Sigh. Bitcoin was an excellent investment, in hindsight - for those who bought before nov/2013 and will be smart enough to sell before it collapses.
But, in hindsight, buying lottery tickets was much better investment - for those who bought the winning tickets.
Bitcoin has been a terrible investment for those who bought in late november, or since january.
Bitcoin is still a terrible investment, because there is no reason to believe that the price will increase, and several reasons for which it may decrease.
No stock trader cares for, or even cares to know, the past history of the price of a stock. One cannot use ancient price history to justify a claim of future price increase. Indeed, resorting to that argument only highlights the lack of any real justification for that claim.
The cause for the increase from 17$ to 1200$ is obvious: the opening of the Chinese market by BTC-China, OKCoin and Huobi. That was basic economics: larger market, fixed supply = higher price. What event is expected to happen that could propel the price much higher than 700$?
I do not want to change anyone's mind here. Adults have the right to spend their money anyway they like -- investing in risky ventures like bitcoin, poker, or lottery tickets, or even throwing it away. And they have the right to try to convince anyone to do the same -- as long as they honestly warn him of all the risks and unknowns, without fallacious arguments or unfounded promises. Gambling with your money is OK, scamming other people is not.
You are quite right, larger markets and a fixed supply increases the price. But how many people do you think use Bitcoin Worldwide? I would hazard a guess and say a little over a million. Now say bitcoin really becomes (a little) mainstream and 5 million people use it. Say those 4 million new bitcoiners start to use it now with the recent drop. You don't think that will propel the price higher? With all the recent media coverage, let's assume that everyday 3600 new people start using bitcoin and everyone of those wants only 1 bitcoin. Now you have an increased demand of 350 (3600-newly mined bitcoins per day ~3251) bitcoin per day. Now, the average bitcoiner hodls to a certain degree, let's say 50% increase in value before selling (I know it's way higher than that, but hey...). Let's also assume that all exchanges run at the same optimal price (again, I know...) and that there are no bad news that rally a dump (...) .
Then you have a demand of 350 per day or 127750 bitcoin per year left to satisfy, and that is with every miner out there selling his bitcoins instantly. Can you guess now why everyone says that the former logarithmic trendline will continue? I just thought you could use a little supply and demand mathematics.
So the answer to your question of what event is needed to propel the price upwards is easy: time. and not even that much of it.