On Friday, August 21, 2015 at 5:56:05 AM UTC-7, wrote:
On Friday, August 21, 2015 at 2:19:19 AM UTC-4, Andy Blackburn wrote:
On Thursday, August 20, 2015 at 4:13:42 PM UTC-7, Papa3 wrote:

Took a look at this - thanks. Observations in-line.

Since you asked. Last day of Dannsville 2014 (avaialbe on the SSA Website). If you pick my file, UH, SM, XC, MS, and W3 at minimum, you can see that I made two critical decisions thanks to FLARM. First, I was able to see where a few guys were out of the gate and headed in that direction. Note: It was a very unusual task (don't go there - we know your feelings on MATs). There was a choice of 3 or 4 waypoints as the first turn. It was also very hazy with a crazy mixed cloudbase with some climbs going up much higher than the surrounding cloudbase. When I started, I was out of visual range of the others who had chosen Loon Lake as the first turn.

(It looks like everybody was in a thermal together, some headed out and you went back and too a couple of turns in another thermal and left three minutes later. When the others (UH, XC, SM) set course for the first turn you were 0.22 miles away, so you probably had a decent idea where they were headed - or could have known - without Flarm. On a hazy day whether you'd have been able to spot them visually (or get within the requisite 1.25 stealth miles) is not clear).

So, that made Decision #1 easy (where to go first). I then picked up MS climbing via FLARM and made a beeline for him. Good climb, but wasn't happy where he was going after that.

(Well, it was a good climb for MS, who was the first one in the thermal - 3.3 knots. The second glider in the thermal was SM, less than a mile in trail who got 3.8 knots. After that was 44, 1.7 miles behind MS, who only got 2.6 knots. You were 4.25 miles back and got there 4 minutes later. For your Flarm leeching prize you were awarded...1.4 knots and 269 feet of climb (this is all per SeeYou). You also made a 90-degree left turn to get to the next thermal that MS found 3.5 miles away. He got a 2.8 knot climb. 44 was Stealth mode leech distance behind and was awarded 2.4 knots. You were a full 3.75 miles behind and by the time you got to this thermal you were alerted to by the magic of Flarm you were able to achieve...1.4 knots. Had you gone straight and run into the same thermal as UH, XC and SM who knows what you'd have gotten - they achieved 1.4-1.6 knots, so a little bit better that you got with your Flarm-inspired deviation. It's not clear if the deviation was off course, or you just turned early - I didn't load the waypoints, or your flight claim).

Good news - several gliders off to the right per FLARM. I'll go there since I already have tactical advantage (i.e. I won the Start Gate). Decision #2 helped by FLARM.

(You had about 7 miles separation when you set out from the prior thermal. From that point on, you and the other three were on a converging course (does your Flarm get 7 miles or was that just happenstance? It was more or less the course you were on already). It looks like you deviated more steeply to meet up with them from about two miles apart, which probably cost you a fraction of a mile. It's not clear that Flarm did you any good on this as you would have met up anyway - at least with Stealth mode - if it was pea-soup hazy maybe you wouldn't have ever gotten an actual eyeball on anyone..)


From there, SM, XC, and I made up a very nice working group that did EXACTLY what good working groups do - one guy would lead out and the other would spread out. SM and I were in 18M span with XC in 15, so all XC had to do was to stay with us and not get dumped. He's way too good a pilot to get dumped, and he ended up winning the day (as he should).

(Loose team flying out on course has been common practice for generations, not really related to Flarm. We could invoke penalties for "team flying" anytime any gliders take two or more thermal in a row together - per their IGC files. It would be pretty easy. However, despite the "cheaty" nature of it, I think people kind of enjoy it. The "stay with the group and win on handicap" is harder than it seems, but even so there have been occasional calls to"legislate" it away).

So, there's a real-world example of where FLARM helped make some critical early decisions that got me connected with the pack and then helped me get connected with a good working group. The 4-5 minutes I gained put me in second for the day, just out of first.

(I'd have to load all the waypoints and the task, but it appears that the Flarm-related activities actually hurt you slightly (slower climbs that the non-leechers, by a good margin. It seems from the flight data that you actually earned your second by flying better on the non-leechy parts of your flight.

BTW, as I go through the "leechy" contest days people have sent me to look at, this is becoming a common theme. The first glider in a thermal pretty consistently gets the best climb. OTOH, followers - particularly as they get more than a mile or two behind - pretty consistently get substantially poorer climbs. I won't claim it as a universal truth but if you think for a minute how pilots decide whether to stop for a thermal they found versus one someone else is already climbing in you can start to see how the performance statistics would get skewed. Chasing someone else's thermal from more than a mile or two out is often a sucker's bet, and the worst part is you don't even know you were snookered until the flight is over and you can look at all the logs.

Veeery interestink.

9B

In many cases it is not all about a better climb, i.e. picking from a couple option to climb a bit faster, but about getting a climb at all, or at least going toward an area that is working.
At Dansville that day the real question out of the start was "will we get any climb at all, or end up at Avoca. Seeing others climbing ahead, and where, was a very big advantage.
Another less clear example is Elmira on day 6 this year. It was desperation start time with a big hole to cross somehow from low altitude. Those of us that got through the early part of the flight went to the blue more to the north. Others went to the really dark stuff more west. If they could have seen us climbing on Flarm, though poorly, I'm sure some would have come to us instead of lawn darting.
UH

If I take UH's and P3's observations together I think we come to an interesting group of insights.

First, there is growing evidence that typically Flarm leeching gives below-average climbs and is therefore a detriment to performing at the highest level on any given task. Many thermals are variable in strengths and leaders tend to find the strong bubbles (or they wouldn't stop to climb). Followers at more than a mile or so have increasing trouble finding the good part of the thermal.

Aside - Eric, as I look at your trace you caught the lead guys not because you used Flarm to find superior climbs (they were actually slightly weaker), but because you flew 3-4 miles less distance by cutting the first turn, which nearly exactly offset starting 3-4 miles behind them. I'm sure it was reassuring to find the pre-start gaggle, but it's not clear that you got any advantage from it - short of not landing out on a weak day, which is hard to prove would have happened since no one landed out on that leg that I could see.

Second, there is the case of desperation mode as UH points out. The "find any lift or land out" situation like Day 6 at HH. This is the situation where a Flarm target just might prevent you from lawn darting. For me it raises a question. Do we want to increase or decrease the probability that missing (or finding) one thermal on an iffy day determines who wins a contest? There are all sorts of provisions in the rules to devalue days like this, but it's no secret that the days where around half the field lands out are the ones that really make or break your position on the scoresheet. It also may partly explain why the HH results didn't map to the PRL rankings very well (UH made this point waaaaay up in this thread somewhere - random weather days scramble the scoreheeet, despite devaluation). I think this point has some merit - though the circumstances are VERY rare. What that means is Stealth mode increases the odds that the best pilots will get knocked out of contention for reasons that are beyond their control (finding a random thermal that everyone else misses). Is that a good thing or a bad thing for the fairness and enjoyment of sailplane racing? Put another way - do we want to decide races mostly in the air or increasingly via the landout?

More food for thought.

Off to Truckee to fly in the smoke of California wildfires.

9B