2009 Proposed US Contest Rules Changes

Do the benefits of "start anywhere" outweigh the disadvantages?

On Jan 10, 2:04*pm, ZL wrote:
Section 11.2.3.2 has the Regional start anywhere rule for 2008. Thats
the scoring rules section. They missed that it was in two places.


thanks

Andy

On Jan 11, 6:31*am, PMSC Member wrote:
Do the benefits of "start anywhere" outweigh the disadvantages?

My impression is that most pilots who have flown with start anywhere
hae liked it. The discussion here has been mostly about the proposed
modification that the RC is proposing before the rule goes into effect
for Nationals in 2009. The focus of the discussion is whether the
change addresses a real problem.

I took a look at the first day at Region 9 in 2008 to try to
understand at least one example of pilot behavior under the start
anywhere rule.

In Std, 15M and and 18M (all using the same start cylinder) 29 pilots
started through the top of the cylinder, 5 started out the side -
mostly close to the courseline. Roughly 2/3 of the pilots started in
the front half of the cylinder, but starts were distributed all over
the cylinder with the largest "gaggle" of starters comprised of 4
gliders.

Climb rates for out the top starters were as follows: 6 at1.8-3.9kts,
9 at 4-4.9 kts, 5 at 5-6 kts, 5 at 6-7 kts and 4 at 7-8 kts. A number
of the start climbs were for more than 4,000' gains. There appeared to
be pretty good correlation between higher climb rates at the start and
pilots at the top of the scoresheet at the end of the contest, so
perhaps the difference in climb is more a result of strategy than
luck. I'll take a look at some more contest days time permitting.

Overall there's nothing of concern and some potential strategic
insights that might be useful here.

9B

On Jan 11, 12:35*pm, wrote:
On Jan 11, 6:31*am, PMSC Member wrote:

Do the benefits of "start anywhere" outweigh the disadvantages?

My impression is that most pilots who have flown with start anywhere
hae liked it. The discussion here has been mostly about the proposed
modification that the RC is proposing before the rule goes into effect
for Nationals in 2009. The focus of the discussion is whether the
change addresses a real problem.

I took a look at the first day at Region 9 in 2008 to try to
understand at least one example of pilot behavior under the start
anywhere rule.

In Std, 15M and and 18M (all using the same start cylinder) 29 pilots
started through the top of the cylinder, 5 started out the side -
mostly close to the courseline. Roughly 2/3 of the pilots started in
the front half of the cylinder, but starts were distributed all over
the cylinder with the largest "gaggle" of starters comprised of 4
gliders.

Climb rates for out the top starters were as follows: 6 at1.8-3.9kts,
9 at 4-4.9 kts, 5 at 5-6 kts, 5 at 6-7 kts and 4 at 7-8 kts. A number
of the start climbs were for more than 4,000' gains. There appeared to
be pretty good correlation between higher climb rates at the start and
pilots at the top of the scoresheet at the end of the contest, so
perhaps the difference in climb is more a result of strategy than
luck. I'll take a look at some more contest days time permitting.

Overall there's nothing of concern and some potential strategic
insights that might be useful here.

9B

That leads to what might not be a bad idea. I would think it would be
fairly easy to have Winscore compile a list of start locations for
each day. If the rules committee or even the contest organizers had
this information it might be easier to anaylize how people are
actually starting.

Brian

On Jan 6, 4:25*pm, Andy wrote:

In a simple graphic analysis of the defined task I estimated a
distance of over 8 miles of start cylinder circumference between the
intersections of the two extreme case arcs and the start circle. *In
other words, the potential front half circumference of approx 31.4
miles is reduced to a no risk arc of approx 23.4 miles.

Don't want to re-open this but just in case someone reads it later I
need to correct the numbers. I should have said:

"The potential front half circumference of 16.7 miles is reduced to a
no risk arc of approx 8.7 miles."

