Jump to content

Recommended Posts

Posted
2 minutes ago, BruceyB said:

There isn't much that is technical about ranking. You take into effect results of what happens and adjust accordingly. If you were going to provide preseason rankings that had to stand-up to the results of the season, you might have a point.. but ranking a group of athletes based on the previous seasons results and than adjusting them based on what happens really isn't that difficult.

I start with this years NCAA tournament qualifiers and how they performed and make an initial 1-33.. guys who beat them/lose to outsiders get adjusted accordingly. I'm not sure how this is supposed to be so hard to do?

You do have to take into account the freshman - some are more of a known quantity than others.

Then you have to weigh tournament results based not only on how wrestlers placed, but on who they competed against to get to placement.

Then you also have to consider the overall records of wrestlers... except that's tricky when some teams decide to throw in more powder puff early season matches than others. Against wrestlers that are often unknown quantities.

Then you have to acknowledge the 'close matches' that went into OT/SV and aren't as decisive as they might appear.

This all for 77 teams, with 10 starters each - which being a starter or not is sometimes not known until the bitter end - and potential injuries. So figure 12 wrestlers to follow for each team. So that's 900+ wrestlers to watch over the course of the season, every match, every week. And you continue to estimate your prediction of how they will do at season's end.

One last thing - in order to do it well - you have to understand wrestling well enough to see how some wrestlers will match up against other wrestlers based on their specific skills, weaknesses and tendencies (all 900+ of them)

With years of wrestling experience under your belt, and months of following multiple matches each week. Then, and only then, can you do a quality job of ranking.

It's not just stats pumped into a computer program as the season goes on. That might work. Just not very well.

Is it "hard" is it "not hard" - that's all relative. 

 

Relative to putting stats into a spreadsheet with a handful of calculations at seasons end to create a model?

Relatively speaking, ranking is MANY TIMES harder.

Posted
10 minutes ago, RockLobster said:

And you continue to estimate your prediction of how they will do at season's end.

Buddy, rankings try to be objective based on results that have happened. They are not predictions. Keep going though..

Posted (edited)
1 hour ago, BruceyB said:

Buddy, rankings try to be objective based on results that have happened. They are not predictions. Keep going though..

Buddy, rankings are essentially the predictions of outcomes of the wrestlers being ranked.

Which is exactly the reason they are useful. We already have the results - rankings interpret them.

Edited by RockLobster
Posted (edited)
22 hours ago, BruceyB said:

I didn't realize doing college rankings was such a lucrative business.

Aye, Laddie...

ddq1zod-04846ff5-8659-4af8-a717-e14ed02e

That's the ticket!

D3

Edited by D3 for LU

Never argue with an idiot. They will only bring you down to their level and beat you with experience.

Posted
9 hours ago, RockLobster said:

Buddy, rankings are essentially the predictions of outcomes of the wrestlers being ranked.

Which is exactly the reason they are useful. We already have the results - rankings interpret them.

Okay if they are essentially predictions, who will be ranked #1 and #2 at 157 to start next season? 

Posted
9 hours ago, BruceyB said:

Okay if they are essentially predictions, who will be ranked #1 and #2 at 157 to start next season? 

I see you're a persistent, I like that in a dim one. Try to follow along closely:

  • Gable Steveson was the #1 ranked wrestler at Hwt this year
  • Wyatt Pornstache-ickson was the #2 ranked wrestler
  • Wyatt defeated Gable in the finals
  • This was considered, by most everyone, AN UPSET win

Why was it considered AN UPSET win? As if it was some type of surprise?

Because the #2 ranked wrestler defeated the #1 ranked wrestler.

You see, the way it works is that the rankings are essentially predictions of the outcomes of matches.

You're welcome.

Posted
7 hours ago, RockLobster said:

I see you're a persistent, I like that in a dim one. Try to follow along closely:

  • Gable Steveson was the #1 ranked wrestler at Hwt this year
  • Wyatt Pornstache-ickson was the #2 ranked wrestler
  • Wyatt defeated Gable in the finals
  • This was considered, by most everyone, AN UPSET win

Why was it considered AN UPSET win? As if it was some type of surprise?

Because the #2 ranked wrestler defeated the #1 ranked wrestler.

You see, the way it works is that the rankings are essentially predictions of the outcomes of matches.

You're welcome.

You are wrong.  No surprise. It is your brand. Rankings are not predictions. You are using rankings to make predictions.

It was an upset because a two time NCAA champ, two time Hodge winner, undefeated Olympic gold medalist lost to a wrestler who had significantly lesser credentials.

It was not an upset when a #9 seed won a title last year either because the ranking was not a prediction.

  • Bob 3

Drowning in data, but thirsting for knowledge

Posted (edited)
10 hours ago, RockLobster said:

I see you're a persistent, I like that in a dim one. Try to follow along closely:

  • Gable Steveson was the #1 ranked wrestler at Hwt this year
  • Wyatt Pornstache-ickson was the #2 ranked wrestler
  • Wyatt defeated Gable in the finals
  • This was considered, by most everyone, AN UPSET win

Why was it considered AN UPSET win? As if it was some type of surprise?

Because the #2 ranked wrestler defeated the #1 ranked wrestler.

You see, the way it works is that the rankings are essentially predictions of the outcomes of matches.

