Jump to content

Wrestlebacks and 2-class individual tournamnet - compare and contrast


Galagore

Recommended Posts

In the past, I have proposed advancing 5 out of each semi state, using the wrestlebacks to 5th place that the IHSAA did one year.  This keeps the "lucky draw" and excitement of the ticket round while also getting another deserving wrestler to state.

 

This could only help the small schools and number of qualifiers.  They would still have every chance they currently have plus some of the "bad draws" would get added in as well.

Link to comment
Share on other sites

1 hour ago, Galagore said:

I hate to be a nit-picker, but not enough to keep my mouth shut...What you all are calling the "law of averages" is in reality the "Law of Large Numbers." This law basically states that, given a large enough number of trials, observed outcomes will eventually arrive close enough to expected outcomes that any difference is negligible.

 

The law of averages is not a real mathematical principal.

 You are correct. but I am not saying that half the qualifiers will be small school kids.

 

It is a fact that for every kid who gets a good draw, there is one who gets a bad draw.

 

So all I am saying is that over the long haul, that will be the case for small school kids. Maligned is implying that more small school kids will be the 5th & 6th best at a given tournament than are 3rd & 4th best. I'm saying I don't believe that to be the case.

 

The numbers he himself provided were that over the past 2 years from NC semi-state there had been (3) 1sts, (2) seconds, (2) thirds & (6) 4ths. But these 4th's could in reality have only been 5th or 6th with wrestlebacks. So just looking at those numbers, it is not unreasonable to assume that of those six, in reality (2) were legitimate 4ths, (2) were 5ths & (2) were 6ths. That is by continuing the pattern of 1sts, 2nds & 3rds.

 

So if there were (4) 3rds & 4ths & (4) 5ths & 6ths, then over time for every 5th & 6th that got a lucky draw, a 3rd & 4th would get an unlucky one. The law of averages does apply here. I concede that I have made assumptions, but so has Maligned.

Link to comment
Share on other sites

12 minutes ago, SIACfan said:

 You are correct. but I am not saying that half the qualifiers will be small school kids.

 

It is a fact that for every kid who gets a good draw, there is one who gets a bad draw.

 

So all I am saying is that over the long haul, that will be the case for small school kids. Maligned is implying that more small school kids will be the 5th & 6th best at a given tournament than are 3rd & 4th best. I'm saying I don't believe that to be the case.

 

The numbers he himself provided were that over the past 2 years from NC semi-state there had been (3) 1sts, (2) seconds, (2) thirds & (6) 4ths. But these 4th's could in reality have only been 5th or 6th with wrestlebacks. So just looking at those numbers, it is not unreasonable to assume that of those six, in reality (2) were legitimate 4ths, (2) were 5ths & (2) were 6ths. That is by continuing the pattern of 1sts, 2nds & 3rds.

 

So if there were (4) 3rds & 4ths & (4) 5ths & 6ths, then over time for every 5th & 6th that got a lucky draw, a 3rd & 4th would get an unlucky one. The law of averages does apply here. I concede that I have made assumptions, but so has Maligned.

 

I am not questioning your logic. However, the law of averages is still not a thing, regardless of how you apply it. You are referring to the Law of Large Numbers.

 

Just being a math teacher here, not taking sides in this particular debate.

 

My stance on class: Two classes for the individual tournament

My stance on wrestle-backs: Yes, we should have them

My stance on the law of averages: It is a gambler's fallacy (this is actually fact, not simply my stance)

Link to comment
Share on other sites

2 minutes ago, Galagore said:

 

I am not questioning your logic. However, the law of averages is still not a thing, regardless of how you apply it. You are referring to the Law of Large Numbers.

 

Just being a math teacher here, not taking sides in this particular debate.

 

OK, the 'law of large numbers' then.

 

Thanks teach!

