Impossible Returns? Hardly . . .

"Certain price patterns are nonrandom and will lead to a predictive effect."

That statement, according to the Random Walk Theorists, is incorrect. I must assume, therefore, that the following returns are therefore "impossible." 

Mr. Simons, 67, who rarely talks to journalists, is hardly a household name like Warren E. Buffett. But Mr. Simons, who got into the hedge fund business after abandoning a stellar career in mathematics, has a track record that is jaw-dropping. This summer, word leaked out that he was starting a new fund – people took to calling it the "$100 billion fund" because its marketing materials say that it could conceivably grow to that enormous size. Not surprisingly, that has caused Wall Street types to be even more curious about him.

Here are Mr. Simons’s numbers: from 1990 to 2004, Renaissance’s primary hedge fund, called Medallion, has delivered annualized returns of 33.21 percent. (The Standard & Poor’s 500-stock index has returned, on average, 10.98 percent during those same years.) Since the end of 2002, the fund, which has $5 billion under management, has disbursed $4.9 billion to its investors – with another $1.5 billion to be delivered at the end of this year. (emphasis added)

Better than 33% returns — audited returns! — for 14 years.  I think it would be a whole lot of fun trying to come up with the title of Mr. Simon’s book:

Theoretically Impossible
Not Very Efficient   
Random Walk THIS!

(Submit your own title suggestions in comments)

How does Mr. Simons achieve his impossible returns? Through very complex algorithms that scan thousands of stocks (and other highly liquid securities that trade in public markets around the world) and executes rapid-fire trades. These algorithms are derived by the firm’s scientists, who search for "repeatable patterns and other signals."

The NYT quoted Mr. Simons saying this:  "Certain price patterns are nonrandom and will lead to a predictive effect . . . we’re very statistically oriented." He stays away from exotic derivatives.

Quite fascinating.

NOTE:  With this post, we add the category "

Hedge Funds."

$100 Billion in the Hands of a Computer
NYT, November 19, 2005

What's been said:

Discussions found on the web:
  1. I, Hans. commented on Nov 19

    Let’s talk it over with Wegelin.

    The Big Picture: Impossible Returns? Hardly . . . : The NYT quoted Mr. Simons saying this: “Certain price patterns are nonrandom and will lead to a predictive effect . . . we’re very statistically oriented.” He stays away

  2. The Prudent Investor commented on Nov 19

    No title suggestions. But where can I apply for a job with him to learn more?

  3. Mike Arnold commented on Nov 19

    How about something along the lines of: A Ponzi By Any Other Name…..hmmmmmmmm?

  4. Barry Ritholtz commented on Nov 19


    Um, no, you are totally wrong. Go back and reread it, you must have skimmed it too quickly.

    14 years of Audited Returns at 33% — AUDITED RETURNS, Real money, real investors.


  5. royce commented on Nov 19

    Maybe I’ve got the theory wrong, but I don’t think you’re giving it a fair shake here. Under the random walk theory, it’s not impossible for someone to turn in performance much higher than the market as a whole, just as it’s not impossible for someone to rack up big wins at a roulette wheel. Out of 10,000 traders, they expect to see a few do spectacularly well.

    The theory also holds that to the extent that this guy has found inefficiencies to exploit in the market, those inefficiencies will be eliminated once people discover his methodology. The random walkers do not, after all, deny the existence of any inefficiencies. They just say that once inefficiencies are discovered, the market is efficient enough to remove them.

  6. John Navin commented on Nov 19

    It’s more of a blackjack game than a roulette wheel. That is, it’s not the inefficiencies, it’s the probability theory. Simons is practicing a variation of “counting the cards.”

    I’m at 31% annualized for four years in the Marketocracy competition, fully-audited, and I’m available for hire.

  7. Barry Ritholtz commented on Nov 19


    1) He’s talking about predicting short term movements of markets and stocks — something the RANDOM WALK theorists say isn’t possible )hence the name Random Walk);

    2) 14 years is quite a long time to be doing mining inefficiencies . . .

    John Navin:

    Your Marketocracy returns are impressive (Congrats!)

    But p[aper trading is not the same as real money management — there’s no slippage, no execution issues, no actual impact of trades moving markets or stocks prices yourself. Then there’s the lack of stress associated with managing actual clients assets — and the subsequent humn errors these things cause . . .

  8. royce commented on Nov 19


    Then his method should be easily replicated using the same methodology by anyone, shouldn’t it? How long will the method hold up once everyone is using it?

  9. PC commented on Nov 19

    Trading repeateable patterns in markets is nothing new. Certain price patters do recur but the key is “they do not always do so each time.” So the total game is much more than just trading repetable patterns.

    When the pattern doesn’t repeat, you cut your losses short and wait for another set up. After a streak of losing trades, you will eventually hit on a winning set up where the pattern does repeat. You then let your profits run on the winning trade. If you cut your losses short and manage your position size correctly during the drawdowns, you will gain a “net overall profit”. So in essence you are playing a game of trading for positive expectancy, statistics, AND psychology.

    Most traders will quit after a few losing streaks and move on to look for other holy grail (there is none). This is the psychology part.

    So if you give the same method to ten traders with the same rules, you will have ten different results. You will discover that the same profitable methodology can be unprofitable in the hands of some traders. So Royce’s comment above that how long will the same methodology hold up when everyone is using it is not valid.

    And when you are running $5 billion like Renaissance, you scan the entire universe for such repeatable patterns. Entire universe will include “normal” exchange listed futures contracts such as financial futures. This together with normal equities should give you plenty to work with and you can achieve good returns.

    However, when you get up to $100 billion, the law of diminishing return kicks in. You will not be able to get 35% “long term recurring” returns no matter how good you are. It’s simply impossible mathematically.

    So investors who look at the old 35% return track record as the reason to go into the $100 billion fund will be disappointed. That is not to denigrate the performance of Simons. It’s the natural consequence of size.

  10. blackpanther commented on Nov 19

    oh, i am sorry it was PC who posted it.

  11. Damian commented on Nov 20

    PC – with the lack of decent returns from most traditional investment opportunities, I’m sure there would be any number of pension funds (particularly underfunded pensions) that would welcome a 15% annualized return. So while the returns might not be as great as 30%+, I’m sure they’d be very, very happy with even a 10% non-correlated return on their investments.

  12. Econbrowser commented on Nov 20

    Hedge fund risk

    Psst– want to earn a 41% annual return over a decade? Then read on.

  13. Justice Litle commented on Nov 20

    Don’t forget — Simons’ 33% returns come net of fees. HUGE fees. His absolute performance — in both percentage terms and absolute dollar terms — is thus even more stunning.

    The roulette wheel mention is spurious. Not only is 14 years a VERY long time for survivorship bias to play out — 3,500 trading days — but ‘high frequency finance’ firms like Renaissance Technologies literally trade hundreds if not thousands of times per day. Those guys could easily have more than 100,000 trades on their books.

    Add up the frequency, duration and astonishing returns, and the odds of Simons’ results being a fluke are approximately on par with Britney Spears being elected President of the United States.

  14. programing commented on May 10

    I can beat that! I have developed a program that returns 45% a year on one’s money. It took many many years to program it but it finally works. Basically that is correct – there are patterns that that my program finds and there is a certain probability associated with it……

  15. justice commented on Jun 19

    It’s more of a blackjack game than a roulette wheel. That is, it’s not the inefficiencies, it’s the probability theory. Simons is practicing a variation of “counting the cards.”

Read this next.

Posted Under