Ultimately you want to find a stock that's gonna move sometime in the near future.  Be it the result of an earnings release, expected upcoming news, or random speculation.  The more it moves the more money you make.  Keep in mind though, As soon as you enter a straddle trade you start losing money.  The longer you are in a straddle trade the more money you lose.  This is because both of your options will be losing time value if the stock continues to go nowhere.  At times of high volume and high volatility the price of both legs of your trade may become temporarily inflated due to the extra speculation.  This is a great time to sell.  It is also a horrible time to buy into a straddle.  You should never buy into a straddle during times of high volitility, like the early morning or when the stock you are interested has a lot of excess volume and hype.  These are great times to sell.

The reason people often say to buy into a straddle 1-2 months before the exciting event (like an earnings release) is because the extra hype over the upcoming earnings release will cause greater volatility as well as the price of both calls and puts to increase as people make their bets on the upcoming earnings.  If you buy into the straddle at this time you will lose a lot of cash.

My advice to you is to observe the price of the options of some stocks that you think will move significantly in the near the future and pretend trade a few straddles to see what happens before trying it out for real.  A good straddle trade involves a fair bit of qualitative decision making.  Also it's best if you know something everyone else doesn't.  Everyone knows when an earnings release is going to happen and making a straddle trade profitable then is a bit more difficult.  However not everyone in the world could have guessed that SNP would go so crazy last week as a result of all of the talk about chinese gas prices, but you could have.

rauchory, in terms of greeks straddles are long gamma, long volatility and short theta. In other words they make money by big moves in the underlying (long gamma), increase in implied volatility (long IV) and loose money if nothing happens and the time just goes by (short theta). There is an interesting book by Jeff Augen "The Volatility Edge in Options Trading". Highly recommended. One of his straddle/strangle approaches is to purchase a straddle/strangle approx 20 days before earnings and sell the day before the earnings. Usually the underlying has not moved much for that time however because of the anticipation of the earnings the options IV went up. The next day the stock may move or may not move based on the earnings but one thing is sure - volaility will come crushing down, so he sells before the release