This really comes down to two viewpoints.

Naughty Alien certainly has a point here. LE3 will not magically make your game come alive. That's entirely down to your leadership and drive. If you can't make the cut, no engine out there, be it real or theoretical, is going to make your game happen.

The other viewpoint is that from a tech standpoint...LE2 is vastly behind it's competition in terms of feature set and ease of use. Compared to Unity or the UDK, LE is showing it's age, especially at the price point*.

Unity Free/Indy is just as capable as LE in terms of what it can accomplish. It's also cross-platform and runs in the browser. It even runs on Linux at this point. And this is all free. It's also arguably easier to use with a shorter learning curve.

UDK can do significantly more out of the box than LE can. After the learning curve you can also accomplish more much more quickly as well. It's cross platform, running on OS X, Windows and iOS (android support coming soon). The editor is top notch and it's got a very healthy community with a vast amount of support. It's also $100 per title*.

Do I like LE? Yes. I mainly use it for small prototypes or fiddling around however. It just doesn't have the toolset to compete with the competition. Especially when I need licenses for my entire team.

----

*I'm aware that in the long run LE is cheaper than Unity or the UDK. However, this is true only under the condition that a game I sell makes a significant amount of money. At that point, having to pay a little extra is what I call a Good Problem™. I've already made enough that I have the cash on hand to cover it and then some.

It's also true that Unity or the UDK are more expensive in the long run provided that the given toolset does NOT cut down on my development time. In practice, this isn't the case, especially with the UDK.

I'm not trying to bring down LE or anything. I was asked a question and I'm just trying to answer it seriously and honestly. I think LE has a lot of potential, but it is not there yet.