It is that time of year where I dust off WrestleSim and see what it is telling us. Official seeds will be out tomorrow so I will update. This is a teaser for the more complete analysis to come.
Myth 1: No, you do not need eight scoring wrestlers and no I am not going to meet you on a bridge to fight about it.
Myth 2: No, you do not need 20 bonus points and no I am not going to meet you under a bridge to fight about it, but I will meet you on a bridge to fight.
Myth 3: Underseeding hurts the underseeded wrestler. Overseeding helps the overseeded wrestler.
Before people get all worked up about seeding, it doesn't impact the results all that much. Only in RARE circumstances (channeling my inner bird with the CAPS!) is the underseeded wrestler affected negatively. It mostly affects his lower ranked opponents and can block a guy from wrestling back to place that otherwise would have made it. Upsets of highly ranked wrestlers matters far more, especially if a guy gets his act together once he is knocked to the backside. Overseeding hurts the overseeded wrestler, it doesn't help him. These coaches that play seeding games and endorse match ducking, totally ass backwards.
QUIZ: Which Penn State match/wrestler showed a tactic to defeat the "rider" who continually runs his guy off the mat? I noticed it and the tactic was beautiful. Hopefully when widely deployed it will eliminate this behavior from the sport. A friend suggested a fix - you get 30 seconds to ride and if you don't get a turn, back to your feet and no escape point. I like this idea. I also like 3 points for a takedown to encourage more neutral action (Or would it? would a guy with the lead shut down the offense and go on defense). I don't know but anything that prevents riding with no turns and neutral with no shots is good in my book. Not everyone can be Andonian but I wish they would try more.
Penn State this year is a heavy favorite (duh). The base model using Intermat's rankings as seeds has (to win) PSU 98.4%, Iowa 1.0%, Missouri 0.4%, Cornell 0.3%, and ISU 0.1%. Everyone else gets the Blutarsky 0.00. This was based on 1,000 realizations of the probabilistic model. The more accurate simulation where adjustments are made to account for some top seeds being heavier favorites than other top seeds has the results shown below. A bit counter intuitive because you may think "Spencer Lee is a lock, the Hawks' chances should go up" but it doesn't work that way because the chances of the lower ranked wrestlers pulling the upset (think #10 seed winning and getting 20 points instead of ~3 points) gets eliminated/reduced. The Hawks chance of second goes up from about 42% to 62% but their chance of first goes down.
UFF
Myth 1: No, you do not need eight scoring wrestlers and no I am not going to meet you on a bridge to fight about it.
Myth 2: No, you do not need 20 bonus points and no I am not going to meet you under a bridge to fight about it, but I will meet you on a bridge to fight.
Myth 3: Underseeding hurts the underseeded wrestler. Overseeding helps the overseeded wrestler.
Before people get all worked up about seeding, it doesn't impact the results all that much. Only in RARE circumstances (channeling my inner bird with the CAPS!) is the underseeded wrestler affected negatively. It mostly affects his lower ranked opponents and can block a guy from wrestling back to place that otherwise would have made it. Upsets of highly ranked wrestlers matters far more, especially if a guy gets his act together once he is knocked to the backside. Overseeding hurts the overseeded wrestler, it doesn't help him. These coaches that play seeding games and endorse match ducking, totally ass backwards.
QUIZ: Which Penn State match/wrestler showed a tactic to defeat the "rider" who continually runs his guy off the mat? I noticed it and the tactic was beautiful. Hopefully when widely deployed it will eliminate this behavior from the sport. A friend suggested a fix - you get 30 seconds to ride and if you don't get a turn, back to your feet and no escape point. I like this idea. I also like 3 points for a takedown to encourage more neutral action (Or would it? would a guy with the lead shut down the offense and go on defense). I don't know but anything that prevents riding with no turns and neutral with no shots is good in my book. Not everyone can be Andonian but I wish they would try more.
Penn State this year is a heavy favorite (duh). The base model using Intermat's rankings as seeds has (to win) PSU 98.4%, Iowa 1.0%, Missouri 0.4%, Cornell 0.3%, and ISU 0.1%. Everyone else gets the Blutarsky 0.00. This was based on 1,000 realizations of the probabilistic model. The more accurate simulation where adjustments are made to account for some top seeds being heavier favorites than other top seeds has the results shown below. A bit counter intuitive because you may think "Spencer Lee is a lock, the Hawks' chances should go up" but it doesn't work that way because the chances of the lower ranked wrestlers pulling the upset (think #10 seed winning and getting 20 points instead of ~3 points) gets eliminated/reduced. The Hawks chance of second goes up from about 42% to 62% but their chance of first goes down.
UFF
