Some observations on the effect of U.S. glider handicaps on SC Nationals

Some observations on the effects of the current Sports Class handicaps
on Sports Class Nationals results. I looked at the final results for
the SC Nats from 1997 to 2006 (all the ones available on the SSA site)

1997: 37 entrants.
Top ten gliders:
VentusC (2)
VentusB (1)
Mosquito (1)
ASW-20 (1)
Discus A (1)
LS-6C (2)
Pik-20 (1) 8th
LS-4 (1) 10th

Highest placing Club Class: 8th

1998: 45 entrants.
Top ten gliders:
ASW-27 (2)
ASW-24 (1)
ASH-25 (1)
ASW-20 (1)
Discus A (2)
VentusB (1)
Genesis (1)
LS-8 (1)

Highest placing Club Class: 23rd (Erik Mann, LS-4)


1999: 30 entrants.
Top ten gliders:
ASW-27 (3)
ASW-24 (1)
ASH-25 (1)
Discus (1)
VentusCB (1)
Ventus2B (1)
Nimbus4 (1)
LS-6 (1)

Highest placing Club Class: 13th (Erik Mann, LS-4)


2000 (Ephrata, Wa): 26 entrants.
Top ten gliders:
LS-8 (2)
LS-6 (1)
ASW-24 (1)
Mosquito (1)
DiscusA (1)
VentusB,C (3)
Ventus2A (1)

Highest placing Club Class: None


2001 (Montague, Ca): 23 entrants.
Top ten gliders:
LS-8 (2)
LS-6 (1)
ASW-24 (1)
HP-18 (1)
Ka6E (1)
VentusA,B,C (3)
Ventus2A (1)

Highest placing Club Class: 7th (Scott Gradwell, Ka6E)


2002 (Lubbock, Tx): 44 entrants.*
Top ten gliders:
LS-3 (1)
Libelle (1)
ASW-20 (2)
Mosquito (1)
Discus2B (1)
VentusA,B (2)
SZF24 Foka (1)
Mosquito (1)

Highest placing Club Class: 1st (Dave Stevenson, SZF24 Foka)


2003 (Elmira, NY): 44 entrants.**
Top ten gliders:
LS-3 (1)
Libelle (1)
ASW-27 (1)
ASW-24 (1)
DiscusA (2)
Discus2A,2B (2)
Ventus2A,B,C (2)

Highest placing Club Class: 1st (Tim McAllister, Libelle)


2004 (Ionia, Mi): 28 entrants.
Top ten gliders:
LS-8 (1)
ASW-28 (1)
ASW-27 (1)
ASW-24 (1)
ASW-20 (1)
Ka6E (1)
Nimbus3 (1)
SZD-55 (1)
Ventus2B,C (2)

Highest placing Club Class: 3rd (Dave Stevenson, Ka6E)


2005 (Parowan, Ut): 48 entrants.
Top ten gliders:
LS-6B (1)
LS-3 (1)
Duo Discus (2)
ASW-27B (3)
Ventus2B,C (2)
Discus2A (1)

Highest placing Club Class: 6th (Manfred Franke, LS-3)


2006 (Mifflin, Pa): 49 entrants.
Top ten gliders:
LS-8 (2)
Nimbus3 (1)
Duo Discus (1)
ASW-27 (2)
ASW-20 (1)
SZD-55 (1)
Ventus2B,C (1)
Discus2A (1)

Highest placing Club Class: 34th (Tim Welles, Std Cirrus)


So, in the last 10 SC Nats, there have been exactly two contests in
which a 'Club Class' ship has won the SC Nats.

In 2002 at Lubbock, Tx, Dale Stevenson won in a SZF24 Foka. That year,
there were very few top line ships entered because the Std & 15m Nats
dates were very close to the SC Nats so they had to choose one or the
other. The 18m Nats coincided with the SC Nats, so they had to choose
as well. Consequently, this one time in the last 10 years, we actually
had a SC Nats attended mostly by SC pilots and SC ships.

In 2003 at Elmira, Tim McAllister won in his Libelle. Again with
apologies in advance to Tim who flew a great meet, I point out that the
average daily score of the winner was less than 500 points, meaning
that on average, each day was severely devalued due to the number of
landouts and less-than-min-distance flights. In fact, there were no
1000 point days, and on at least one day, the top score was something
like 250!

