Why do some companies manage expectations better than others?
Here are some factors to consider.
The Audience: In providing guidance to markets, companies have conventionally thought of equity research analysts (especially sell-side) as their primary audience.
Ultimately, though, it is investor expectations that drive the game, and while analysts may influence those expectations, they are more follower than leaders.
The companies that are best at moulding expectations talk to investors (though they might use analysts as their messengers).
The information: For a company to try to guide expectations, it has to have better information than investors do -- *and* be able to convey that information to markets.
In particular, rather than just suggest that earnings expectations are too high, providing information on specifics such shipments or margins to back up the guidance will increase the impact it has.
Companies argue that RegFD, the SEC's rule restricting selectively providing information to analysts, has restricted their capacity to provide meaningful guidance, but it's fairly obvious to understand that the regulation does not prevent companies from making public disclosures to all investors.
Credibility:  To be credible, companies that try to manage expectations have to be seen as trying to do both manage them down (when they are too high) and up (when they are too low).
Too many companies seem to think that managing expectations just implied lowballing earnings and revenue numbers.
A study of company guidance statements, the primary device for managing earnings expectations found that almost 80% of guidance provided by firms in the third quarter of 2012 was negative & designed to lower expectations (rather than raise them).
A company that always tries to talk down expectations pre-release, while later beating expectations each period, is like the boy who cried wolf:  more likely to be ignored than listened to.
Results: The endgame occurs when the actual news (earnings or other) comes out, and investors measure it relative to expectations.
If investors feel that companies are fudging the number or cooking the books to deliver actual earnings that beat expectations, they will at some point stop reacting to the news. 
[Meta]: It is the fact that information disclosures have oft become a game has led some companies to consider Joshua's WarGames lesson on global thermonuclear war:  choose not to play the game at all.  
Many opt out, either because they view it as a distraction from their mission of creating long-term value for stockholders -- or because they think it is futile.
The payoff to companies from playing the game may have become smaller over time ... perhaps more companies could consider the option of not playing.