My error was caused by starting with the full circle circumference
instead of the half circle circumference. The numbers were fine tuned
by 9B later in the discussion.

Andy

On Jan 12, 3:45*pm, Andy wrote:

"The potential front half circumference of 16.7 miles is reduced to a
no risk arc of approx 8.7 miles."

I'm digging myself a big hole here, but did say I was math averse!

Should be 15.7 and 7.7 of course.

On Jan 11, 11:35*am, wrote:
On Jan 11, 6:31*am, PMSC Member wrote:

Do the benefits of "start anywhere" outweigh the disadvantages?

My impression is that most pilots who have flown with start anywhere
hae liked it. The discussion here has been mostly about the proposed
modification that the RC is proposing before the rule goes into effect
for Nationals in 2009. The focus of the discussion is whether the
change addresses a real problem.

I took a look at the first day at Region 9 in 2008 to try to
understand at least one example of pilot behavior under the start
anywhere rule.

In Std, 15M and and 18M (all using the same start cylinder) 29 pilots
started through the top of the cylinder, 5 started out the side -
mostly close to the courseline. Roughly 2/3 of the pilots started in
the front half of the cylinder, but starts were distributed all over
the cylinder with the largest "gaggle" of starters comprised of 4
gliders.

Climb rates for out the top starters were as follows: 6 at1.8-3.9kts,
9 at 4-4.9 kts, 5 at 5-6 kts, 5 at 6-7 kts and 4 at 7-8 kts. A number
of the start climbs were for more than 4,000' gains. There appeared to
be pretty good correlation between higher climb rates at the start and
pilots at the top of the scoresheet at the end of the contest, so
perhaps the difference in climb is more a result of strategy than
luck. I'll take a look at some more contest days time permitting.


I took a look at a second day at R9 in 2008 and a day from 2007.
Disclaimer: three contest days is a small sample and all three days
were flown at Parowan under conditions that allowed starts through the
top of the cylinder.

Some interesting (to me) observations:

Under the start anywhere rule the percent of pilots starting through
the top increased to 85% from 60%. Under the old rule most out the top
starts were within a mile or two of where the edge of the cylinder
meets the first leg courseline. Under the new rule starts were broadly
distributed, with a slight bias toward the front half of the cylinder.
As you'd expect, spreading out starts makes for smaller gaggles -
under the old rule it appeard there were at any given time 2-3 big
gaggles with 5-10 gliders in them - the biggest gaggle being at the
edge of the cylinder on courseline.

Now the really interesting part (repeat disclaimer here). Under the
old rules starters out the top had a slightly faster average climb
that starters out the side - less than half a knot. Also, there was
less scatter in out the top climb rates than initial climbs after
starting out the side - this stands to reason - under the old rules
climbing out the top was normally crammed together into a couple of
gaggles.

Under the new rules the opposite is true. Starters out the top had
nearly a knot slower initial climb on average than the admittedly
small number of edge of cylinder starts. The scatter (as measured by
standard deviation as a percent of average climb rate) was about 50
percent higher for starts out the top.

All of this stands to reason in my rationalizing mind. If you have
more room inside the cylinder to look there is some likelihood that a
few pilots will hit boomers and get an advantage, but the lower
average climb rate for starts out the top also indicates that perhaps
pilots got a bit fixated on starting this way - climbing out the top
at 3 knots when there are 10-knotters about is not a great decision.

All of the above is very rough and can be rightly criticized for a
host of reasons. On thing that is clear is that the first climb out of
the start is not the whole story - pilots with weaker thermals tend to
leave them sooner, so you really need to get an aggregate view of
average climb rate to the top of the lift band - even if it's spread
over 2-3 thermals. That's too complex an exercise to do with SeeYou
and a spreadsheet.

9B

On Jan 13, 6:41*pm, wrote:
On Jan 11, 11:35*am, wrote:





On Jan 11, 6:31*am, PMSC Member wrote:

Do the benefits of "start anywhere" outweigh the disadvantages?

