2015 Stunt Nats Seeding

[Howard Rush]

Here is circle seeding for the 2015 stunt Nats.  I did it using the same seeding formula that was used for the previous Stunt Nats.  The seeding formula is based on Nats placing for the last 10 years. US team members not attending the Nats during a world champs year are included, ranked the same as Nats winners. I did the calculation this year.  I put the seeding into my Nats stunt tabulation program.  I'll forward copies of the seeding spreadsheet to anybody who wants it. Likewise, I'll send the tabulation program to anybody who wants it.  I encourage everybody to look at these spreadsheets and critique them.  Here are the top 30 seeds:

1   David   Fitzgerald
2   Derek   Barry
3   Paul Walker
4   Orestes Hernandez
5   Doug Moon
6   Kenny Stevens
7   Matt Neumann
8   Howard Rush
9   Richard Oliver
10   Bob Hunt
11   Kaz Minato
12   Brett Buck
13   Bill Rich
14   Bud Wieder
15   Eric Taylor
16   Frank   McMillan
17   Robert McDonald
18   Bill Werwage
19   Ryan   Young
20   Scott   Reynolds
21   William DeMauro
22   Steve Moon
23   Joe Gilbert
24   Mike Schmitt
25   Dennis Vander Kuur
26   Jason Greer
27   Michael McHenry
28   Mike Greb
29   Germanico Becerril
30   Dan Banjock

For the top 20 in Open and Advanced for the last ten years, first place gets 20 points, second place gets 19, and so on. US Team members who were out of town for the WC get 20 points each. Scores get multiplied by 10 for 2014, 9 for 2013, and so on. Advanced scores are then multiplied by .5. Bud Wieder's score includes both his Open and Advanced placings, for example.  I combined the Expert and Advanced placings for 2013 and 2014.  Top score is seeded #1. Guys who haven't placed in the top 20 in either Advanced or Open are unseeded. Their assignment to one of the four groups for qualifying rounds is done by random draw.
 
The tabulation program tosses out seeds that are not entered at the current Nats and moves those who are entered up to fill in gaps.  For example, if Dave Fitzgerald doesn't show up for this year’s Nats, the tabulation program will bump everybody else up a notch.

This seeding is only used to distribute contestants among four circles for qualifying.  It might make the circles more uniform, but it has little effect on the outcome of the contest.

Although seeding has little effect--none at all on the top handful of fliers-- using a formula published in advance removes any arbitrariness about circle assignment.

[Brett Buck]

1   David   Fitzgerald
2   Derek   Barry
3   Paul Walker
4   Orestes Hernandez
5   Doug Moon
6   Kenny Stevens
7   Matt Neumann
8   Howard Rush
9   Richard Oliver
10   Bob Hunt
11   Kaz Minato
12   Brett Buck
13   Bill Rich

  To beat everyone to the punch...
   "What the hell, I was in the flyoff last year, why the hell am I 12th, why even bother to show up, what a rip-off! And Billy is 18th, this is BS!"


   Brett

[PJ Rowland]

What about my Georgian mate.. how many nats wins between 1 and 3.. lol. Stats huh.

Ohh brett.. your a great number 12ver...

[Tim Wescott]

For the top 20 in Open and Advanced for the last ten years, first place gets 20 points, second place gets 19, and so on. US Team members who were out of town for the WC get 20 points each.

So, if I assume a working goal of getting seeded well at the 2017 Nats, and I don't make the top 20 this year, then I should go to the team trials selection in hopes of making the top 3 there, so I automatically get 20 points for going to Oz?

I dunno Howard -- that seems kinda expensive.

This seeding is only used to distribute contestants among four circles for qualifying.  It might make the circles more uniform, but it has little effect on the outcome of the contest.

Little outcome on the first few places, perhaps, but it should help to reduce the number of people getting bumped out of the top 20 who really ought to be in.

[Trostle]

Little outcome on the first few places, perhaps, but it should help to reduce the number of people getting bumped out of the top 20 who really ought to be in.

Great observation.  That is the reason there is a seeding process.  It is surprising how many people there are who choose not to recognize that.

