You discuss the behavior of stock prices after an earnings announcement. There is a significant amount of academic research on this topic (called post-earnings-announcement drift). It basically finds that stock prices tend to move sharply initially, but continue to gradually follow in the same direction as the initial move for several weeks thereafter. I'm not sure if this behavior can necessarily be connected to options markets. For instance, there is research into the size of the PEAD when firms have options on their stock and when they don't. The stocks that have options tend to move to the "correct" price faster than the ones without options.
As for the direction of the stock price move, markets typically look to whether the firm beats estimates or not. If the firm beats estimates, then the stock price rises, and vice-versa. Hence, it is not unrealistic to expect stocks to either go up or down sharply following earnings.
When you say that options aren't priced for a 2% move down and a 5% move up, that's like saying the implied volatility curve is not as skewed before earnings announcements. Normally, implied volatility will have a skew/smile as a result of out of the money put contracts being expensive (people are buying protection to hedge themselves against a decline in prices). This presentation: http://users.iems.northwestern.edu/~armbruster/2007msande444/presentation4.pdf suggests that there is no clear pattern of the behavior of implied volatility before and after earnings announcements. If there were movement, it could be informed trading in anticipation of a better/worse earnings release and call, or it could be due to a broader change in market risk appetites, or it could just be noise trading.
Nevertheless, there's a lot of research related to options pricing around earnings, such as: