I've spent some pretty amazing sums learning about the market, and I've found there are two kinds of tuition: One involves blindly forging ahead and losing a lot, and another involves reading the right books, making the right decisions, refusing to trust the book and your instincts at exactly the wrong moment, and losing just a little less. I heartily recommend the second course, if only because Ben Graham, Edward Lefevre, and (sometimes) Bill O'Neil can write.
I've never met an investor -- myself included -- who properly understood what he read without losing money missaplying it first. Case in point: My ridiculous attempts to apply classic Graham & Dodd techniques to the market by indiscriminately buying anything with a low P/E, and calling current earnings X S&P's P/E 'intrinsic value'. Or the time I decided to test the CANSLIM method because of a five-minute breakout (this, it should be noted, can work if you can watch your money every second of the day. I spent most of the day busy enough not to think about what I'd lost until I got home).
But books are crucial, because after the first 30% or 40% loss, everything clicks and you'll find what works for you. For me, that's a chaotic mishmash of classic value investing, ornery contrarianism, and an elitist attempt to be the first person to properly frame the next big market change (if you'd realized, in the late 70's, that a bond could be a peculiarly tax-efficient 'stock', and that seemingly worthless bonds valued that way were actually pretty valuable, imagine the kind of killing you'd have made). I don't think I could have gotten to this stage nearly as fast without digesting 90 or so years of market wisdom, spewing out a few vomitous years of losses, and finally realizing how to get it exactly right.