A corrolary reason is that the seeding process helps assure that each competitor has an equal chance to qualify for the semi-finals so that "luck of the draw" does not superceed skill. 

Keith

[John Leidle]

    Since I'm not listed shall I sign up for Expert this year?
  John

[Howard Rush]

Little outcome on the first few places, perhaps, but it should help to reduce the number of people getting bumped out of the top 20 who really ought to be in.

I did a Monte Carlo thing on that, too.  I forgot the details, but seeding didn't have much effect on the 20th-best or 21st-best guy.  The biggest effect was to keep guys who were 8th best or so from getting bumped out of the top 20. 

[Tim Wescott]

Monte Carlo?

Sigh.

http://www.washington.edu/students/crscat/math.html.  Search on "390".

[Brett Buck]

Monte Carlo?

Sigh.

  That's certainly a helpful comment, sorry we are all too slow to keep up with you. And before you launch into another lecture, I know exactly what Monte Carlo analysis is and what the limitations are.

  I actually have a real counterpoint to Howard's premise, which I will write up soon, but merely expressing your emotions like you find it all very tiresome   really isn't helpful.

    Brett

[Howard Rush]

I think Tim is just funnin' me.  Do I ever pick on him?  Maybe when he feeds me a good straight line.

[Dennis Moritz]

When you miss a NATs or two or judge instead of compete, your ranking plummets. I think that's true. But how did DF wind up on top. I thought he didn't fly, missed a few, when CD.


Sent from my iPhone using Tapatalk

[John Leidle]

   I don't do much to help I admit but if I were ranking pilots I'd put any current or former World Champ at the top, then any Nats winners then Team trial members then the rest. I weigh a lot on experience and the history of coming through in the tough rounds these guys have done just that.
   John

[Howard Rush]

I don't know about that, John.  They're starting to let riffraff on the Team.

[Paul Walker]

   I don't do much to help I admit but if I were ranking pilots I'd put any current or former World Champ at the top, then any Nats winners then Team trial members then the rest. I weigh a lot on experience and the history of coming through in the tough rounds these guys have done just that.
   John

Where would you place someone who won the Nats 20 years ago but has not been in attendance since?

BTW John, the ranking system basically does just what you suggested, and with a simple formula!

Who is there that you don't think should be there?

[Howard Rush]

But how did DF wind up on top. I thought he didn't fly, missed a few, when CD.

Except for 2013, he either won the Nats or was on the team instead of attending the Nats (also 20 points) every year since 2008.  From 2005 to 2007 he was either on the team or placed at least third.  

[John Leidle]

  Hi Paul,  I'm not sure any one is missing certainly not me. But I believe Bob Hunt, Brett Buck & fo dam show Bill Werwege should be in the top 10 . 

[Howard Rush]

Then they ought to go to the Nats more and fly well when they get there. 

[Paul Walker]

  Hi Paul,  I'm not sure any one is missing certainly not me. But I believe Bob Hunt, Brett Buck & fo dam show Bill Werwege should be in the top 10 . 

Yes, by YOUR entirely subjective criteria. The objective criteria looks at their performance in the past 10 years. That's why some you think should be there aren't.

[Howard Rush]

And it really doesn't matter much.  Seeding is not a measure of your worth as a person.  If it were, it's obvious that this year's is wrong in seven cases.  Seeding just spreads the better fliers among the four qualifications circles to take some of the luck out of who qualifies for the semifinals.  We could add former world champions to the formula, but the formula is good enough as is.

[Tim Wescott]

I think Tim is just funnin' me.

Uh -- would I do that?

[Tim Wescott]

That's certainly a helpful comment, sorry we are all too slow to keep up with you.

I'm sorry if my attitude offends you Brett.  I'm quite confident that Howard (or you, for that matter) could pick up this statistics stuff if he were feeling motivated -- it only requires a good knowledge of calculus and a willingness to think sideways* **.  If the comment sounded a bit snarky it's because I was trying to make him feel properly motivated***.

* Far less sideways than if you're a quantum mechanic.

