FlyerTalk Forums - View Single Post - CPUs improving ?
View Single Post
Old Mar 29, 2013 | 1:00 pm
  #39  
entropy
FlyerTalk Evangelist
30 Nights
30 Countries Visited
1M
20 Years on Site
 
Join Date: Feb 2002
Location: San Francisco/Tel Aviv/YYZ
Programs: CO 1K-MM
Posts: 10,859
While this is probably the subject of a different thread, there's clearly something afoot on some of these flights where I'm way down the list. This week, DEN-FLL flight. Flight was mid-day and mid-week to a tourist destination in the height of spring break season. This was not a typical Thursday or Friday night hub-to-hub flight. I finished #7 on the upgrade list on a pmUA 757 (24 seats in F). According to the mobile app, about half the folks in F were "upgraded" by some means or another. The other half apparently "booked" some variant of F or don't show on the mobile app as upgraded.

There's no way to convince me that:
A dozen people paid a full F fare to sit up front on a four hour flight.
There were really 17 or 18 people ahead of me on the CPU list--it's highly unlikely that there were that many Y or B fares, or GSes, or even 1Ks on higher fare classes.

So I'm forced to conclude either they are (a) extremely successful at selling F seats at reduced rates, either through V-UP type fares or ToDs, either at booking or at check-in, and/or (b) the CPU algorithm isn't really as advertised. I'm slowly becoming convinced that (b) is a possibility as well somehow.
My flying is way down this year. but one observation I made was a recent SFO-YYZ flight, a flight for which there are no reasonable V-up (or discounted F seats), as a 319 was R5 a few days before the flight. I was already in F, my partner (Plat) was on a W fare in coach. The 1K window passed, still R4, plat window, R2, then it goes R0 at the gold window, then the flight swaps to a 320 (+4 seats). He ended up #6 on the list. I have a hard time believing that there were 4 plats/1K/GS on a higher fare (observing the lack of people in exit rows), and the presence of R seats post-GS/1K

I'd like to see who cleared, how and why... because I doubt it was according to the CPU published criteria.
entropy is offline