<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.6000.16546" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial>Allow me to start a new thread for the purpose of
analyzing the issues Jim and others highlighted in the D3 thread. Most of
us have been frustrated over the years by inaccurate scoring, both high and
low. Forget the notion that it doesn't matter if an inaccurate judge is
consistent - that person is just consistently wrong and it does matter, the
rules require both accurate and consistent scoring. I also don't believe
that judging ability depends entirely on class flown, masters and FAI folks
aren't inherently smarter than others. Experience does improve accuracy and it's
important to know what maneuvers / schedule that will be judged (called
preparation). However, it's not important what class is flying the maneuver - a
half loop or immelman or stall turn is the same in intermediate as F3A. The
NSRCA Judge Cert program has improved the quality of judging immensely! So - now
that I've gotten these generalities out of my system, let's take a look at why
F3A scores may vary a lot by judge / region.</FONT></DIV>
<DIV><FONT face=Arial></FONT> </DIV>
<DIV><FONT face=Arial>I've heard the opinion expressed that the whole point
scoring in F3A dictates that no downgrade is applied until an error of 15 deg or
more is observed. Conversely, others feel that any error in F3A requires, at
least, a one point downgrade. Hmmmm - that'll make a difference! F3A
adopted whole point scoring in an effort to force judges to use the entire 10 -
0 range of scores, rather than the upper 3 or so as was typical. This is
probably where the 1pt for any error notion comes from. But it was
difficult to quantify how much to downgrade many errors, and a wide variation
occurred between judges of equal skill, some saw a 5 deg error worthy of a point
deduction - others would see a 30 deg error as worthy of only a point. Probably
the most useful metric available to judges is the 1/15 rule! </FONT></DIV>
<DIV><FONT face=Arial></FONT> </DIV>
<DIV><FONT face=Arial>However, the 1/15 rule fails to define just what should be
done for errors of less than 15 deg. Honest differences of opinion exist and
these become more important the better a maneuver is flown. I suspect that some
evaluation of the wording of the rules might help. F3A requires "marking"
(scoring) in whole points, but uses the word "downgrade" regarding the judges
assessment of the "mark". While examples of egregious errors are noted in whole
points, there is no exclusion forbidding the judge to use smaller downgrade
increments to arrive at the whole point score. So why is the downgrade for
errors smaller than 15 deg undefined? Well - pattern folks are certainly smart
(or we wouldn't be doing this - right?) and have no problem recognizing the
downgrade applicable to 30, 45, etc. errors basis the 1/15 metric, there
shouldn't be any difficulty in the other direction either as 5 deg = 1/3
pt, 7.5 = 1/2 pt, etc. </FONT></DIV>
<DIV><FONT face=Arial></FONT> </DIV>
<DIV><FONT face=Arial>A problem arises when a judge is between whole points with
the proper downgrades. (Things would be a lot easier if F3A adopted 1/2 pt
scoring - I've made the arguments and some are listening - but don't expect a
change any time soon.) The scenario might be a simple turnaround maneuver with a
slight 5 deg error of some sort which deserves a 1/3 pt downgrade. Some will
score this a 10, others a 9. The F3A rules dwell on major defects and leave
these situations nearly unaddressed. Consider that a 9 is unfair - might as well
make a 15 deg error. Some will say a 10 is unfair as the maneuver is imperfect
and we are striving for perfection. OK - the F3A rules state "<SPAN lang=EN-GB
style="FONT-SIZE: 10pt; FONT-FAMILY: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-GB; mso-fareast-language: FR; mso-bidi-language: AR-SA"><FONT
size=3>A high score should be given only if no major defects are found and the
maneuver is well positioned." <FONT face=Arial>You decide - I would probably go
with a 10, as there's no "major" defect, and feel comfortable rounding to the
nearest whole number. Unfortunately, unintentional bias (basis pilot reputation,
quality of current flight, etc.) can slip in here and result in like maneuvers
being rounded up and down for different pilots - here's a place where judge
consistency must be applied. </FONT></FONT></SPAN></FONT></DIV>
<DIV><FONT face=Arial><SPAN lang=EN-GB
style="FONT-SIZE: 10pt; FONT-FAMILY: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-GB; mso-fareast-language: FR; mso-bidi-language: AR-SA"><FONT
size=3><FONT face=Arial></FONT></FONT></SPAN></FONT> </DIV>
<DIV><FONT face=Arial><SPAN lang=EN-GB
style="FONT-SIZE: 10pt; FONT-FAMILY: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-GB; mso-fareast-language: FR; mso-bidi-language: AR-SA"><FONT><FONT
face=Arial><FONT size=3>Centering is another area where downgrades often vary in
that some focus on a center "key point" of a maneuver and downgrade heavily if
that point is missed. F3A rules state "<SPAN lang=EN-GB><FONT
face="Times New Roman">This may be in the range of 1 to 4 points
subtracted" </FONT><FONT face=Arial>with regards to centering errors,
without defining a metric. Most assume 1 pt / 25%. OK - the middle of the
inverted portion of a 4 pt roll is way off - the 90 deg roll ends at the pole -
most would ding this a couple of points. But, a 4 pt roll may be 1000 ft long,
so a 2 pt downgrade would be appropriate for a 500 ft error, the example is
probably less than 200 feet - so a 1pt downgrade would be more appropriate. If
the overall length of a maneuver is considered, centering downgrades are often
less than scored. This concept is even more important for "narrow" center
maneuvers.</FONT></SPAN></FONT></FONT></FONT></SPAN></FONT></DIV>
<DIV><FONT face=Arial><SPAN lang=EN-GB
style="FONT-SIZE: 10pt; FONT-FAMILY: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-GB; mso-fareast-language: FR; mso-bidi-language: AR-SA"><FONT><FONT
face=Arial><FONT size=3><SPAN
lang=EN-GB></SPAN></FONT></FONT></FONT></SPAN></FONT> </DIV>
<DIV><FONT face=Arial><SPAN lang=EN-GB
style="FONT-SIZE: 10pt; FONT-FAMILY: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-GB; mso-fareast-language: FR; mso-bidi-language: AR-SA"><FONT><FONT
face=Arial><FONT size=3><SPAN lang=EN-GB>So - there are a couple of places where
judges can disagree in scoring and these will generate large differences in
scores by flight and, possibly, by region. Our NSRCA judging program has done
and is doing a good job of ensuring we all recognize errors. The 1/15 rule
provides a good metric and works well with the AMA 1/2 point scoring
system. This same 1/15 metric leaves us hanging a bit in F3A when used
with whole point scoring. Maybe a solution is for us to use 1/2 pt. scoring in
F3A events in AMA contests. Certainly we might include direction in our judging
program to ensure folks judging F3A handle this issue
consistently.</SPAN></FONT></FONT></FONT></SPAN></FONT></DIV>
<DIV><FONT face=Arial><SPAN lang=EN-GB
style="FONT-SIZE: 10pt; FONT-FAMILY: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-GB; mso-fareast-language: FR; mso-bidi-language: AR-SA"><FONT><FONT
face=Arial><FONT size=3><SPAN
lang=EN-GB></SPAN></FONT></FONT></FONT></SPAN></FONT> </DIV>
<DIV><FONT face=Arial><SPAN lang=EN-GB
style="FONT-SIZE: 10pt; FONT-FAMILY: 'Times New Roman','serif'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-GB; mso-fareast-language: FR; mso-bidi-language: AR-SA"><FONT><FONT
face=Arial><FONT size=3><SPAN
lang=EN-GB>Earl</SPAN></FONT></FONT></FONT></SPAN></FONT></DIV></BODY></HTML>