** And Howard seems to enjoy thinking sideways.

*** My 4th-year statistics class was one of the funnest classes I have ever taken.  It's not every day that you get presented with big buckets of math that are weird, esoteric, digestible, and useful all at once.  I still crack that book at least once a year, for paid work.

[Tim Wescott]

Seeding just spreads the better fliers among the four qualifications circles to take some of the luck out of who qualifies for the semifinals.

It appears to do more than that -- it also rewards people for Nats attendance, and for going to the Worlds.  If you just wanted to spread out the better fliers, you'd have a formula that gives the same score for the same level of flying ability to someone who only shows up at every other Nats as for someone who shows up at all of them (counting attendance at the World's as a Nats trip).

I'm not saying anything against the practice -- just noting that as a practical effect of the formula.

[Brett Buck]

It appears to do more than that -- it also rewards people for Nats attendance, and for going to the Worlds.  If you just wanted to spread out the better fliers, you'd have a formula that gives the same score for the same level of flying ability to someone who only shows up at every other Nats as for someone who shows up at all of them (counting attendance at the World's as a Nats trip).

I'm not saying anything against the practice -- just noting that as a practical effect of the formula.

  It barely matters at all, to anyone. The reason we do it at all is to remove complaints about "the hard circle" - not actually prevent "hard circles". It's the NATs, they are all "hard circles". It is done in the manner shown because it was a relatively simple *entirely objective* approach, as opposed to the previous almost entirely objective process (that addressed some of the "concerns" expressed in this thread, but because it was subjective, subjected the people doing it to endless harangues over "rigging the system").

   Of course I was joking above (in a clearly failed attempt to forestall the same argument we have had for 5-6 years now). I personally don't care *at all* where I come out. Put me dead last every year - I don't care, if I can't finish in the top 5 in a 12-15-man contest, I don't deserve to move on. Same with anyone else, as far as I am concerned.

     I don't see how it rewards any legitimate contender. If you can win, you should be able to beat the other guys on your circle in qualifying  The reason it uses past finishes is because that's the only reasonably objective information available.

    Brett

[Paul Walker]

It does prevent the possibility of having one circle with,
Paul, David, Doug, Brett, Derick, and Howard together. One of those would not make it, and that's not right. Yes I know that if you can't make the top 5 in your circle you should not win, but this is extreme.

I doubt that a pure random draw would come up with this combination. I'm sure Howard or Tim can tell you to seven dignificant digits.

[Tim Wescott]

Four circles, right?  So the smallest Nats attendance where that could happen would be 17 guys.  If 17 people show up (including those six guys -- who's "Drieck"?) then there's a 1 in 57480192000 chance* that would happen -- that's 134980364111084491013883777744962985451250075566080** times more likely than the chance that you'll get dealt a natural royal flush in 5-card stud, but it's not zero. If 20 people show up it's 1 in 39230231040000.  I got all tangled up in the possibilities at 21 -- I could do a Monte Carlo simulation, but that's not manly (and perhaps a bit time consuming), so I'll just give up instead.

* Assuming my math is right -- this is the third or fourth iteration of this post.

** You DID ask for significant digits, and since the probabilities are all in the form 1 in <whole number>...  Of course, this is neglecting the chance that someone dies of a heart attack between the circle selection and the end of the contest, which appears to be a heck of a lot more likely than randomly placing your six Golden Guys in one circle.

[Dennis Moritz]

At one time there were comments about so and so distributing fliers on circles to bias advantage in favor of so and so and so and so. Never made sense to me. No doubt never entirely made sense to those who proclaimed the arcane theory. Since every time I heard the spin it spun differently. Given the inclination towards conspiracy theory, Howard's system has an exemplary virtue. It's math.

[Howard Rush]

Four circles, right?

To reduce the probability of having a hard circle (and other undesirable stuff) when attendance is low, we'll use fewer circles.  I should write this up separately.  For either event, if there are fewer than nine contestants, we'll use one circle: no seeding.  For 9-16 contestants, two circles (as was done at the last Nats for Expert and Advanced).  For 17-24 contestants, three circles.  For more than 24, four circles. 

[Tim Wescott]

Wow.  If you're going to get 19 or fewer contestants in Open let me know -- top 20, here I come!

