My experience was this. On a call to Delta on 4/20 I was told it would be Monday 4/26 before a seat was available. 15 minutes later I booked a DELTA/KLM flight to my destination that afternoon through travelocity (at a high cost).<p>When I complained to Delta after returning I was told the travel websites buy seats in block and resell them, so that is why they could not give me a seat. Oh, and there were several empty seats on the flight.
The logistics involved are hard. Really really hard. These are not boxes that are stranded. They're people. And people move around, change their minds, make other plans, book themselves multiple times, etc.
Assuming airlines are trying to maximize their bottom lines, if they want to capitalize on this spike in demand to offset the lower demand that resulted from the flight restrictions, they might be wanting to slowly fill up flights by gradually lowering higher prices.<p>The idea would be to only have a few empty seats. If the flights are full, their prices might be too low. Eventually everything will return to historical equilibrium as stranded passengers exit the market.<p>Altruistic airlines might go out of their way to get stranded passengers back to their homes, which could build customer loyalty and long-run revenues, however.<p>(This is just all wild speculation.)
The models are optimized for the 99% of the cases without any special circumstance. Airlines overbook by x% because they know that a lot of people won't show up, they increase the price over time etc.<p>I can see that an event like this can throw the whole system out of whack. It's not obvious how the parameters should be altered. Without any data about how many people will buy more than one ticket in a situation like this, predictions are impossible.