American Airlines’ Bizarre $16 Fares

B6 is doing something similar (and DL isn't far behind). WN is at least keeping a floor under fares. My best guess would have been that the computers are possessed and the human fare-setters are overloaded...but there is no way in H-E-Double-Hockey-Sticks that these fares were filed as of two months ago. As to folks "risking life and limb if AA were still distance-based", I feel compelled to point out that a LOT of program cross-credits are still distance-based. If you've got a few flights on BA (for example), racking and stacking flights at those prices might actually make sense as a status run.

@Mark Anley. That's for Air Canada's new "Cockpit Class" they've introduced. You'll be in seat A0, the Co-pilot's position.

Well Air Canada is staying competitive... $1000 from YYZ>YVR round-trip (economy).

Potreroflyr & others If that is true, why did the Imperial College original model predict peak cases 12 to 18 months from now WITH bending the curve and now saying mid April? Put simply, several large cases came out and suggested much lower # of deaths AND countries are starting to ramp up to test for antibodies. With anti-body tests, we will know exactly which wave we are on (~10 on his original model, vs 18-21 for the Stanford, Oxford, his revised study). Stanford and Oxford both asked Governments to begin antibody studies ASAP to prove where we are once and for all and how many have immunity already. Lastly, why are none of these bend the curve studies actually putting a value on each piece of bending the curve? Some of the bend the curve part has NO economic costs (washing hands, surfaces, white collar working from home, social distancing, etc) while others literally will cost the world $20+ trillion in economic output (shutting everything down). How much do they estimate R0 is reduced by each of these?

Richard - Yes, that is what the Imperial College study guy said, but his data does not suggest that. The main reason for the reduction was the increase in R0 with more assymptomatic cases. His original model needed us to bend the curve for 12-18 months. The new one was just a few weeks. You can run a basic expotential table in excel yourself. 3-4 weeks bending the R0 from 3.0 to 1.5 and then opening back up, you end up in the same place just a month later, so that can't be it, especially since about 80-90% of those that end up a vent die regardless. Regardless, the Oxford Study and Stanford studies, which ironically came out just a couple days before the Imperial College revised their estimates down by at least 96%, started at the significantly lower estimate, and Amherst college did a study of 13 experts 2 weeks ago and most expected about a quarter million deaths in the US +-150k. In short, the US media has run with the bad estimates from this thing non-stop and isn't discussing the estimates that show this will likely be the equiv of a very bad flu in terms of deaths but not catastrophic. Lastly, without doing anti-body studies, they really don't know as they have to make unknown assumptions about those with mild cases or assymptomatic all together that are not counted in the denominator. As someone that has worked with #s for a living for more than 10 years now, graduating summa cum laude with a BS in Economics and received a MBA from a T10 program, I can tell you its easy to make a model spit out whatever you want it to, especially with several unknown variables.