Paul's "six golden guys in one circle" probability goes up dramatically if you only have one circle.

[Brett Buck]

At one time there were comments about so and so distributing fliers on circles to bias advantage in favor of so and so and so and so. Never made sense to me. No doubt never entirely made sense to those who proclaimed the arcane theory.

   Particularly since the supposed beneficiaries and the putative  "screw-ees" blasted through qualifying every year with absolutely no drama, just like you would have expected. It made no sense whatsoever. As soon as this became evident and the argument suddenly seem silly to everyone, the same same people moved on to judge selection. Again, this despite the putative  "screw-ee" qualifiying for the flyoff year after year in a remarkable run.

  A far more plausible answer to this is that said "screw-ee" got to Saturday morning fairly easily, convinced himself that the system was rigged and psyched himself out. You could watch the body language from 50 yards away - defeated before the first flight! The same thing happened at the Team Trials, sans the qualifying.

  This is critically important, and you see it to varying degrees all the time - the *very instant* you get it in your head that some sort of external agency is controlling your destiny, you are sunk. It doesn't matter if it is true or not, you have to keep the problem under your own control. The only thing you can have any control over is your own actions and performance, you can't control anyone else's, and you have to evaluate very issue as a challenge for yourself. You cannot make excuses, blame the universe, etc, because it doesn't matter.

    It's particularly true in stunt. People defeat themselves over and over for years, and never get out of it. Virtually ANYONE can be competitive if they approach it correctly, but most people don't, largely because of they let this stuff get to them and don't react in a way that is useful.

   I hate stupid "success" homilies with a passion but - "whether you think you can, or think you can't, you are correct".

    Brett

   

[Brett Buck]

To reduce the probability of having a hard circle (and other undesirable stuff) when attendance is low, we'll use fewer circles.  I should write this up separately.  For either event, if there are fewer than nine contestants, we'll use one circle: no seeding.  For 9-16 contestants, two circles (as was done at the last Nats for Expert and Advanced).  For 17-24 contestants, three circles.  For more than 24, four circles. 

  Note the extensive discussion of it here:

http://stunthanger.com/smf/index.php/topic,27617.0.html


     I think Steve Yampolsky came up with the algorithm for ADV/EXP/Open but it still works for ADV/Open, I think. I still think it makes more sense to reduce the "Top 20" to fit the entry, but it's far too ingrained and has too much tradition to make that a popular option. Right now there seems to be no problem with sufficient entry, so it probably doesn't need a solution.

    I thought it generally worked better with 2 Open and 2 ADV even with abundant entries. But they changed it in 2004 anticipating a wild imbalance in entries (which is exactly what happened). It would have been a disaster with a 2 and 2 format - couldn't have gotten through two rounds of Open in one day. We ran the 4-circle mixed format since then.

    Brett

[phil c]

And it really doesn't matter much.  Seeding is not a measure of your worth as a person.  If it were, it's obvious that this year's is wrong in seven cases.  Seeding just spreads the better fliers among the four qualifications circles to take some of the luck out of who qualifies for the semifinals.  We could add former world champions to the formula, but the formula is good enough as is.

Sorry Howard, more like wrong in 30 cases.  Since the formula prevents obvious blunders by individual judgements, as you say, it's good enough.

[Derek Barry]

(including those six guys -- who's "Drieck"?)

Some guy from the south that spends too much time fishing and not enough time flying.

Derek

[Derek Barry]

What about my Georgian mate.. how many nats wins between 1 and 3.. lol. Stats huh.

Ohh brett.. your a great number 12ver...

Thats what a lot of top five finishes and no wins will get you. 2nd spot on a list that only slightly matters.

Derek

[Derek Barry]

Assuming everyone shows up (which never happens) this would be the top five seeds in 4 circles.