Link to comment
Share on other sites

I myself was a ticket rounder last year and i had a pretty tough draw at regionals and semi state, maybe it was just me but i felt like that without wrestlebacks you have to be on your A game 100% of the time and it made me better as a competitor with that sense of urgency so to speak. upsets are fun to see and the Cinderella runs are awesome i love how our system is set up even if some wrestlers that could punch their ticket to state in other quads of the bracket. but if the IHSAA chooses to change it id still support it

Link to comment
Share on other sites

On 2/18/2021 at 10:42 AM, SIACfan said:

 

Additionally, I question the accuracy of the 79% 'Lucky Quarters' rate. Where does this come from? If it is accurate than that means that (4) out of every (5) 4th place qualifiers is not the participant who deserved to advance & only reinforces the idea that wrestlebacks are needed. But I concede that just because wrestlebacks are needed doesn't mean you are wrong about losing small school qualifiers. But my contention is that ensuring the correct kids are advancing is more important than the school size they come from.

 

It's just probability calculations. Assuming 19 per regional, assuming random placement of best 4 on different teams, and assuming the best 4 lose only to each other, probabilities of all scenarios:

One per regional (therefore avoiding each other till semi's): 1.7%

All four in same regional (therefore avoiding each other till semi's): 1.2%

Three in one regional, one out of another (and avoid each other till semi's): 4.3%

Two from one regional, one each from two others (and avoid each other till semi's): 9.6%

Two from each of two regionals (and avoid each other till semi's): 4.6%

Total probability of the best 4 avoiding each other till the semi's: 21%

Probability of at least one quarter bracket with no top 4 kids: 79%

 

Obviously this includes rigid assumptions. But it illustrates how incredibly often at least one quarter-bracket has none of the best kids. 

 

To be clear from something one of you said a few comments back: I'm not trying to make any specific point for or against class anything. I'm just stating the mathematical facts. One aspect I'm realizing I haven't communicated clearly and that you're wrongly assuming: The distribution of small school kids--percentage-wise according to their class--is less in the top semi-state placements. There is not even distribution among small schools of 8th, 7th, 6th, 5th-best, on down the list. The distribution is falling as you get to the higher placements compared to the distribution of kids that are 5th, 7th, 10th best. This is not true of the distribution of kids coming from big schools (look at the last 5 years at NC SS...same story as the last two). Because of this descending distribution, a bigger percentage of kids that get to state from small schools were brought in by those lucky quarters and a bigger percentage of big-school qualifiers relative to their class would have gotten in no matter what. Because this is true, again, relative to their class, each time we swap a 5th or 7th best kid for a 3rd or 4th best kid there's a small increase in probability our small-school qualifier number has gone up relative to big-school qualifiers.

We start the state series with 66% small school kids (1A-2A). Their distribution compared to big schools falls at every level and placement. We have no where near 66% of small school kids placing 1st or 3rd at semi-state. The rate has fallen dramatically by then. Imagine for simplicity's sake, we've fallen to 15% small school at the 10th best level in a given semi-state and 6% at the semi-state champ level. That means we have approximately 10-12% small school kids at the 5th-7th best range and only 8-9% at the 3rd to 4th best range. Every time we swap someone from that lower quality range in place of someone from the better quality range because of no wrestle backs, we slightly increase the chance we took a small-school kid to state because of that descending comparative distribution relative to large-school kids.

 

Link to comment
Share on other sites

Even at state, this descending distribution of 1A/2A kids continues:

 

43 non-placers (38%), 28 placers (25%)

 

11 7th-8th (39%), 8 5th-6th (29%), 5 3rd-4th (18%), 4 2nd-1st (14%)

 

In the above group, for example, if you take five random placer-level kids and swap them for five random non-placers, you increase the probability of having small-schoolers in the placer group. Once more, take 20 3rd-4th quality kids at every semi-state and swap them for 20 5th-7th quality kids because of no wrestle backs, you slightly increase the population density of small-schoolers. 

Link to comment
Share on other sites

Here’s my take. NO CLASSES.

