mamboni
21st April 2010, 07:08 AM
Goldman’s One-Sided Coin Trick
Defying all logic and “winning†90% of the time
Eric Fry
Reporting from Laguna Beach, California...
If you flip a penny 1,000 times, it'll land "heads" about half the time and "tails" about half the time. If you bet on "Even" 1,000 times at a roulette table, you'd win a little less than half the time (thanks to the "0" and "00" slots on the wheel). These probabilities are relatively simple and intuitive.
Stock market probabilities differ somewhat. Insight improves the odds. Lots of insight improves the odds greatly...over a long-term timeframe. But over the short-term, insight provides a very limited and unreliable benefit. "The markets can stay irrational," John Maynard Keynes famously observed, "for much longer than you can stay solvent."
That said, successful stock market traders tend to "win" about 55% to 60% of the time. This win percentage is not so different from that of successful sports bettors. Again, these probabilities make sense. If you possess relevant insights into the "markets" you are trading - be they S&P 500 futures or professional football games - you can improve your odds of success...a little. Chance and pure, dumb luck still play a prominent role...unless you happen to be a trader at Goldman Sachs. For reasons that neither logic nor probabilities can explain, Goldman's trading desk "wins" more than 90% of the time...or at least it did during 2009.
(see Figure 1 below)
The inexplicably successful Wall Street firm lost money on only 19 trading days last year, which means it made money on 244 days out of 263. And Goldman did not simply make some money, it made lots of money. The firm booked a daily profit of more than $100 million on 131 trading days - that's almost ten times the number of $100 million days it booked in 2004.
(see Figure 2 below)
Even during the rough and tumble days of 2008, Goldman still managed to amass an implausible record of success by booking a daily trading profit 63% of the time and racking up $100 million profits on 90 trading days. A cynical observer could easily deduce that: 1) the "level playing field" on which Goldman purports to operate is as crooked as can be and that; 2) Goldman's miraculous trading success in 2009 may have something to do with the disappearance and/or emasculation of former competitors like Bear Stearns, Lehman Bros. and Merrill Lynch.
"Traders are supposed to live by their wits, making judicious bets on the market," observes financial commentator, Sean Paul Kelley. "Good traders who don't have inside information tend to win about 55% of the time and lose money 45% of the time, the difference being their profit resulting from their trading acumen.
"Goldman Sachs doesn't work this way," says Kelley. "They have bright people no doubt, and somewhere on the trading floor these people on occasion make good and bad judgment calls. From what it looks like, however, their traders are benefiting from two advantages: information not available to the market, and muscle. These two things give the firm an edge that almost guarantees substantial 'trading profits' quarter after quarter.
"The information part comes from a variety of sources," Kelley continues. "We've seen the scandal over High Frequency Trading, where Goldman and other firms have computers positioned at the New York Stock Exchange, getting information on trades a millisecond before they are posted publicly. Goldman sees where the market is going second by second, positions itself for very short term profits, and in effect extracts a tax on trading by individual investors and mutual funds. Goldman Sachs is the biggest player in this business... For credit products, mortgage securities, and equity derivatives, Goldman Sachs extracts similar information from its clients interested in buying or selling these products...
"None of these information sources or uses are illegal at this point," Kelley concludes, "but this is hardly the profile of your typical day- trader pitting his wits against the fickleness of the market; this is the profile of a hedge fund with critical information and size advantages, using them to maximize profit."
As Kelley correctly observes, none of Goldman's known trading practices are illegal. On the other hand, legal trading practices have never before generated such a sustained record of improbable success. So just maybe, Goldman's brilliant trading record emerges from something other than gee-whiz computer programs and the brilliant instincts of trading jocks.
Remember, Goldman generates its trading results from the activities of hundreds of traders, operating in dozens of different financial markets. And yet, somehow, the collective activities of this gun- slinging diaspora produce a daily profit 93% of the time. That's either very impressive or very illegal.
