant than I, and one just as dumb.

"But nothing can be everything. Exceptions always exist." - Chuck Klosterman
"All my numbers are wrong. But they are right on average." - WKN

All of these numbers are built off of historical data for the whole field. For example, a #1 seed only wins 47% of the time, hence 3.1 expected champs from the Penn State set of seeds. Though Penn States experience with #1 seeds is much better having won 13 of the last 18 times they were in that position (72%), so it makes sense to take the over there.

I won't bore you with all the details, but if you want to know more here is a link:

Expected Points - Friend or Foe? | Wrestling.Guru

The first thing I did when playing around with historical wrestling data back in 2020 was figure out what the average outcome was by seed. A #1 seed doesn&# ...
wrestling.guru wrestling.guru

These numbers only hold then if everyone performs exactly like the historical average, but that never happens. Which is why I led with those two quotes.

Now as for recent history, Penn State basically performs to seed (-0.1 spots from 2019-2025 and -0.4 spots in 2025 alone), but that means they outperform this model because this model assumes more underperformance at very high seeds than Penn State typically delivers. That is where the adjustment factor above came from. But I only did that at the team score level not at the AA level. So, it makes sense in Penn State's case to lean on the over for AAs too.
 
Last edited:
  • Like
Reactions: kingstown, Cali_Nittany1, Psalm 1 guy and 2 others
Col

Col

Sophomore
Oct 17, 2021
71
133
33
Misalorales said:
I know I've talked to you about my new fangled idea on projecting points. I decided I need another season of pissing with the numbers and formula so I'm not putting myself out there yet BUT I have the range somewhere between 176 and 210. Which is precisely why I need another year. That range is absurd.🤣 That said, you know what I'm using, I do think they're going to break that record again because of the in season dual dominance. And I have played with different standard deviations like you advised and other things and gotten it to align with past results much more so than it doesn't.
Click to expand...
I think you are on the right approach with the standard deviations, but be careful as I am not sure they are constantly the same number year to year. If not it might fit into a standard line or curve (like a exponential or logarithmic curve) which you can use to predict future values. Having said all of that it is probably consistent.... but I would limit the data to the current rules (i.e. 3 pt takedown). This means you don't have enough data points yet to make a good fit. You could use @Wrestleknownothing model that converts the old rules to the new (for the most part) to do a quick sanity check on it. And the math will continue until all are having fun....
 
  • Like
Reactions: Psalm 1 guy and Misalorales
W

Wrestleknownothing

All-Conference
Oct 30, 2021
1,689
4,485
113
Col said:
I think you are on the right approach with the standard deviations, but be careful as I am not sure they are constantly the same number year to year. If not it might fit into a standard line or curve (like a exponential or logarithmic curve) which you can use to predict future values. Having said all of that it is probably consistent.... but I would limit the data to the current rules (i.e. 3 pt takedown). This means you don't have enough data points yet to make a good fit. You could use @Wrestleknownothing model that converts the old rules to the new (for the most part) to do a quick sanity check on it. And the math will continue until all are having fun....
Click to expand...
Yeah, the SD was tricky. The team scores are not normally distrbuted, but what I found is that the deviation between my estimates and the actual scores for the top 20 teams is normally distributed. It breaks down with lower scores.
 
  • Like
Reactions: psumac and Misalorales
Psalm 1 guy

Psalm 1 guy

All-Conference
Nov 3, 2019
1,159
4,436
113
Wrestleknownothing said:
Based on Flo rankings here is my attempt to squeeze too much information into a single place.

I am not sure how I feel about it. Feel free to tell me how you feel about it.



Let's start with the obvious.
  • PSU is sitting on 159 points. But as @Psalm 1 guy and I discussed over on HR, PSU has an annoying habit of outperforming the model.
  • Therefore, I have created a PSU-specific adjustment (call it a fudge factor, call it bullcrap). This year it is 25.5 points.
  • That would put PSU at 184.5 points. A record even I could recognize.

The less obvious.
  • At 159 points there is only between a 0.3% - 5% chance anyone catches them. The lower number is the chance anOSU scores over 159. The higher number is the chance PSU scores under 93.7. There are obviously a lot of other scenarios for the ambitious in the crowd to work out.
  • At 184.5 the chances drop to 0.005% - 2%.
  • Given that there is only a 10% spread among Ohio State, Iowa State, Oklahoma State, and Nebraska second through fourth is a jump ball.
  • Iowa - sigh. I fear the rankings have not caught up to their current state. I fear this estimate may be on the high side. There are just so many unknowns about this team. Who are the starters at four (?) weights? How injured are Ferrari and Kueter and what will they look like after being out so long?
If you are a betting man I think you have to take the over on PSU and the under on Iowa. Not sure who you would bet that with, but....
Click to expand...
Previously on this forum I laid out a very homerish best case scenario of 202.5 points for this season's NCAA tournament, which included a bunch of bonus points. I would be absolutely shocked if that occurred, but I feel much more confident that Penn State will break the official, unofficial or normalized NCAA tournament team scoring record : )
 
Last edited:
  • Like
Reactions: BriantheLion, JoeBagobagels and Wrestleknownothing
P

psumac

Sophomore
Sep 18, 2017
58
128
33
Wrestleknownothing said:
Yeah, the SD was tricky. The team scores are not normally distrbuted, but what I found is that the deviation between my estimates and the actual scores for the top 20 teams is normally distributed. It breaks down with lower scores.
Click to expand...
nothing better than a normally distributed error plot
 
You must log in or register to reply here.
Top Bottom