 

as for wrestle backs, they should be as follows. 
 

sectionals, full wrestle backs to top 6 (just like they are now. )

 

regionals keep them the way they are. 
 

semi state. Full wrestle backs to 3/4 and 5/6 after at the ticket round. So single elimination round 1 then ticket round matches get to wrestle back for 3-6 and we find a true alternate. 
 

state, keep it the way it is. 
 

let you know your thoughts on that

Link to comment
Share on other sites

7 hours ago, Jayruss said:

Not going to lie I love single class! Wrestlebacks will definitely be an improvement, but keep the single class!

 

I think most (self included) agree that they love our single class system. However, what I love and what is best for our sport are two separate entities in my mind. A two class system would help preserve the sport, and that is what I love about it.

Link to comment
Share on other sites

11 hours ago, maligned said:

It's just probability calculations. Assuming 19 per regional, assuming random placement of best 4 on different teams, and assuming the best 4 lose only to each other, probabilities of all scenarios:

One per regional (therefore avoiding each other till semi's): 1.7%

All four in same regional (therefore avoiding each other till semi's): 1.2%

Three in one regional, one out of another (and avoid each other till semi's): 4.3%

Two from one regional, one each from two others (and avoid each other till semi's): 9.6%

Two from each of two regionals (and avoid each other till semi's): 4.6%

Total probability of the best 4 avoiding each other till the semi's: 21%

Probability of at least one quarter bracket with no top 4 kids: 79%

 

 

I am confused (I know, I know, big surprise).

 

You have listed the (5) possible scenarios for a semi-state.

-1 of the 4 best kids comes from each regional

-all 4 come from one regional

-3 from one regional & 1 from another

-2 from one regional & 1 from two others

-2 from two regionals

 

You also stated the assumption is they will only lose to each other.

 

Thus if 1 comes from each regional then it is 100% that they won't face each other until the semi's. Likewise, if they all come from the same regional then it is 100% that they don't face each other until the semi's. So clearly that is not the percentages you are presenting.

 

So are you saying that there is only a 1.7% chance that 1 comes from each regional, that it is a 1.2% chance that all come from the same regional, that it is a 4.3% chance that there is 3 from one regional & 1 from another, a 9.6% chance that it is 2 from one regional & one from two others, & lastly a 4.6% chance that 2 come from two regionals?

Link to comment
Share on other sites

22 minutes ago, SIACfan said:

 

I am confused (I know, I know, big surprise).

 

You have listed the (5) possible scenarios for a semi-state.

-1 of the 4 best kids comes from each regional

-all 4 come from one regional

-3 from one regional & 1 from another

-2 from one regional & 1 from two others

-2 from two regionals

 

You also stated the assumption is they will only lose to each other.

 

Thus if 1 comes from each regional then it is 100% that they won't face each other until the semi's. Likewise, if they all come from the same regional then it is 100% that they don't face each other until the semi's. So clearly that is not the percentages you are presenting.

 

So are you saying that there is only a 1.7% chance that 1 comes from each regional, that it is a 1.2% chance that all come from the same regional, that it is a 4.3% chance that there is 3 from one regional & 1 from another, a 9.6% chance that it is 2 from one regional & one from two others, & lastly a 4.6% chance that 2 come from two regionals?

Right, these scenarios represent the probability each type happens AND that the top 4 subsequently fall into 4 different brackets. Obviously there's a 100% chance the best 4 dont meet if they come from 4 regionals--but there's only a 1.7% chance they are distributed that way. Much bigger chance of 3-1-0-0 or 2-1-1-0, for example, but then bracket only falls favorably part of the time. So again, each scenario is probability of the distribution happening AND the bracket working out with one of the top 4 per quarter.

Edited by maligned
Link to comment
Share on other sites

9 minutes ago, maligned said:

Right, these scenarios represent the probability each type happens AND that the top 4 subsequently fall into 4 different brackets. Obviously there's a 100% chance the best 4 dont meet if they come from 4 regionals--but there's only a 1.7% chance they are distributed that way. Much bigger chance of 3-1-0-0 or 2-1-1-0, for example, but then bracket only falls favorably part of the time. So again, each scenario is probability of the distribution happening AND the bracket working out with one of the top 4 per quarter.

