Because I hate to tell you, someday you'll collide with real life. If you put your mind to it, can you become president? Chances are simply, no. A lot of people have put their mind to it, like Romney, McCain, Hillary, Kerry... it's not like they were lacking in motivation or desire.
So... no, sometimes, no matter how hard you want something, you are not going to achieve it. I know there's a big self-help movement, believing in attracting positive energy, believing in yourself, and whatnot... but a lot of people can't seem to always tell the difference between realistic expectations and self-delusion.
They probably just didn't want it enough. If they'd really thought positively, if they'd just visualised themselves in the position they wanted to be in, they'd all be president now.
And if you're going to say "but what about their opponents?". Well, if they believed in themselves enough, then they'd be president too.
If Romney and Obama both believed in themselves, then they'd both be president.
Now that's the power of positive thinking.
Having said that, positive thinking and motivation are very important factors in goal achievement. It's often the case that when someone fails to accomplish something, it's because there was simply something else more important to them that inhibited their effort.
On the other hand, it is arguable that negative thinking can be a fairly strong predictor of failure. That is perhaps the most useful take-away from the otherwise relatively useless field of positive thinking: if you keep thinking you're going to fail, keep visualising and expecting failure, chances are very good that you will indeed fail, even if you had all the other necessary factors for success.
Yes, it's true that most people who want to be President, won't make it. But it's also true that every single person who thinks it is impossible to become President won't make it either.
Before you do anything, you must discover what you love and what you're good at. You can then do that thing. It feels like "anything" because if you had a choice of anything, that's what you'd choose. But you can't actually do things that you wouldn't want to do anyway.
This is easy to do actually. Unless you define error as a value that depends on multiple measurements. I prefer to just go with standard deviation and keep the ambiguity out of it.
Except compute an uncomputable function. Or transmit information faster than the speed of light. Or simultaneously measure the speed and momentum of a particle with errors whose product is less than half the planck constant.
But apart from that, anything.