[Gary Dillard]

OK, time to bump this and give a little insight as to how SPAM got 25 and 16 on their alliance. First of all, if your team hasn't done detailed scouting before, you need to download this file and see how SPAM does it - they have a good system that has taken years to develop and it really works. The cards and the data haven't changed much, but the ranking algorithms have.

Friday night going through the data, teams were ranked top to bottom with an OPR-like metric, based on observing individual teams not alliance scores from the website and assigning weight to coopertition bridge, accounting for attempts, etc. I don't remember the exact order, but 33, 25, 180 (yes, they rank themselves) and 148 were top 4; 16 was well down the list. We then put next to each team number if they were long or wide, and if they had ever triple balanced. We went down the list and resorted, giving weight to wide TB's but comparing to the other teams above them on the list. First pick went to the best available on the list, team 25 (a long bot), so our next choice had to be a wide bot in order to triple balance. Here's where the qualitative data trumps the quantitative data. We knew Bomb Squad was having problems, but we kept asking ourselves, if they start working do we really want to be against them? To make it to and through Einstein, you need to be robust and proven, and Bomb Squad had been there. It was a high risk, high payoff decision and it fit in with our match strategy for starvation. We were stunned that they were still available, and elated that we got them.