 

OK, so if you ignore the caveat of the top 4 avoiding each other until semi's then the percentages of the 5 scenarios have to equal 100%.

 

I find it hard to believe that the chance of 2 of the 5 scenarios happening (1 from each & all from one) is only 2.9% and the other 3 scenarios happen 97.1% of the time.

Link to comment
Share on other sites

1 hour ago, SIACfan said:

 

OK, so if you ignore the caveat of the top 4 avoiding each other until semi's then the percentages of the 5 scenarios have to equal 100%.

 

I find it hard to believe that the chance of 2 of the 5 scenarios happening (1 from each & all from one) is only 2.9% and the other 3 scenarios happen 97.1% of the time.

You're right. I screwed up. There's a 43% chance of the bracket falling "fairly"--not 21%. (What's funny is..I did these same calculations like 10 years ago and thought it was more like 60/40. I figured out now what I did wrong this time.) Sorry.

 

Again, assuming 19 teams per regional, random distribution, and top 4 only beat each other, here are probabilities for "fair" semi-state brackets:

 

1-1-1-1: 10.2% instance, 100% bracket correctness = 10.2% overall "fairness" contribution

4-0-0-0: 1.2% instance, 100% bracket correctness = 1.2% overall 

3-1-0-0: 17.2% instance, 33% bracket correctness = 5.7% overall 

2-1-1-0: 57.7% instance, 33% bracket correctness = 19.2% overall

2-2-0-0: 13.7% instance, 50% bracket correctness = 6.8% overall

 

Overall fairness total: 43%

 

Chance of an "unfair" bracket = 57% 

 

Edited by maligned
Link to comment
Share on other sites

13 minutes ago, maligned said:

You're right. I screwed up. There's a 40% chance of the bracket falling "fairly"--not 21%. (What's funny is..I did these same calculations like 10 years ago and thought it was more like 60/40. I figured out now what I did wrong this time.) Sorry. Here they are:

 

1-1-1-1: 10.2% instance, 100% bracket correctness = 10.2% overall "fairness" contribution

4-0-0-0: 1.2% instance, 100% bracket correctness = 1.2% overall 

3-1-0-0: 17.2% instance, 33% bracket correctness = 5.7% overall 

2-1-1-0: 57.7% instance, 33% bracket correctness = 19.2% overall

2-2-0-0: 13.7% instance, 25% bracket correctness = 3.4% overall

 

Overall fairness total: 39.8%

 

Chance of an "unfair" bracket = 60.2% 

 

 

Ok, now I have even more questions.

 

How is a 3-1-0-0 scenario only 33% bracket correct? The 3 from the same regional are going to be in different quarter brackets & the other top guy (regional champ) may or may not fall into one of the 2 non-champs from the loaded regional. Thus there has to be at minimum 75% bracket correctness with the possibility of 100%

 

Also the same is true for the 2-1-1-0 scenario. The 3 regional champs who are top 4 guys will all be in a different quarter brackets & the non-champ top guy may or may not fall into of the other tops guys brackets. So again there is at a minimum 75% bracket correctness with the possibility of 100%.

Link to comment
Share on other sites

1 minute ago, SIACfan said:

 

Ok, now I have even more questions.

 

How is a 3-1-0-0 scenario only 33% bracket correct? The 3 from the same regional are going to be in different quarter brackets & the other top guy (regional champ) may or may not fall into one of the 2 non-champs from the loaded regional. Thus there has to be at minimum 75% bracket correctness with the possibility of 100%

 

Also the same is true for the 2-1-1-0 scenario. The 3 regional champs who are top 4 guys will all be in a different quarter brackets & the non-champ top guy may or may not fall into of the other tops guys brackets. So again there is at a minimum 75% bracket correctness with the possibility of 100%.

Definition of "bracket correctness": Landing all 4 of our "top 4" in separate quarters.

 

For the 3-1-0-0 scenario:

You have a champ, 2nd, and 3rd from one semi-state who will be placed so that the champ is on one half and the 2nd and 3rd are on the other side in separate quarters. The probability then is very simply a 1 in 3 chance that the remaining guy (a champ from another regional) lands in a quarter not occupied by one of the 2nd or 3rd place guys from the 3-man regional.

