I want to start this post with a disclaimer, for fear it might be censored ;-).  I’m a reasonably good racing  pilot, but I was just barely off the bottom of the score sheet at the Uvalde pre-worlds, and had never tried team flying before.  These are my personal thoughts based on flying with the U.S. team (and with Gary Carter for the last three days), and do not necessarily have anything to do with reality.

• Team flying can be extremely effective, especially with three or more pilots on weak and/or blue days
• Team flying is HARD to do effectively, and can easily turn out to be more of a distraction than a help.
• Team flying is only effective if all pilots concerned can stay in visual contact with each other.  Once visual contact is lost, then team members are essentially on their own; the lead pilot (assuming the members know which pilot is in the lead) can call back with general climb/condition reports, but that’s about all.
• Team flying cannot be learned in a day, or even a contest.  I tried to fly with the U.S. 15m team and/or Gary Carter (HK) on almost every day at Uvalde.  Afer 8 contest days, I still sucked big time.  Individual team members (notably Sarah Kelly and Francois Pin) seemed to ‘click’ better than others, but even they were having problems staying together.
• It is extremely easy to get separated, and extremely hard to get back together again.  Gary Carter and I flew 3 days together on our own frequency.  We had similar gliders (an ASW-27 and a Ventus 2bx) and similar (regional level) skill levels.  As long as there was a definite line to follow, we had no problem staying together, even if one pilot got a little ahead.  We used John Cochrane’s technique of staying together horizontally, with the guy on top moving ahead slightly to find and core the next thermal for the lower guy.  In our 3 days of flying, we found that (when we were doing well) that the lead would change back and forth quite often, and the trail guy usually could catch back up again fairly quickly.  However, when there was some ambiguity about the line to take, we often found ourselves getting separated.  In other words, we sucked at reaching a consensus on which way to go – we tended to revert back to our individual flying habits and go our own way.  It tended to start with what we thought would be a very temporary, very small deviation (“OK, I’ll try this cloud and you try that one”), but invariably developed into “I’ve lost you” within just a few minutes.  At that point, we would start trying to get back together, and this trick almost never worked.  Even when we were convinced we were at the same mileage from the next turnpoint and in the same offset relative to the course line, we almost never saw each other.  And, the communications effort significantly detracted from the real business at hand – going farther and faster.