From the above, I conclude the following: If you are a Sports Class
pilot with a Club Class ship, your chances for doing well in a SC Nats
depend on two factors. First, if the SC Nats conflict with the 15m,
18m and Std Nats, you might stand a chance. Secondly, if you can
arrange the weather so that at least half the fleet lands out every
day, then you might stand a chance. Otherwise, forget it.

The LS-4, by some estimates the most popular CC ship in the world has
only been in the top ten once (10th place overall), and that was 10
years ago. Manfred Franke, a world-class SC pilot by any measure, has
managed three top 10 placements in 10 years, but has never been a SC
Champion. Two of his three top 10 placements were in 2002 and 2003
(see above).

I'm not making these numbers up, they're right from the SSA site. IMHO
the actual record of the last ten years shows a consistent handicap
bias toward top line glass. I understand that a lot of work and
analysis went into generating the current handicap numbers, and I
appreciate that. However, I think it might be time to revisit these
numbers in the light of actual performance.

Just my $0.02,

Frank (X3)

The trouble with looking at the glider is that it doesn't consider the
pilot. In the US, the vast majority of top-level pilots own
recent-generation gliders. When they go to a sports class event,
usually because it's fun or geographically convenient, they bring their
gliders with them. This analysis assumes that pilots are randomly
assigned to gliders, which isn't the case.

Try this: relate the position in the scoresheet to the glider handicap
and to the ranking number of the pilot. I suspect you'll find ranking
of pilot a far stronger determinant of who wins than the handicap of
the glider. Or try rescoring the race with any handicaps you like.
Alas, you will find the best pilots still winning. The handicaps
cannot be off by more than a few percent. The margin between first and
(say) 10th is much greater than that.

John Cochrane BB

BB wrote:
The trouble with looking at the glider is that it doesn't consider the
pilot. In the US, the vast majority of top-level pilots own
recent-generation gliders. When they go to a sports class event,
usually because it's fun or geographically convenient, they bring their
gliders with them. This analysis assumes that pilots are randomly
assigned to gliders, which isn't the case.


John beat me to the punch. I'd suggest we ask our friends in Europe
to give us some insights on pilots vs. ships. For instance, I know
from reading S&G that several of the top juniors and young guns in
England show up on the scoresheet for various national races -
standard, 15M, Club. My assumption is that: a) distances to travel
are significantly less, so it is feasible for someone to compete in
more than one nationals and b) the prevalence of club class gliders in
large syndicates or clubs (vs. the US penchant for individual
ownership) means it is much easier to line up a Club Class ship if you
want one.

So, let's model this in the US. Suppose we ran a true Club Class
nationals in an accessible location which did not conflict with any of
the other FAI class nationals. Would top-seeded pilots show up? I
posit that the answer is a qualified "yes"; some significant number
would. They would do what Tim Welles did this year and beg, borrow,
or steal one. Is that 5? 10? Dunno. But, getting back to the
lament about "club class gliders doing well", I think the answer is
that the "gliders" would do well in direct proportion to the top pilots
flying them.

Just a case in point: My performance in my LS4 in 1998 and 1999 vs. my
performance in 2006 in LS8 was much less a function of aircraft than it
was of piloting skill. I've now flown 3 Sports Nationals, improving
my placing in each. Along the way, I happened to switch to an LS8, but
trust me, that had nothing to do with getting into the top tier. I'm
a low learner, and in the first 10-12 or so years of racing, I could be
counted upon to make at least 3 really stupid mistakes per contest and
a dozen sort of stupid mistakes.

Going back to my tennis analogy from earlier. I finally got so fed up
with one guy who came back a dozen times to have me restring his new
racket (ie. "the equipment is the problem"), I made a deal. If I
beat him using my old Wilson Stan Smith (wood racket = club class) vs.
his Wilson Hammer (graphite aero = ASW-27) with him spotting me 1 point
per game (handicap), then he would sign up for a series of six lessons
(shameless commercial ploy). 6-2, 6-0. Hey, I was 21, and 6
lessons at 50 bucks a pop bought me a season's worth of aero tows!

