Understanding Performance – Socrates v Medallion?


QUESTION: Marty; I invested in your Deutsche Bank hedge fund and the performance was about 3 times that of even the Renaissance’s Medallion fund. Your employees said for the public fund you closed positions early because you were making too much in 1998. Yet that was still about 3 times what Medallion produced in 1998. Medallion is closed since 2005 and nobody has been able to duplicate their returns no less your’s. You said at the WEC you had no interest in returning to managing funds. Why is it that the only two quantitative funds to be successful, you and Renaissance, do not take on more clients?

Thanks for a great conference. Will see you in Hong Kong



1998-ltcm-contagion1998-yen-long-term-capital-management-crashANSWER: Performance declines the larger a fund becomes. There is a limit to the amount of money one can manage on the same scenario. Conspiracy theorists do not want to hear that. But this is reality. Someone can return 50% with $10 million and lose money with $100 million. As I will point out, the scope of trading is paramount to fund management.

To set the record straight, yes I had to close out positions early in 1998 in the public fund because we made way too much money. That may sound nuts, but in a public open fund you cannot post gains in the hundreds or percent for a two months. It would upset the entire industry cause all sorts of problems even with regulators. The model correctly forecast the Long-Term Capital Management Crash. I sold $1 billion worth of Japanese yen at 147 against the Yearly Bullish Reversal in addition to numerous other markets. They began calling me Mr. Yen for that trade.

1998-sp500-july-20In the share markets, I had even sold the S&P500 on the very day of the Economic Confidence Model peak – July 20th, 1998, which was the precise high in that market. Getting so many markets precisely correct presents a problem because most people do not comprehend that markets are (1) interlinked, and (2) precise. They immediately want to say you manipulate markets since the vast majority do not understand how the world economy functions and go back to instantly assuming manipulation. I was accused of manipulating the world economy because that is easier to assume than perhaps things do not work the way people believed.


Add to all of the markets we traded precisely for that move, the fact that the London Financial Times put on the front page of the second section the story that we had forecast the Russia was about to collapse at our London World Economic Conference that summer brought in even the CIA. When they called, it was in the middle of the panic and wanted me to go to Washington to build this model for them.


The Case-Shiller real estate index peaked precisely to the day of the Economic Confidence Model on 2007.15 (February 26th, 2007). However, it also marked the precise day of the infamous sale at the top of Goldman Sachs’ notorious ABACUS 2007-AC1 $2 billion Synthetic CDO, for which they were charged with fraud. How many times does this model have to pinpoint turns precisely to the day (even 1987 Crash), before people concede perhaps we are all connected and this is how the world works rather than me waking up with a premonition. There are people who pour over everything I write desperately trying to prove me personally wrong rather than looking at the reality that maybe there is something very important lurking behind the appearance of randomness and this is not my opinion against their’s.

You have to understand that performance will decline with any fund the larger the fund grows. Yes our track record with Deutsche Bank outperformed the Medallion Fund back then. However, to be fair, the Medallion Fund began only in 1988 where as our model was already being used for more than 15 years before that so we had much more experience. Comparing one to another may be interesting, on the surface, but it really comes down to the scope of the trading. We used the entire world even during the 1980s. This is why we have created the Global Market Watch to reveal how everything is connected to help take that giant leap forward for humanity.

Strategy is very important in trading. This is the core purpose of the World Economic Conferences. It is not that you trade every specific market we cover. They key is you can see how the world is unfolding and then focus your attention on the market sector you prefer, yet draw confidence from watching the strategy from the whole. It is not that the strategy fails, but you cannot execute an unlimited amount of trades of the same position.

Performance will not remain the same by simply throwing limitless amounts of money at the same market. There is not enough depth to any market whereby it can absorb an endless supply of positions or even if it were confined to a limited group of markets. This is why Medallion is a closed fund. If they accepted more money, their performance would decline. It is not a one for one relationship. Yet it is also why I developed the Global Market Watch to expand the strategy to the entire world without limitations. This is also why we have had the largest and most diverse client base from around the entire world. We are not just forecasting the United States or a single market. Those who focus on just forecasting a single market, like gold, typically hate us and only try to disprove whatever we do because they have nothing else to offer.

The Medallion Fund is the ultimate black box. They really do not report what they made money on and they are also renown for having just quants and not fund managers from Wall Street. Why? Once you interject human reasoning, performance declines. Some will claim to be managing money based upon our model, but in reality they will fail because it is simply their interpretation and human emotion will intervene.


Funds have long tried to mimic myself as well as Medallion. They have never been able to because they are not really quantitative and human decisions, including emotions, override things. The Brady Commission investigated the 1987 Crash. They assumed at first it was computer trading. What surfaced was that most computer models said sell when our Double Weekly Bearish Reversals were elected. However, the fund managers did not sell because they did not believe their systems. When the market gaped down, panic set in and people sold BECAUSE they had no idea what was going on.

Personally, I have learned over the years that the very best trades are when the computer projects the outcome and it makes me feel like this has to be wrong. For example, BREXIT and Trump are examples of that situation. The computer will beat anyone – including me. We are all infected with emotions. They are the hardest thing to overcome.

The comparison is not whether Socrates would beat Renaissance Medallion, but why is it that the only two systems that are not based upon human decisions are at the top of the heap? How many times does the ECM alone have to work for decades before it is no longer a coincidence?