There is a big problem with trying to game the system with the airlines: The pricing is done using control theory. In fact, they have the most mature applicatio

There is a big problem with trying to game the system with the airlines: The pricing is done using control theory. In fact, they have the most mature application of control theory to pricing out there, probably by virtue of being to first to apply it in that way. Practically, that means that you can only game the system if nobody else knows how. If you are the only one that knows how, the algorithms probably can't detect and compensate for the error. As soon as news breaks out, the algorithm starts compensating. We saw this about 10 years ago when there were a few companies that started promoting the ability to search for Y-class fares...it was all but a year or so before the pricing stabilized. The same problem is currently happening for the hidden-city ticketing trick [1](the algorithms are compensating, although the airlines put in some manual systems when it first broke news). The same goes for these ideas of "the best time to buy". Not only is historical data not good enough, but the more well known the data becomes, the worse the predictions become. This has already happened to some degree with the "buy as early as possible" crowd. Now it is more expensive to buy early than it is to buy late...something that wasn't previously true. The lesson? If you figure out a way to game the system, don't tell anyone else. [1] http://www.wsj.com/articles/SB10001424127887323539804578259740155116054 http://www.wsj.com/articles/SB100014241278873235398045782597...