The term
"geographically convenient"
keeps coming up for reasons to take a gun to a knife fight.

Parallel universe:
It's Geographically convenient for me to fly the Sierra Nevada and
White Mountains.
My passport says United Kingdom on the front.
If I fly long/fast/high from any of my home sites, it could be
certified as a British record.

This is wrong, too.
Jim

You could only claim a "British National" record, not a "UK Local" record.
That is what British National records are about. Many years ago Nick
Goodhart claimed a British National height record flying out of Bishop
California, and why not?

W.J. (Bill) Dean (U.K.).
Remove "ic" to reply.

"JS" wrote in message
ups.com...

The term
"geographically convenient"
keeps coming up for reasons to take a gun to a knife fight.

Parallel universe:
It's Geographically convenient for me to fly the Sierra Nevada and
White Mountains.
My passport says United Kingdom on the front.
If I fly long/fast/high from any of my home sites, it could be
certified as a British record.

This is wrong, too.
Jim

Listing the top ten scores and then listing the top club class finisher
in the field is not the best way to represent data, IMHO. A person more
skeptical as myself might say that you are pushing an "agenda". I've
looked at the data and come up with a different picture. The reason that
so few club class ships score well in sports class contests is that
there are so few club class ships entered.

Let's look at 2006, for instance (other years paint a similar picture).

2006 (Mifflin, Pa): 49 entrants.
Top ten gliders:
LS-8 (2)
Nimbus3 (1)
Duo Discus (1)
ASW-27 (2)
ASW-20 (1)
SZD-55 (1)
Ventus2B,C (1)
Discus2A (1)

Highest placing Club Class: 34th (Tim Welles, Std Cirrus)
(Tim finished 27th, BTW. And, IMHO, the highest placing club class
finisher was Rick Roelke [25th] in a 304CZ)

The gliders finished (in order): LS-8, Duo Discus, Ventus 2Bx, ASW-27,
ASW-27, LS-8, Nimbus 3DM, ASW-20, SZD-55-1, Discus 2B, Discus 2B, Ventus
2B, SZD-55-1, Ventus 2CxT, Discus 2B, Duo Discus T, Duo Discus, ASW-27,
ASW-28, ASW-27, Discus 2A, ASW-27, Discus 2B, ASW-20, DG-800S, ASW-20,
Ventus 2C, Discus 2B, 304CZ, ASW-24, Std Cirrus, Duo Discus, ASW-27,
ASW-27, ASW-27, LS-8, ASK-14, Discus CS, LS-8, H-201 Std Libelle,
Genesis II, Discus 2B DG1000, ASW-20, Ventus CM, ASW-27, ASW-20B, Nimbus
4, ASW-27, ASW-27, Discus CS, ASW-20C, ASW-15A, ASW-27, Ventus 2Bx

Am I surprised that a club class glider did not finish in the top 10?
Not at all, given that there were only a few entered. By all means, lets
revisit the handicap system periodically and correct when necessary. No
system is ever perfect.

Tony V. LS6-b "6N"

Tony Verhulst wrote:
Listing the top ten scores and then listing the top club class finisher
in the field is not the best way to represent data, IMHO. A person more
skeptical as myself might say that you are pushing an "agenda". I've
looked at the data and come up with a different picture. The reason that
so few club class ships score well in sports class contests is that
there are so few club class ships entered.

Assume that the Club Class ships at the 2006 Sports Class Nationals we

SZD-55-1
304CZ
Standard Cirrus
ASK-14
Libelle
ASW-15

If I have added up the numbers correctly, the Club Class ships had an
average standing of 30th, while the other ships had an average standing
of 25th. Include the SZD-55-1 flown by a guest, and the average Club
Class standing is 27th. Eliminate the SZD-55-1 entirely, and the
average Club Class standing is 33th. Also eliminate the "outlier" (the
ASW-15 that came in 48th), and the average Club Class standing is 30th.

This suggests that the newer ships don't have much of an advantage.
When you take into account that most of the top pilots were not flying
Club Class ships, the Club Class ships appear fairly competitive.

