<html><div style='background-color:'><DIV class=RTE>
<P>Arch, et al - </P>
<P>This is a GREAT discussion - wish it had been (formally in the big forum ) done before the rules proposal was submitted . </P>
<P>My thoughts: </P>
<P>1. Right now I believe the AMA rules are silent on whether / when to discard any judge's score. It's in the FAI rules (I think), but not in AMA<BR>2. I agree wholeheartedly with Mark that we need to see and understand mathematically the effect of throwing out high and low scores, esp when using a small # of judges like 5. <BR>3. I suspect that the low judge is more often the correct one than the high judge. Especially in Finals situations it's possible to get into the mode of 'they're here, so they must be doing it all correctly' ... <BR>4. I'm not sure that throwing out ALL the high scores and/ or ALL the low scores will weed out / negate a biased judge's scores - whether we assume he's biased TOWARD one pilot or AGAINST one pilot, all we do by arbitrarily tossing out all highs or lows is to tend toward a mediocre judging result. We need to attack and kill bias through other
ways - training, coaching, etc rather than some mathematical method that won't eliminate his effect anyway...<BR><BR>My gut feeling on this is that if we throw away scores esp from small judging groups we're not only wasting a lot of guys' time making them sit in the chair for nothing, and we're driving AWAY from the stated goal of recognizing and rewarding perfection in model aerobatics - and instead, moving toward a system that trains us to give / receive average / mediocre scores. </P>
<P>Rick <BR>Just another pattern guy</P>
<P> </P></DIV>
<DIV></DIV>
<BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #a0c6e5 2px solid; MARGIN-RIGHT: 0px"><FONT style="FONT-SIZE: 11px; FONT-FAMILY: tahoma,sans-serif">
<HR color=#a0c6e5 SIZE=1>
<DIV></DIV>From: <I>"Arch Stafford" <rcpattern@comcast.net></I><BR>To: <I>"'Atwood, Mark'" <atwoodm@paragon-inc.com>,<discussion@nsrca.org></I><BR>Subject: <I>RE: Bad Rules Proposal</I><BR>Date: <I>Tue, 25 Oct 2005 09:34:45 -0400</I><BR>>The single major problem with this system is that now, ONE PERSON can decide<BR>>the entire contest. This realistically only affects a handful of people and<BR>>unfortunately, we are human. In the Masters finals specifically you end up<BR>>with people judging twice. Lets say one of those judges likes or dislikes a<BR>>person, now that person can score the one person he likes, 2 points higher<BR>>on every maneuver, while scoring every other person correctly. In a contest<BR>>where 5 positions were decided by 8 normalized points,
one person can decide<BR>>those positions. Bias does exist, right, wrong, intentional, unintentional,<BR>>it doesn't matter. I would rather have 3 out of 5 judges in the middle get<BR>>it right, than 4 out of 5, with the 5th one WAY out now totally affecting<BR>>the contest. I will give you the one judge that may get a snap/spin correct<BR>>on a zero, and the others don't, fine...but what about the guy who gives<BR>>10's just because he wants to. I have no problems with a guy that scores<BR>>everyone consistently, but if you have one judge who does play favorites one<BR>>time, then he has decided the entire contest by himself.<BR>><BR>>Arch<BR>><BR>><BR>>-----Original Message-----<BR>>From: Atwood, Mark [mailto:atwoodm@paragon-inc.com]<BR>>Sent: Monday, October 24, 2005 10:55 PM<BR>>To: Arch
Stafford; discussion@nsrca.org<BR>>Subject: RE: Bad Rules Proposal<BR>><BR>>I'm interested in hearing more on this....<BR>><BR>>Let me express my thoughts which are very subject to influence.<BR>><BR>>I've heard this argument before...that we must drop the highs and lows<BR>>to eliminate potential bias and to help offset the occasional "Snap<BR>>Nazi".<BR>><BR>>First, I'd like to express that dropping both high and low score on a<BR>>panel of 5 judges is statistically absurd (from a math perspective).<BR>>Even with the old 10-12 judge format it was dicey, but clearly more<BR>>effective. That being said, I would pose the argument that dropping<BR>>those scores does more harm than good.<BR>><BR>>In the case of the "biased" judge...the assumption is that he/she would<BR>>be the high or low...which is unlikely to be the
case. There IS<BR>>typically a high judge...averaging 9's in the finals...and a low judge<BR>>(not sure where that average falls but it's probably down in the 7.5<BR>>range for a finals flight... Regardless...if it any of the 3 lower<BR>>scoring judges that's biased in favor of a flight...adding .5 pt or even<BR>>a full point to every score still won't get his scores thrown out. What<BR>>you do successfully accomplish is tossing out the same 2 judges...for<BR>>virtually every flyer...therefore completely wasting their time AND<BR>>defeating the purpose of having 5 judges in the first place.<BR>><BR>>As for the snap nazi problem...I've seen far more examples of someone<BR>>getting scores from 4 judges and a zero from one...for doing the wrong<BR>>maneuver (i.e only one judge got it right) than I have the other
