On Sep 24, 1:00*am, John Smith wrote:
noel.wade wrote:
After all, the G-102 and Junior have much
better penetration and handicapping alone can't account for the
differences in their polars under various conditions!

How about a dynamic handicap? Using the logger files, you could evaluate
not only the wind profile, but also the strenght and diameter of
thermals and their separation, cloud base, etc. etc. and apply not a
handicap number, but rather a handycap matrix which is fair for all
gliders in all conditions. And don't forget to define a pilot factor as
well, otherwise the experienced ones would have an unfair advantage!

You know that you have your parameters and the formula correct when,
after normalizing the results, everybody gets 1000 points.

Interesting.

I thought about this idea a bit a couple of years back. I think it
would be possible to come up with something analytically on the basis
of pilot seeding, average climb strength, average altitude, wind and
maybe some task-specific attributes. You do have to be able to
separate out pilot skill from sailplane performance when you do it and
I suspect there is some autocorrelation between the two. Something
simpler might be increasing the spread in handicaps as task distance
increases, since task distance is a decent indicator for a lot of the
other factors you'd want to consider (except wind).

The problem, and the reason I dismissed the idea, is that it is
cumbersome, would potentially lead to a lot of arguing about how each
day gets measured and further obfuscates the scoring process -- not to
mention the burden on the scorer.

The other problem is that one of the issues for very low performance
ships in particular is the risk of outlanding - which goes up as the
tasks get more challenging. Because you are adjusting handicaps on the
basis of long-term averages, you would likely see lots of points
lavished on the pilots in low-performing ships just for getting around
the course on challenging days. As a consequence you could end up with
a handicap system that pushes pilots in low-performing ships to the
top of the scoresheet in the, say, 2 out of 5 contests where they can
get around the course every day, but finds them at the bottom of the
scoresheet in the contests where they have a landout. On average they
are in the middle, but they end up winning a disproportionate share of
contests. BB wrote a very interesting article on this in terms of
overall contest strategy irrespective of handicaps. The current
handicap system has a bit of this built in already - increasing the
spreads might make it worse rather than better.

The conclusion I came to is that the current system works well enough
for most Std and 15M ship of late-70s vintage on up. The Sports Class
is and should be optimized around a typical mid-80s 15M and Std
gliders (basically, Club Class, that is). If you want to fly
something outside these parameters you need to accept the fact that
the scoring system can't totally level the playing field for you, but
in the end you probably aren't expecting to be on the podium - and
that's okay. At least you still get to fly with everybody else and
learn about racing. I have a slightly different view on 2-seaters, but
that's a topic for a whole different thread.

The idea of restricting team selection to people flying Club Class
gliders only I think is a red herring because the scoring system and
contest rules are optimized around these gliders already so the odds
that someone flying an ASK-14 is going to get on the US team is
impossibly low (if they did they'd have earned it). The arguments
about low performing gliders dragging the tasking down should be
resolved through better training of CDs on how to call tasks and in
particular by realizing that you can't optimize task calling for the
lowest common denominator - that's not the main purpose of Sports
Class. If someone wants to fly a 2-33 in a Regionals they better have
a big crew and a trailer ready to go.

9B