CNBC’s million-dollar portfolio challenge begins this week. Participants can trade a fictional account of stocks and currency. Prizes are given over each of the 10 weeks, and then a grand prize winner is awarded a million.

In 2006 I wrote a column “How Your Children Can Win the Stock Market Game.” I’m going to use that same logic to help you maximize your chance of winning the CNBC challenge.

A balanced portfolio of a dozen exchange-traded funds (ETFs) would be an ideal portfolio if this were real money and you wanted the best chance of meeting your retirement goals. Your portfolio would include thousands of stocks and you would receive a satisfying market average.

But average doesn’t win. Among the participants from around the world, only one will win a million dollars. Second place gets a beautiful Maserati. Third place gets nothing. So like a downhill skier in the Olympics, you need to risk every bone in your body to go for the gold. Do everything wrong, and you just might win.

If the rules permitted it, you would want to bet everything on a penny stock and hope it went to three pennies. But the contest has limited the number of wildly swinging stocks by listing about 2,700 stocks that participants are allowed to trade.

These stocks, although still quite volatile, are highly traded equities. You are not allowed to put more than 25% of your portfolio value into any one stock. Studies show that any stock position greater than about 15% of your portfolio adds a tremendous volatility without any additional return. This random volatility gives our portfolio the best chance of being the best or the worst in the contest.

Because the riskiest portfolio has a greater chance of winning, we should invest in just four stocks. Such combinations number about 2.2 trillion, smaller than the national debt but much larger than the number of participants.

You can have five such portfolios. Each one has a separate chance to hit the stock lottery. You could go bankrupt with four of them and still win the grand prize with the fifth.

Many such portfolios will have winners that will cancel losers and end up in the middle of the heap. Most of the movements in the stock market are just noise. Think of stock returns like the end of a whip being violently cracked up and down. Most of the movement is just noise. But the person cracking the whip is standing on an escalator, and therefore there is a reversion to the mean escalation as stocks return about 6.5% over inflation annually.

We must avoid those combinations with a greater chance of positions that cancel each other. A biotech stock and a mining stock will cancel each other if biotechs do well and mining does not. But a portfolio of four biotech stocks might all soar together if the sector as a whole does well. Therefore you stand a better chance of winning if each of your five portfolios has a theme that causes them to move in sync.

Finally, you don’t want to invest in utilities. These stocks do not offer market returns; neither do they offer market volatility. For the same reason, I would avoid real estate investment trusts (REITs). These are currently great investments, but they are also fairly slow moving. Even in the drop-and-rebound markets of 2008 and 2009 caused by REITs, they were some of the least volatile movers.

Extrapolating, it is also unlikely that good value-oriented, blue chip, dividend-paying stocks are going to win. Maybe they would if you pick the right ones. But if the markets rise, the best bets will be stocks that are the most in sync with such market movements as measured by beta.

Large companies have a more difficult time continuing their growth than smaller companies. Thus picking the largest companies such as Apple, Exxon Mobil, BHP Billiton or Microsoft will probably not result in the largest growth if the markets go up. Unknown smaller companies will probably outperform these larger companies.

So we should pick just four stocks in the same sector of the economy or country with as high a beta and as low a market cap as possible. For example, we could pick all precious metals and mining companies with an emphasis on silver, which could be even more volatile than gold. Four such eligible silver stocks are Coeur d’Alene Mines Corp (CDE), Silver Standard Resources Inc (SSRI), Endeavour Silver Corp (EXK) and Silver Wheaton Corp (SLW).

This four-stock portfolio is as highly concentrated and as poorly diversified as allowable. Although it is not what you would choose for your own portfolio using real money, it has a good chance of being a statistical outlier as part of the game.

Your investments are real. To meet your financial goals, you don’t need to beat the investment returns of everyone else. You want a decent return to enable you to retire comfortably and ensure a cash flow that will support your standard of living.

Of course CNBC has its own reasons to sponsor the contest. It will make money off participants in at least two ways.

Every day up to three multiple-choice trivia questions will give you an additional 2,000 CNBC bucks for each correct answer. Questions must be answered by 8 p.m., and the correct answers will be reported throughout the day on CNBC. Increasing viewers is always profitable. But the contest ends on November 25, and the fall sweeps take place from October 28 to November 24.

CNBC may also make money by requiring participants to trade $100,000 of their portfolio in Forex currency trading, unfamiliar to many people. FXCM, which appears to be sponsoring part of the challenge, wants you to become familiar to increase their currency trading business.

If you decide to participate in the contest, avoid watching CNBC all day long just to learn the trivia answers. Most of the daily financial information is just noise. One advisor calls it financial pornography.

And if you opt to trade currencies as part of the contest, avoid the practice in real life. Purchasing unhedged foreign bonds provides a better currency exposure with a fixed-income portfolio paying you interest.

Buy your four stocks the first day and hold them throughout the competition. You have $900,000 to invest in stocks, so buy $225,000 of each stock. Use your $100,000 for currency trading to invest in one currency against another. For example, invest in the New Zealand dollar against the Japanese yen. Then forget about the game for the remaining 10 weeks.

I love strategy games and what children can learn by playing them. But a market trading game teaches all the wrong lessons. Most investors are concentrated in large-cap stocks in the United States. This is better than four individual stocks in one sector of the economy, but it is the same mistake. Guard your real portfolio against such concentration.