 

For the 2-1-1-0 scenario:

You have 3 champs and one runner-up. Once we place the champ and runner-up from the "2" regional, there's a 1 in 3 chance the other two champs will land so that they're not in the same quarter as the concerned runner-up.

 

Link to comment
Share on other sites

21 minutes ago, SIACfan said:

Furthermore, the 2-2-0-0 scenario has to be at least 50% correct.

You're right for that one. Exactly 50%. For some unknown reason, I didn't input the 1/3 scenario where the two champs are on the same side. One third possibility for that, plus 2/3 * 1/4 = 1/6 possibility of the two champs being in opposite halves AND the two runners-up both falling into the quarters not occupied by our concerned champs. One third plus one sixth = one half. 

 

I'll adjust numbers above. Slightly below 60% failure rate.

Edited by maligned
Link to comment
Share on other sites

31 minutes ago, SIACfan said:

 

Ok, now I have even more questions.

 

How is a 3-1-0-0 scenario only 33% bracket correct? The 3 from the same regional are going to be in different quarter brackets & the other top guy (regional champ) may or may not fall into one of the 2 non-champs from the loaded regional. Thus there has to be at minimum 75% bracket correctness with the possibility of 100%

 

Also the same is true for the 2-1-1-0 scenario. The 3 regional champs who are top 4 guys will all be in a different quarter brackets & the non-champ top guy may or may not fall into of the other tops guys brackets. So again there is at a minimum 75% bracket correctness with the possibility of 100%.

For 3-1-0-0, it's only correct if the top 4 are separated into the 4 different quarter brackets.  So that 1 has to be in with the 4th place finisher.  The only other options are with the 3rd place finisher or the 2nd place finisher.  Thus, 33% chance of being correct.

 

2-2-0-0 uses the same logic, but with more scenarios.

Link to comment
Share on other sites

On 2/14/2021 at 10:38 AM, Galagore said:

Can anyone explain to me the logic that leads someone to say that we should not have two (or more, but I like two) classes in our individual tournament because it is exciting and/or the goal is to crown one true champion in each weight (two most used pieces of logic that I have read). Many of the same people say that we should have wrestlebacks because a lot of kids who deserve to go to state don't get to go. Semi-finals, ticket round, and Friday night are my favorite rounds to watch, and it is because of the non-wrestleback, single-class system. The state tournament should not be organized based on what I like to watch, so I am in favor of both wrestlebacks and a two-class system.

 

Truly asking because I am a curious bird who likes to know how people think and not simply trolling UncleJimmy (didn't even tag you so you could ignore this thread if you wish)...How does one come to the conclusion that class is bad and wrestlebacks are good?

Bottom Line.....Big Schools don't want to wrestle each other at the beginning of the tournament series.  

If they do, that means less qualifiers, less placers, less post season success.  

 

Link to comment
Share on other sites

1 minute ago, bulldog145 said:

Bottom Line.....Big Schools don't want to wrestle each other at the beginning of the tournament series.  

If they do, that means less qualifiers, less placers, less post season success.  

 

 

Interesting...

Link to comment
Share on other sites

I understand, I was thinking the percent that the brackets are correct & you are calculating the percentage of time the brackets are 100% correct.

 

But I believe the percentage of time the brackets will be 100% correct under the 2-2-0-0 scenario is 28.6%. If you label the 4 brackets A, B, C & D and the 4 top guys as 1, 2, 3 & 4 and say the 2 champs (guys 1 & 2) are in brackets A & B, and if #4 can't be in A-bracket & #3 can't be in B-bracket then the possible scenarios are as follows:

 

A   0   0   3   3   0   0   3

B   0   0   0   0   4   4   4

C   3   4   4   0   3   0   0

D   4   3   0   4   0   3   0

 

So there are 2 of 7 (28.6%) scenarios where the brackets are 100% correct.

 

 

 

Edited by SIACfan
Link to comment
Share on other sites

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.