Tony Verhulst wrote:
Listing the top ten scores and then listing the top club class finisher
in the field is not the best way to represent data, IMHO. A person more
skeptical as myself might say that you are pushing an "agenda". I've
looked at the data and come up with a different picture. The reason that
so few club class ships score well in sports class contests is that
there are so few club class ships entered.

Zounds! I'm mortally wounded by your accusation that I, an owner of a
LS4 and avid Sports Class competitor, have allowed bias to creep into
my otherwise completely objective analysis ;-). Well, actually, I had
no real idea how to look at the data - I just thought that if I
actually looked, something might leap out at me. Something did, but
it wasn't what I expected. What got my attention is the two times SC
pilots did well, and what it took for that to happen.

You and the others are of course right that the skill of the pilot has
a lot to do with who is, or is not, in the top of the scoresheet. In
theory, if the handicaps are done correctly, equally skilled pilots
would score equally, regardless of what ship each was flying. All I
was trying to do with the data is to point out that that doesn't appear
to be the case. A very highly skilled pilot (Manfred Franke) in a
very well-prepared LS3 consistently scored lower than very highly
skilled pilots in ASW27s (or Ventus2s or D2s, etc). To borrow P3's
analogy for a moment, I believe the data shows that the guy with the
wooden racket consistently lost, even though a handicap system was
devised to equalize the performance difference between the wooden and
graphite rackets. Either he wasn't that good to begin with, or there
is something wrong with the handicaps.

I think you either have to agree that either the handicaps are somewhat
biased in favor of the ASW27 crowd, or that Manfred really isn't in the
top rank of pilots. Maybe there are so many better ASW27 pilots that
there will always be several at each SC Nats. This may in fact be the
case, but how to tell for sure?. One way would be to have the ASW27
pilots fly LS3s, and that (I think) is what the Club Class does in
Europe.

In closing, I offer a very tongue-in-cheek suggestion for a definitive
experiment: P3 and HF arrange to swap gliders for the entire 2007
season, and fly SC in several of the same regionals and the 2007 SC
Nats at CCSC. If HF wins, P3 has to publicly state that the current
handicap system is unfairly biased toward top-line glass. If P3 wins,
I'll publicly accept the current handicap system as fair and equitable,
and rabble-rouse no more (at least on this subject). Start the
towplanes, and may the best pilot win! ;-).

Frank(X3)

Frank wrote:

I think you either have to agree that either the handicaps are somewhat
biased in favor of the ASW27 crowd, or that Manfred really isn't in the
top rank of pilots. Maybe there are so many better ASW27 pilots that
there will always be several at each SC Nats. This may in fact be the
case, but how to tell for sure?. One way would be to have the ASW27
pilots fly LS3s, and that (I think) is what the Club Class does in
Europe.


Frank(X3)

Frank,

I think the statisticians would say that you are running into the "rule
of small numbers"; ie. with such a small sample it's hard to make any
reliable generalizations. To put it a different way, how many pilots
are capable of winning a nationals at any given contest (leaving aside
ships for a minute)? Realistically, maybe 1/3 of the competitors are
truly national caliber; the others are there to prop up the bottom of
the scoresheet. I'm sure others would argue that the number of
potential champions is lower .

So, let's say that there are 10 folks on average in a SC nationals who
have the potential to win. So, the unadjusted probability is 10% that
someone like Manfred will win under this scenario. Now, throw in the
fact that certain guys seem to win contests again and again (I'll use
Gary Ittner as an example in the 15M, flying his "old" Ventus C against
all of the new ASW-27s and V2s as a for instance). So, the
probability for the "rest of us" is really pretty small. The fact
that 2 true Club Class ships have won in the last 10 years means that
they are statistically over-represented and the handicaps need to be
revised in favor of the more modern ships :-))))

Seriously, I think a couple of other posts all alluded to the fact that
there are some good reasons why the current handicap system in the US
is a compromise. If you really want to get under the numbers, create a
simple model in a spreadsheet. Under average Eastern conditions (say a
course of 150 miles with 300fpm average lift), you can calculate the
time to climb and the time to cruise using the published polars. The
7% handicap (actually more like 7.5%) the LS3 has over the ASW-27 is
VERY significant. However, get the cruise speeds up with either
ridges or strong thermals, and the newer gliders have an advantage.
I thought the South African model described earlier is interesting, as
it tries to use the actual results to deterimine the day's handicap.
On the other hand, some folks will find it difficult to accept that
the results are adjusted so dynamically...

