Not only are you badly misinformed, you are spreading information that could encourage someone to risk their money when they don't know what they're doing.

First off, something that requires a university education and a year of full-time training is not what most people think of when they hear the phrase "very little skill involved". When a member of the general public tries trading for the first time, they have far less than even that (and even economics professors have less of it than they think.)

Secondly, what you're describing is algorithmic trading, not discretionary trading. I guarantee you that that math whiz fresh out of college is not allowed to risk significant amounts of money on personal hunches. (I would guess they don't learn discretionary trading at all; they have their hands full with the algorithms.) I'm sure there's a lot of infrastructure to ensure that any strategy they come up with is well back-tested before going live. Again, the general public (including the professors) isn't doing algo trading and doesn't have access to that infrastructure and that support system.

Tennis was not the best analog. Trading is more like poker. First, they both have the property that the more you win, the bigger advantage you have, and conversely. Second, as they say, if you don't know who the sucker at the table is, it's you.

The "skill" involved in trading isn't really that difficult to learn - it's tradecraft, which you have to learn in any industry. And you're right; the math whiz fresh out of college doesn't get a ton of money to play with. He gets a small amount and if he does well, is progressively trusted with more. The bulk of this type of trading is about risk balancing. Institutional traders will make more money here simply because they trade using proprietary platforms that get simply them better prices on securities. The algo guys optimize the ordering systems such that the prop traders get first pick at everything.

Trading at high levels (billion dollar positions and above) is a lot like poker; but the bulk of securities trading is really no different than playing a video game. I used to know a guy who worked for a big HFT firm, and that's exactly how he described it. Humans are excellent at pattern recognition, and so the HFT guys use humans to make decisions about when to get in/out of an asset. Computer programs show patterns to the analysts, who issue a "yes/no" decision and who are judged solely on their performance.

But yes, you are also right that mere mortals like us have no chance because it's not an even playing field. I mean, the algo traders pay millions of dollars a year to get Reuters feeds 60 seconds early, so by the time news hits the market, the price has already adjusted. Hell, the trading platforms that are processing your orders sell access to their incoming order feed, and the algo traders love to pre-empt your trades to drive the price up.

A more apt definition would be that trading is like poker where the guys with huge piles of chips also have X-ray vision. And they get to skim the pot even when they lose. If you have X-ray vision and pot-skimming rights, then yeah, it's just poker. If you don't, then you should just go play Blackjack (i.e. buy-and-hold investment where you play against the market as a whole, not any individual players).