Circle 1               Circle 2           Circle 3           Circle 4
David                  Drieck            Puale              Orestes
Doug                   Kenny            Matt                Howard
Rich                    Hunt               Kaz                Brett
Bill Rich               Bud W.           Eric                Frank
Bob McDonald       Werwage        Ryan               Scott

[RC Storick]

To reduce the probability of having a hard circle (and other undesirable stuff) when attendance is low, we'll use fewer circles.  I should write this up separately.  For either event, if there are fewer than nine contestants, we'll use one circle: no seeding.  For 9-16 contestants, two circles (as was done at the last Nats for Expert and Advanced).  For 17-24 contestants, three circles.  For more than 24, four circles.  

God forbid the top 20 fliers year after year would get on a circle and they might be eliminated from continuing the same thing and passing on to the next round. The reason why the top 20 never changes is they don't fly against anyone but the same people year after year. I know the reason for the contest is to pick the best flyer so that person should rise to the top regardless if its a seeded circle or a luck of the draw circle. Out of those years (when I was 21st) would have been my best shot but I am not popular. Oh well

Before you I get slammed for BS "BB" it doesn't matter to me any longer as I am flying classic and not open. Brett said why show up. I tried 10 years to make it and didn't make the club. Good luck

[Brett Buck]

Before you I get slammed for BS "BB" it doesn't matter to me any longer as I am flying classic and not open. Brett said why show up.

   I don't think you need my help on this one.

   Again, offer still stands, I will assist in any way that will actually help, if you change your mind.

   Brett

[Tim Wescott]

Thats what a lot of top five finishes and no wins will get you. 2nd spot on a list that only slightly matters.

Think of it like championship points in NASCAR.

[Howard Rush]

Assuming everyone shows up (which never happens) this would be the top five seeds in 4 circles.

Circle 1               Circle 2           Circle 3           Circle 4
David                  Drieck            Puale              Orestes
Doug                   Kenny            Matt                Howard
Rich                    Hunt               Kaz                Brett
Bill Rich               Bud W.           Eric                Frank
Bob McDonald       Werwage        Ryan               Scott

It's a little different because of the folding, which keeps circle 1 from getting higher seeded guys than the others:

Circle 1               Circle 2           Circle 3           Circle 4
David                  Drieck            Puale              Orestes
Howard                Matt             Kenny              Doug            
Richard                Bob Hunt        Kaz                Brett
Frank                 Eric Taylor       Bud W.           Bill Rich               
Bob McDonald       Werwage        Ryan               Scott

[Derek Barry]

It's a little different because of the folding, which keeps circle 1 from getting higher seeded guys than the others:

Circle 1               Circle 2           Circle 3           Circle 4
David                  Drieck            Puale              Orestes
Howard                Matt             Kenny              Doug            
Richard                Bob Hunt        Kaz                Brett
Frank                 Eric Taylor       Bud W.           Bill Rich              
Bob McDonald       Werwage        Ryan               Scott

I don't understand the folding? Or why it only affects circle 1?

[Tim Wescott]

I don't understand the folding? Or why it only affects circle 1?

It affects all the circles.  Instead of circle #1 getting guy #1, guy #5, etc., circle #1 gets guy #1, guy #8, guy #9, etc.  Circle 2 gets 2, 7 and 10 instead of 2, 5 and 10.  Basically, the average placing of the guys seeded in each circle are the same with folding, but without the average is highest in 1 and lowest in 4.

So it prevents circle #1 from being the "hard" circle.  Sorta.

[Brett Buck]

I don't understand the folding? Or why it only affects circle 1?

  It doesn't only affect circle 1. It goes 1-2-3-4-4-3-2-1 instead of 1-2-3-4-1-2-3-4. Look at the NCAA brackets  #1 gets #16, 2 gets 15 etc. and 8 gets 9.

   This makes Circle 4 the "hard" circle and 1 the "easy circle". Which I fully intend to whine about at the first opportunity -  all they are doing is gettinga bunch of hot-shots to beat up on me, and Drieck gets only gets guys with 4 WC between them! More depressing is that now even pure mathematics is against me, used to it was just random chance, and anyone who had ever met me, or not.

  I said it before and I will say it again, this is the worst Sekrit Kabal I have ever been in!

