US contest scoring formula is broken

One thing that soaring
should do in my opinion is change the time cycle for rule changes.
Keep the rules constant for 3 or 4 years, then change them. *That
would reduce the constant whiplash of new rules every year.


So, if you see a really awful rule, a dangerous rule, or even a typo
or clear mistake (it happens), we should be forbidden to change it for
three years? That makes no sense at all.

And speaking of complication, now every single rule would need a date
by it -- "rule 10.2.3.5 (no change before 2011) ..." Or maybe you
want a new rule specifying "major changes" vs. "typos" which have
different waiting periods. And a new commitee to monitor all the
different waiting periods....

What rules change from the past several years would you really have
liked to force everyone to wait for? Would it have been good to force
start anywhere to wait three years just to watch the clock run out?
How about the new finish, which prohibits thermaling under the finish
gate or big zoomies right in to the belly of another glider? Should we
have watched people do that for three years just for funzies? How
about the new rule giving 600 points for distance rather than 400?
Would it have been good to just sit for 3 years? Why?

The RC does keep changes to a minimum, there is the cycle of "try it
in regionals first" which catches major glitches, and an extensive
"get the word out" program.

John Cochrane

On Jun 24, 8:01*pm, John Cochrane
wrote:
*One thing that soaring
should do in my opinion is change the time cycle for rule changes.
Keep the rules constant for 3 or 4 years, then change them. *That
would reduce the constant whiplash of new rules every year.

So, if you see a really awful rule, a dangerous rule, or even a typo
or clear mistake (it happens), we should be forbidden to change it for
three years? That makes no sense at all.

And speaking of complication, now every single rule would need a date
by it -- "rule 10.2.3.5 (no change before *2011) ..." Or maybe you
want a new rule specifying "major changes" vs. "typos" which have
different waiting periods. And a new commitee to monitor all the
different waiting periods....

What rules change from the past several years would you really have
liked to force everyone to wait for? Would it have been good to force
start anywhere to wait three years just to watch the clock run out?
How about the new finish, which prohibits thermaling under the finish
gate or big zoomies right in to the belly of another glider? Should we
have watched people do that for three years just for funzies? How
about the new rule giving 600 points for distance rather than 400?
Would it have been good to just sit for 3 years? *Why?

The RC does keep changes to a minimum, *there is *the cycle of "try it
in regionals first" which catches major glitches, and an extensive
"get the word out" program.

John Cochrane

Typo's or clear mistakes could be fixed, but nothing else. I think we
have the really dangerous rules gone, we have been tinkering with
these rules for a long time now.

The rules right now are pretty fair and pretty safe. Leave them alone
for a while. And no, you do not need a date on each rule. You
publish the rules for 2009 - 2012 and leave them alone. Then in 2013
you get to publish a new set of rules. In the meantime, everybody
gets to just fly without trying to figure out the latest rule changes.

By the way, quit using the Regional pilots are guinea pigs.

Todd Smith
3S

I've had disagreements with the Rules Committe over the years (no
comments, guys). But not about the scoring formulas per se. I do think
they're way too complicated for most pilots to understand but UH's
advice is good: just fly the glider. There are very few situations
where doing something cute or counterintuitive just to enhance your
score will benefit you (the airport landing bonus is one of the few
examples, and I'm sure I'll get pushback on that). I flew many years
with ONLY ASTs and definitely don't want to go back! I don't like it
when CDs call MATs on days when the weather is fairly uniform and
predictable--it's lazy--but on risky days, I like having the option of
going where it's soarable rather than flying directly into a T-storm
over the TP.

My primary issue is with the way the Rules change every year and the
impact this has on the scoring program. The more changes you make to
the scoring formulas, the more chances there are for bugs to creep
into the scoring software. And most pilots will never pick this up
unless the errors are substantial.

And right now the burden is on the pilot to protest any scoring
problems, whether they relate to the scorer's mistakes or to the
software itself. The only real way to do that is to take Winscore with
you to the contest and score side by side with the scorer. Winscore
does a good job, but it's not perfect. It's process dependent (read:
the scorer has to do some manual stuff). And it sometimes has bugs.
When a bunch of us had questions about the scores at a regional
contest last year, I installed Winscore and rescored the contest and
found situations where the scores were wrong based on the rules. Yeah,
a savvy scorer might have caught it, but no pilot would have. This
particular item didn't change the overall winners but it did change
the daily winners and overall placings. I strongly suspect it affected
other contest results last year, as well.

In the interests of fairness and responding to pilot requests, the
Rules Committee makes changes to the scoring nearly every year. I
would argue that they try too hard. Anyone who's ever been involved in
software development or the implementation or modification of a
software application knows that every single one of those changes has
the potential to introduce bugs. Yeah, most of them get caught during
testing, but not all. It's not like there are millions of Winscore
users paying big bucks every year to compensate someone for making
certain it's 100% reliable. That said, we should be making Rules
changes only when there's a really good reason, and as infrequently as
possible.

The Rules Committee has done a good job of developing Rules that work
pretty well in most cases. Stop fooling around with them. Sometimes
trying to make something perfect just makes it worse.

Chip Bearden
ASW 24 "JB"
USA