way<BR>>around. Dropping scores would only serve to squelch the good judge in<BR>>that scenario.<BR>><BR>>This IS a problem...the zero's on Snaps and Spins especially. I've<BR>>become aware that some see zeroing a snap as an easy, uncontested way to<BR>>ruin someone's flight. Pretty petty...but it seems to be the new sheik<BR>>form of bias. Personally...I'd LOVE to see someone propose getting rid<BR>>of the "ZERO" criteria since we can't seem to agree on it at all and<BR>>simply make poor snaps and poor spin entry's a downgrade...it would<BR>>eliminate a lot of bickering...but that's another subject.<BR>><BR>>As for the dropping of highs and lows...I would like to hear a<BR>>compelling and statistically relevant argument (and yes...I would like<BR>>to see the math) that shows how this helps our judging
with anything<BR>>less than 12 judges.<BR>><BR>>-Mark<BR>>AMA Dist III Board member<BR>><BR>>-----Original Message-----<BR>>From: discussion-request@nsrca.org [mailto:discussion-request@nsrca.org]<BR>>On Behalf Of Arch Stafford<BR>>Sent: Monday, October 24, 2005 10:18 PM<BR>>To: discussion@nsrca.org<BR>>Subject: Bad Rules Proposal<BR>><BR>>Just wanted to throw out a new rules proposal that I saw recently. I<BR>>have<BR>>no issues with any part it except the part that they use all judges.<BR>>This<BR>>is ridiculous in the finals.there is no reason to not drop the high and<BR>>low<BR>>scores from each maneuver. This is the way the proposal was submitted.<BR>>It<BR>>can't be changed, it just has to be voted on. This realistically only<BR>>affects a small number of people. In my
opinion we need to make sure<BR>>that<BR>>we contact the competition board to vote this down. A lot of these<BR>>rules<BR>>just get voted in by people that it doesn't affect. I hope people will<BR>>contact the competition board in your AMA district to vote this down,<BR>><BR>><BR>><BR>><BR>><BR>>Thanks,<BR>><BR>><BR>><BR>>Arch<BR>><BR>><BR>><BR>>RCA-07-15 - Insert in amend AMA rule handbook page 72, under section 13:<BR>>Determining the winner.<BR>><BR>><BR>><BR>>Proposed insert at the end of 13.1: For all AMA classes all judges'<BR>>scores<BR>>are to be included in the tabulation of scores regardless of the number<BR>>of<BR>>judges used in a normal, matrix, or finals round.<BR>><BR>><BR>><BR>>If you have concerns either way on this proposal, you should
contact<BR>>your<BR>>AMA Contest Board representative and let him know about it. I've<BR>>included<BR>>the whole proposal below. BTW, all the proposals are available on the<BR>>AMA<BR>>site in the Competition section.<BR>><BR>><BR>><BR>>RCA-07-15 - Insert in amend AMA rule handbook page 72, under section 13:<BR>>Determining the winner.<BR>><BR>><BR>><BR>>Proposed insert at the end of 13.1: For all AMA classes all judges'<BR>>scores<BR>>are to be included in the tabulation of scores regardless of the number<BR>>of<BR>>judges used in a normal, matrix, or finals round.<BR>><BR>><BR>><BR>>Proposed insert new 13.2:<BR>><BR>><BR>><BR>>At large contests such as a National level contest, the number of<BR>>contestants may exceed the time available to run a complete round in<BR>>front<BR>>of the
same judges. In this case the D may elect to use the matrix<BR>>system<BR>>for the 6 round event.<BR>><BR>><BR>><BR>>The Matrix system is intended for use in situations where the number of<BR>>pilots exceeds that which can be run on 1 site, in front of 1 set of<BR>>judges,<BR>>and within the time limitations of the event. For example, the Masters<BR>>class at the NATS often falls into this category.<BR>><BR>><BR>><BR>>The explanation, construction and scoring instructions for the Matrix<BR>>system<BR>>are proposed in 13.3. Where possible and practical, each competitor<BR>>will<BR>>fly 6 matrix rounds. The best 4 of 6 rounds will be used to determine<BR>>the<BR>>finalists. If the event cannot run a finals round due to inclement<BR>>weather<BR>>etc., the best 4 of 6
rounds will be used to determine the winner.<BR>><BR>><BR>><BR>>Finals format - If a Finals event is included, the number of finalists<BR>>will<BR>>be 20% of the total or a practical number to match the time available.<BR>><BR>><BR>><BR>>The Finals format is also subject to the time available. The CD can opt<BR>>to<BR>>run a 4 round final, or a 3 rounds (or less) finals. If a 3 round final<BR>>is<BR>>selected one normalized preliminary overall position score will be<BR>>carried<BR>>over. To allow for weather problems the best 2 of 2, or 2 of 3 will<BR>>determine the winner. If a 4 round finals is chosen then there will be<BR>>no<BR>>carry over overall position scores. Again, to allow for weather issues<BR>>the<BR>>best 1 of 1, 2 of 2, 2 of 3 or 3 of 4 normalized finals
scores will<BR>>decide<BR>>the winner. Equal judging exposure will be applied and only completed<BR>>rounds will be counted in the final standings.<BR></FONT></BLOCKQUOTE></div></html>