   Brett

[john e. holliday]

Settle the whole thing easily.  Go back to the NAVY days when I heard that when you got ready to fly you gave them a card and got in line, if there were some ahead of you.   5 PM flying stopped and scores were posted.  At least that is what I heard.   But, reality some will complain no matter how it is done.  I know a few that made top twenty in just their year.   But, what do I know as I've only flown Classic and Intermediate at past NATS.

[Robby Hunt]

 
   This makes Circle 4 the "hard" circle and 1 the "easy circle". Which I fully intend to whine about at the first opportunity -  all they are doing is gettinga bunch of hot-shots to beat up on me, and Drieck gets only gets guys with 4 WC between them! More depressing is that now even pure mathematics is against me, used to it was just random chance, and anyone who had ever met me, or not.

 

Hey, at least your not being tracked by LeFors...

[Derek Barry]

 It doesn't only affect circle 1. It goes 1-2-3-4-4-3-2-1 instead of 1-2-3-4-1-2-3-4. Look at the NCAA brackets  #1 gets #16, 2 gets 15 etc. and 8 gets 9.

   This makes Circle 4 the "hard" circle and 1 the "easy circle". Which I fully intend to whine about at the first opportunity -  all they are doing is gettinga bunch of hot-shots to beat up on me, and Drieck gets only gets guys with 4 WC between them! More depressing is that now even pure mathematics is against me, used to it was just random chance, and anyone who had ever met me, or not.

  I said it before and I will say it again, this is the worst Sekrit Kabal I have ever been in!

   Brett

Got it! Thanks.

Derik

[Derek Barry]

 It doesn't only affect circle 1. It goes 1-2-3-4-4-3-2-1 instead of 1-2-3-4-1-2-3-4. Look at the NCAA brackets  #1 gets #16, 2 gets 15 etc. and 8 gets 9.

   This makes Circle 4 the "hard" circle and 1 the "easy circle". Which I fully intend to whine about at the first opportunity -  all they are doing is gettinga bunch of hot-shots to beat up on me, and Drieck gets only gets guys with 4 WC between them! More depressing is that now even pure mathematics is against me, used to it was just random chance, and anyone who had ever met me, or not.

  I said it before and I will say it again, this is the worst Sekrit Kabal I have ever been in!

   Brett

There is a strong chance that two of the top five on your circle will not be flying at the Nats this year so who knows what we will get in the final draw. The most important thing is that I will not have to fly on circle 1 this year, I prefer 2 and 4.

Derek

[Scott Richlen]

Why is Banjock ranked 30th?

I thought you added a Bi-Slob round on Friday?  And it is about time!  How can a person be a National Champion just because they fly some exotic electric-powered, aero-dynamically pristine, miniature aerobatics ship?  What's the challenge in that?

And why can't we add a Ringmaster Round on Saturday?  That would show who can really fly!

[Brett Buck]

There is a strong chance that two of the top five on your circle will not be flying at the Nats this year so who knows what we will get in the final draw.

   I'll find some other reason to whine/excuse when I see the final arrangement, I just wanted to get "first whine".

The most important thing is that I will not have to fly on circle 1 this year, I prefer 2 and 4.

  Oh, that's the most important thing?  You should fly over there all the time, so all the others seem easier. You will not prefer 4 if the wind is coming from the West, and the circus tent is in place. Ask Todd about that from 2003, they are probably still finding bits of camo-painted balsa out in the soybean field. At least it was self-cleaning.

  I have done mostly OK over there. The only problem I had was in 2004, when Billy and I were on that circle, I flew an OK flight in the howling wind right up the slope that would have put me first or second - and then they cancelled the round!

   Same in 2006, Ted and I both flew really good flights, again in substantial air. Ted's in particular looked like the wind had been cancelled. Great flight, great landing, rolled about 3/4 of a circle around into the wind, then took off again about 4 feet up!  Good times...

    Brett

[Brett Buck]

And why can't we add a Ringmaster Round on Saturday?  That would show who can really fly!

   Sadly, what it would show is that David can really fly, he's a Ringmaster god. Just be sure and make him fly *his own airplane* - his record of destruction with other people's Ringmasters (Ted...)  is not promising.

  If there was room in the car, I would bring the Skyray and we could to a one-airplane contest.

    Brett