As far as the LS-3 swap, I'll take a different tack. I'll see what I
can do to line up an LS-4 to go head-to-head with you. Loser each day
buys the beer :-))

P3

I think the statisticians would say that you are running into the "rule
of small numbers"; ie. with such a small sample it's hard to make any
reliable generalizations. To put it a different way, how many pilots
are capable of winning a nationals at any given contest (leaving aside
ships for a minute)? Realistically, maybe 1/3 of the competitors are
truly national caliber; the others are there to prop up the bottom of
the scoresheet. I'm sure others would argue that the number of
potential champions is lower .

So, let's say that there are 10 folks on average in a SC nationals who
have the potential to win. So, the unadjusted probability is 10% that
someone like Manfred will win under this scenario. Now, throw in the
fact that certain guys seem to win contests again and again (I'll use
Gary Ittner as an example in the 15M, flying his "old" Ventus C against
all of the new ASW-27s and V2s as a for instance). So, the
probability for the "rest of us" is really pretty small. The fact
that 2 true Club Class ships have won in the last 10 years means that
they are statistically over-represented and the handicaps need to be
revised in favor of the more modern ships :-))))

Well, as my Mammy used to say, "Be careful what you wish for, cuz you
just might git it!". I was hoping that my rabble-rousing would generate
some thoughtful discussion, and apparently I'm getting my wish! ;-).

Yes, I agree that my data base is woefully small. However, more than
about 10 years of data kinda defeats the purpose of the argument
anyway, as the gliders and the pilots will have evolved significantly
over that time . Actually it was just this paucity of data that caused
me to focus on HF as the poster child for SC (my apologies to Manfred),
as this is the clearest case I can see where a very good pilot in an SC
ship has difficulty competing against very good pilots in ASW-XXs
(where X is greater than 20 or so). BTW, regarding the probabilities
you mentioned, it would seem that if Manfred's per-contest probability
was on the order of 0.1 and he competed in more than 5 or 6 nats, he
should have (in the sense that anything over 0.5 is 'likely') won one
by now.

Seriously, I think a couple of other posts all alluded to the fact that
there are some good reasons why the current handicap system in the US
is a compromise. If you really want to get under the numbers, create a
simple model in a spreadsheet. Under average Eastern conditions (say a
course of 150 miles with 300fpm average lift), you can calculate the
time to climb and the time to cruise using the published polars. The
7% handicap (actually more like 7.5%) the LS3 has over the ASW-27 is
VERY significant. However, get the cruise speeds up with either
ridges or strong thermals, and the newer gliders have an advantage.

I have a basic problem with the above modeling approach, and that is
that I find it difficult to believe that a 25 year old glider can still
achieve its published polar (assuming it ever did), no matter how well
prepared. I borrow from my power background, where aircraft
performance numbers are routinely inflated by the manufacturers and are
just as routinely missed by the actual aircraft. However, newer
aircraft do come closer than older ones, pretty much regardless of
type. Having said that, I really don't know if Manfred's LS-3 can
still achieve its numbers (maybe it even exceeds them now!).

I thought the South African model described earlier is interesting, as
it tries to use the actual results to deterimine the day's handicap.
On the other hand, some folks will find it difficult to accept that
the results are adjusted so dynamically...

Ah, the handicapped bowling league approach. I actually think this
might be a viable approach, in that it tends to make it easier for a
newcomer to do well (but, maybe this is yet another thing I should be
careful about wishing for...)



As far as the LS-3 swap, I'll take a different tack. I'll see what I
can do to line up an LS-4 to go head-to-head with you. Loser each day
buys the beer :-))

Oops! You have clearly mistaken me for someone willing to put his
glider (or at least his beer) where his mouth is! ;-). However, given
the fact that I will almost certainly never be able to afford anything
better, and that I do plan to fly it in our SC Nats next year - bring
it on! ;-))))

Frank (X3)