I know that UDUstats tracks the success/failure rates for various upgrades, but I thought we might get more "scientific" results if we kept track of these rates using MilePoint's built-in polling ulitity. I have allowed "selection of multiple responses" in the hope that the utility will let anyone take the poll multiple times (that is the unknonwn), but even if folks can take the poll only once, I believe that things will work out just fine, as a long as each of us votes accurately (read: truthfully). Note that to test the idea, this is limited only to p.s. (JFK-SFO or JFK-LAX) flights. If it works, we could have similar polls for CPUs, RPUs, GPUs or mile upgrades for any flight or per region. The limitation, of course, is that this would provide only the distribution for folks who use MilePoint. But, hey, we are the important ones, right? If this works, we might be able to have a lot of fun by using more extended (e.g., bootstrap) analyses to estimate the true distribution, after opening up the poll to include all types of upgrades and regions. For now, do vote as often as the system will let you, after you clear/do not clear an upgrade on p.s. only in 2012. If the system lets you vote only once, that is okay too, but please to vote, even retroactively (i.e., p.s. trips that have already occurred in 2012). The poll will be open for 365 days... I just got the idea for this and will start because I just had both the outbound (RPU) and inbound (GPU) upgrades clear for JFK<->LAX p.s. flights that I am taking in about a week. Here we go... UPDATE: Unfortunately, and as I had feared, you will be allowed to vote only once, so let's refine this a bit more. For a two-way trip, please provide success or failure only for the outbound segment. For one-way, well, just tell us if it is success or failure.