The truth will come to light eventually.
Defying all logic and “winning†90% of the time
Eric Fry
Reporting from Laguna Beach, California...
If you flip a penny 1,000 times, it'll land "heads" about half the time and "tails" about half the time. If you bet on "Even" 1,000 times at a roulette table, you'd win a little less than half the time (thanks to the "0" and "00" slots on the wheel). These probabilities are relatively simple and intuitive.
Stock market probabilities differ somewhat. Insight improves the odds. Lots of insight improves the odds greatly...over a long-term timeframe. But over the short-term, insight provides a very limited and unreliable benefit. "The markets can stay irrational," John Maynard Keynes famously observed, "for much longer than you can stay solvent."
That said, successful stock market traders tend to "win" about 55% to 60% of the time. This win percentage is not so different from that of successful sports bettors. Again, these probabilities make sense. If you possess relevant insights into the "markets" you are trading - be they S&P 500 futures or professional football games - you can improve your odds of success...a little. Chance and pure, dumb luck still play a prominent role...unless you happen to be a trader at Goldman Sachs. For reasons that neither logic nor probabilities can explain, Goldman's trading desk "wins" more than 90% of the time...or at least it did during 2009.
(see Figure 1 below)
The inexplicably successful Wall Street firm lost money on only 19 trading days last year, which means it made money on 244 days out of 263. And Goldman did not simply make some money, it made lots of money. The firm booked a daily profit of more than $100 million on 131 trading days - that's almost ten times the number of $100 million days it booked in 2004.
(see Figure 2 below)
Even during the rough and tumble days of 2008, Goldman still managed to amass an implausible record of success by booking a daily trading profit 63% of the time and racking up $100 million profits on 90 trading days. A cynical observer could easily deduce that: 1) the "level playing field" on which Goldman purports to operate is as crooked as can be and that; 2) Goldman's miraculous trading success in 2009 may have something to do with the disappearance and/or emasculation of former competitors like Bear Stearns, Lehman Bros. and Merrill Lynch.
"Traders are supposed to live by their wits, making judicious bets on the market," observes financial commentator, Sean Paul Kelley. "Good traders who don't have inside information tend to win about 55% of the time and lose money 45% of the time, the difference being their profit resulting from their trading acumen.
"Goldman Sachs doesn't work this way," says Kelley. "They have bright people no doubt, and somewhere on the trading floor these people on occasion make good and bad judgment calls. From what it looks like, however, their traders are benefiting from two advantages: information not available to the market, and muscle. These two things give the firm an edge that almost guarantees substantial 'trading profits' quarter after quarter.
"The information part comes from a variety of sources," Kelley continues. "We've seen the scandal over High Frequency Trading, where Goldman and other firms have computers positioned at the New York Stock Exchange, getting information on trades a millisecond before they are posted publicly. Goldman sees where the market is going second by second, positions itself for very short term profits, and in effect extracts a tax on trading by individual investors and mutual funds. Goldman Sachs is the biggest player in this business... For credit products, mortgage securities, and equity derivatives, Goldman Sachs extracts similar information from its clients interested in buying or selling these products...
"None of these information sources or uses are illegal at this point," Kelley concludes, "but this is hardly the profile of your typical day- trader pitting his wits against the fickleness of the market; this is the profile of a hedge fund with critical information and size advantages, using them to maximize profit."
As Kelley correctly observes, none of Goldman's known trading practices are illegal. On the other hand, legal trading practices have never before generated such a sustained record of improbable success. So just maybe, Goldman's brilliant trading record emerges from something other than gee-whiz computer programs and the brilliant instincts of trading jocks.
Remember, Goldman generates its trading results from the activities of hundreds of traders, operating in dozens of different financial markets. And yet, somehow, the collective activities of this gun- slinging diaspora produce a daily profit 93% of the time. That's either very impressive or very illegal.
The truth will come to light eventually.