The point is buying straddles before earnings without any consideration or discussion of implied volatility is trading in the dark and hoping for luck. It is easy to say "If the move is larger than the expected move then you make money". For example, if the stock goes up when I buy calls then you make money. If the stock stays right at the strike price for 3 weeks after shorting a straddle then you make money. If.... you see the point. It is not what theoretically could happen it is what is more likely to happen. In any straddle, if the move is greater than what the premium projects prior to expiration then you make money, that is just stating the profile of a straddle.
GOOG as the best example where IV crush can eat significantly into potential profits given the risk of the position. A GOOG straddle at $100 with high implied IV will get crushed even if the stock moves $50 at the open. We have seen IV go from 50 to 30 after an earnings announcement and the stock does not always gap and run 150 points. One cannot say solely the best time to buy straddles is before earnings, the better answer is based on a person's expectation of volatility along with expected movement. It does not matter if you look at it as vega or theta, if the play is earnings then most of the time the holding is through earnings/short-term- i.e. a couple of days. If you are holding longer then it is vega and theta that work against a long straddle. In the long run you would be fighting vega loss if you bought without any consideration of relative implied volatility for short time frames.
There is no simple easy answer as to when is the best time to buy a straddle because too many other factor come into play. Not discussing IV when it comes to straddles is the same is ignoring direction when buying calls or puts.