You're welcome.

To illustrate everyone's point one more way: Mendez, as national champ, is undefeated at 141 until Big 10s next year. Ono is also undefeated at 141, but techs every single guy he faces in the first period the entire season. Mendez will be ranked #1 on every site, because he cannot, on ranking principles, drop until he loses. However, everyone and their mother will be "predicting" an Ono over Mendez result at Big 10s--including the rankers who have Ono slotted #2 on their website. Rankings and predictions are closely correlated--but rankings are win/loss reward based, not predictive, in their purpose.

Edited by maligned
  • Bob 1
Posted
On 3/24/2025 at 7:50 AM, Wrestleknownothing said:

And for @BruceyB, I ballparked it and came up with the "PSU Bump Included Total" of ~183. Based on that PSU underperformed PSU by 6 points. The calls for Sanderson's head are real.

Had Kerkvliet not been injured before the tournament you would have essentially been correct as he probably would have ended up third, if not second, instead of MFF'ing to 6th.

  • Bob 1
Posted
On 3/27/2025 at 6:48 AM, Wrestleknownothing said:

You are wrong.  No surprise. It is your brand. Rankings are not predictions. You are using rankings to make predictions.

It was an upset because a two time NCAA champ, two time Hodge winner, undefeated Olympic gold medalist lost to a wrestler who had significantly lesser credentials.

It was not an upset when a #9 seed won a title last year either because the ranking was not a prediction.

Sorry, but you're wrong.

Lower seeded wrestlers beating higher seeded wrestlers is the very definition of an UPSET.

You already know this. 

Figs #8 over Ayala#3 = Upset
Arujau #6 over Fix #1 = Upset
Carr #4 over Mesenbrink #2 = Upset

And, yes...

#9 Starocci over #6 Welsch = Upset

You're confusing "wrestling" with "gambling on wrestling."

The ranking/seed is essentially the prediction of the outcome prior to the start of competition.

It may be a good or bad prediction, but that's what it essentially is.

Posted
On 3/27/2025 at 9:11 AM, maligned said:

To illustrate everyone's point one more way: Mendez, as national champ, is undefeated at 141 until Big 10s next year. Ono is also undefeated at 141, but techs every single guy he faces in the first period the entire season. Mendez will be ranked #1 on every site, because he cannot, on ranking principles, drop until he loses. However, everyone and their mother will be "predicting" an Ono over Mendez result at Big 10s--including the rankers who have Ono slotted #2 on their website. Rankings and predictions are closely correlated--but rankings are win/loss reward based, not predictive, in their purpose.

And another poster confusing "wrestling" with "gambling on wrestling"

Rankings, just like predictions, are based on past performance. (Either would be bad if they weren't.)

The primary difference between rankers and gamblers is that the rankers only use actual data.

Posted
On 3/27/2025 at 2:11 PM, maligned said:

To illustrate everyone's point one more way: Mendez, as national champ, is undefeated at 141 until Big 10s next year. Ono is also undefeated at 141, but techs every single guy he faces in the first period the entire season. Mendez will be ranked #1 on every site, because he cannot, on ranking principles, drop until he loses. However, everyone and their mother will be "predicting" an Ono over Mendez result at Big 10s--including the rankers who have Ono slotted #2 on their website. Rankings and predictions are closely correlated--but rankings are win/loss reward based, not predictive, in their purpose.

We don't even know if Ono can get away from the bottom position. FS guys often struggle in folk, you know that. 

Posted
18 hours ago, Wrestleknownothing said:

To the simple minded, yes.

If #9 Starocci over #6 Welsch was so obviously going to go Starocci's way (which I agree with)

Then... think for a moment. It's simple. The rankings have always been essentially predictions.

So, perhaps the rankings were done poorly.

Posted
19 hours ago, RockLobster said:

And another poster confusing "wrestling" with "gambling on wrestling"

Rankings, just like predictions, are based on past performance. (Either would be bad if they weren't.)

The primary difference between rankers and gamblers is that the rankers only use actual data.

I've never gambled even a penny on wrestling. But I love to analyze all sports and tinker with predictive ratings. Results are key to those and form the basis for those. But win/loss results alone, in a robotic, traditional rankings methodology are not predictive and don't pretend to be so. You can ask rankers. They will all say that conventions of ranking prevent them from putting who they think are truly the best wrestlers in their proper places in many instances. Again--this is because rankings are a recency-biased win/loss rewards system with standardized conventions. Things like margin of victory, past season performance, non-official results, and known performance-hindering injuries are not incorporated in any way--even though they would dramatically improve the predictive nature of rankings. I know you know this: people who set gambling odds don't care about rankings in any way. They incorporate much more data than wins and losses into their odds. They aren't limited in the facts and data they're allowed to use.  If you picked win/loss favorites based on rankings compared to based on oddsmakers' assessments--the rankings fall far short. Again, because they're not designed, in principle, to be predictive.

Posted
On 3/31/2025 at 6:07 AM, Boring said:

We don't even know if Ono can get away from the bottom position. FS guys often struggle in folk, you know that. 

I can't tell if this comment is tongue-in-cheek. I was presenting an exaggerated scenario to illustrate a point. I wasn't making any comment, in any way, about expectations of actual performance level by Ono this year. Sorry if that wasn't clear.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...