My impression is that most pilots who have flown with start anywhere
hae liked it. The discussion here has been mostly about the proposed
modification that the RC is proposing before the rule goes into effect
for Nationals in 2009. The focus of the discussion is whether the
change addresses a real problem.

I took a look at the first day at Region 9 in 2008 to try to
understand at least one example of pilot behavior under the start
anywhere rule.

In Std, 15M and and 18M (all using the same start cylinder) 29 pilots
started through the top of the cylinder, 5 started out the side -
mostly close to the courseline. Roughly 2/3 of the pilots started in
the front half of the cylinder, but starts were distributed all over
the cylinder with the largest "gaggle" of starters comprised of 4
gliders.

Climb rates for out the top starters were as follows: 6 at1.8-3.9kts,
9 at 4-4.9 kts, 5 at 5-6 kts, 5 at 6-7 kts and 4 at 7-8 kts. A number
of the start climbs were for more than 4,000' gains. There appeared to
be pretty good correlation between higher climb rates at the start and
pilots at the top of the scoresheet at the end of the contest, so
perhaps the difference in climb is more a result of strategy than
luck. I'll take a look at some more contest days time permitting.

I took a look at a second day at R9 in 2008 and a day from 2007.
Disclaimer: three contest days is a small sample and all three days
were flown at Parowan under conditions that allowed starts through the
top of the cylinder.

Some interesting (to me) observations:

Under the start anywhere rule the percent of pilots starting through
the top increased to 85% from 60%. Under the old rule most out the top
starts were within a mile or two of where the edge of the cylinder
meets the first leg courseline. Under the new rule starts were broadly
distributed, with a slight bias toward the front half of the cylinder.
As you'd expect, spreading out starts makes for smaller gaggles -
under the old rule it appeard there were at any given time 2-3 big
gaggles with 5-10 gliders in them - the biggest gaggle being at the
edge of the cylinder on courseline.

Now the really interesting part (repeat disclaimer here). Under the
old rules starters out the top had a slightly faster average climb
that starters out the side - less than half a knot. Also, there was
less scatter in out the top climb rates than initial climbs after
starting out the side - this stands to reason - under the old rules
climbing out the top was normally crammed together into a couple of
gaggles.

Under the new rules the opposite is true. Starters out the top had
nearly a knot slower initial climb on average than the admittedly
small number of edge of cylinder starts. The scatter (as measured by
standard deviation as a percent of average climb rate) was about 50
percent higher for starts out the top.

All of this stands to reason in my rationalizing mind. If you have
more room inside the cylinder to look there is some likelihood that a
few pilots will hit boomers and get an advantage, but the lower
average climb rate for starts out the top also indicates that perhaps
pilots got a bit fixated on starting this way - climbing out the top
at 3 knots when there are 10-knotters about is not a great decision.

All of the above is very rough and can be rightly criticized for a
host of reasons. On thing that is clear is that the first climb out of
the start is not the whole story - pilots with weaker thermals tend to
leave them sooner, so you really need to get an aggregate view of
average climb rate to the top of the lift band - even if it's spread
over 2-3 thermals. *That's too complex an exercise to do with SeeYou
and a spreadsheet.

9B

Andy,

Great analysis, thanks for taking the time. One thing you may need
to watch for if you do additional analysis, some of us had the boards
out in the bottom of the thermals to core the center and ensure we
were under the top for two minutes before starting the climb out the
top. This may lower the average climb value displayed in SeeYou.

TT

Will somebody close to 9B please throw a Rubik's Cube over his
transom. Thanks!

2NO

On Jan 14, 5:41*am, Tuno wrote:
Will somebody close to 9B please throw a Rubik's Cube over his
transom. Thanks!

2NO

Is that a new shape for the start cylinder?

Anything worth doing is worth over-doing, I always say.