Someone sent me this its a great read
PART 1
The hardest but also most important lesson to learn in trading is how to handle losses gracefully. Most traders will inevitably encounter a string of losses at some point, so those who can't lose without being thrown off their game won't survive the market. The traders who have realistic win/loss expectations and a trading system they trust have the best chance of prevailing over tough market conditions. Here we look at what kind of losses traders can expect and how they can adjust their focus and strategy to deal with these losses.
Losing Battles
Every trader worth his or her salt knows that trading against the trend is not a good idea. So it seems logical that the best trading systems would be those that follow the trend: when the trend is going up, take long trades only, and when it's going down, it's time to go short. That being said, you'd think that trend-following systems would have the best win/loss ratios, right?
In his new book, " A Short Course in Technical Trading", Perry Kaufman offers some sobering statistics on the matter. According to this veteran program-trading expert and author, "You can expect 6 or 7 out of 10 trend trades to be losses, some small some a little larger." And yet, Kaufman says that trend-following systems are some of the best trading systems around. In other words, trend-following systems won't yield huge profits, but they'll still do better than most systems.
It will probably come as a shock to those who have spent countless hours searching for a winning system, but Kaufman makes it absolutely clear in his book that having realistic win/loss expectations means expecting losses - lots of them. He states, "As a trend trader, you should expect mostly small losses, some small profits, and a few large profits."
If it gets across this point alone,"A Short Course" is a worthwhile addition to your library. Kaufman provides an example to demonstrate a phenomenon traders in the game for the long haul have come to learn the hard way:
"In a normal distribution of 1,000 coin tosses, half of them would be single runs of heads or tails. Half of those, 25%, would be a sequence of either two heads or two tails. Half of the remaining, 12.5%, would be sequences of three in a row, and so on. Therefore, in 1,000 coin tosses, you can expect only one run of 10 heads or tails in a row."
In other words, in 1,000 trading days - or about four years - a trader could expect to experience 10 wins (or losses) in a row only once, that is, if trading were as random (normally distributed) as a series of coin tosses, which it is not.
So, your odds of winning with trend-following systems are better than your odds of winning a series of random coin tosses, but there are other challenges to having more winning than losing trades. Although markets are not random, you can still expect short-term random movements within a trend, major reversals at the end of each trend, and the time lag most trend-following systems experience when getting into and out of the market.
As a result, thanks to lags and unexpected short-term random movements, you are still subject to the effects of randomness. Given enough time, an experienced trader can expect to suffer 10 or more losses in a row. It is not a matter of if, but when.
When asked about realistic trading expectations, Thomas Stridsman, author of " Trading Systems That Work" and " Trading Systems and Money Management", had this to say:
"What is more important than how large your winning trade is when you win, or how many winners or losers you might have in a row. It is the mathematical expectancy of your strategy. That is, how much are you likely to win on average on all trades, winners and losers combined, and how much this value is likely to fluctuate in the short term.
"For an even further increase of peace of mind, you also probably are better off looking at profitable time periods, such as weeks or months, rather than profitable trades. Simply looking at a win/loss ratio is not enough."
If you think that doing either more or fewer trades would be a more successful strategy, think again. Kaufman demonstrates in "Short Course" that the more trades the trader performs, the lower his or her profits over the long haul. On average, longer-term trades generate more eventual profits. However, if you're a long-term trader, your risk of getting one or more big losses is increased, since you are in the markets longer and therefore exposed to risk for more prolonged periods. Bottom line, no matter what your trading style or preferred time in a trade, you will lose and lose big on more than one occasion.
Kaufman has the data to back up his claims. He has performed thousands of tests on various systems, and some of these are presented in Short Course. In one example, he tested Microsoft for 10 years ending Jan 2001 and covering a period when the stock moved from a pre- split price of $1.04 to a high of $60 in Dec 1999. It should be pretty easy to beat the odds following that kind of trend, right?
Using an 80-day moving average during the period to generate buys and sells the system, trading both long and short positions, generated a total of 88 trades. Of these, only 36 trades - or 41% - were profitable. And Kaufman comments in the book that "[t]hat's actually good for a trend system, which often has closer to 35% good trades".
The hardest but also most important lesson to learn in trading is how to handle losses gracefully. Most traders will inevitably encounter a string of losses at some point, so those who can't lose without being thrown off their game won't survive the market. The traders who have realistic win/loss expectations and a trading system they trust have the best chance of prevailing over tough market conditions. Here we look at what kind of losses traders can expect and how they can adjust their focus and strategy to deal with these losses.
Losing Battles
Every trader worth his or her salt knows that trading against the trend is not a good idea. So it seems logical that the best trading systems would be those that follow the trend: when the trend is going up, take long trades only, and when it's going down, it's time to go short. That being said, you'd think that trend-following systems would have the best win/loss ratios, right?
In his new book, " A Short Course in Technical Trading", Perry Kaufman offers some sobering statistics on the matter. According to this veteran program-trading expert and author, "You can expect 6 or 7 out of 10 trend trades to be losses, some small some a little larger." And yet, Kaufman says that trend-following systems are some of the best trading systems around. In other words, trend-following systems won't yield huge profits, but they'll still do better than most systems.
It will probably come as a shock to those who have spent countless hours searching for a winning system, but Kaufman makes it absolutely clear in his book that having realistic win/loss expectations means expecting losses - lots of them. He states, "As a trend trader, you should expect mostly small losses, some small profits, and a few large profits."
If it gets across this point alone,"A Short Course" is a worthwhile addition to your library. Kaufman provides an example to demonstrate a phenomenon traders in the game for the long haul have come to learn the hard way:
"In a normal distribution of 1,000 coin tosses, half of them would be single runs of heads or tails. Half of those, 25%, would be a sequence of either two heads or two tails. Half of the remaining, 12.5%, would be sequences of three in a row, and so on. Therefore, in 1,000 coin tosses, you can expect only one run of 10 heads or tails in a row."
In other words, in 1,000 trading days - or about four years - a trader could expect to experience 10 wins (or losses) in a row only once, that is, if trading were as random (normally distributed) as a series of coin tosses, which it is not.
So, your odds of winning with trend-following systems are better than your odds of winning a series of random coin tosses, but there are other challenges to having more winning than losing trades. Although markets are not random, you can still expect short-term random movements within a trend, major reversals at the end of each trend, and the time lag most trend-following systems experience when getting into and out of the market.
As a result, thanks to lags and unexpected short-term random movements, you are still subject to the effects of randomness. Given enough time, an experienced trader can expect to suffer 10 or more losses in a row. It is not a matter of if, but when.
When asked about realistic trading expectations, Thomas Stridsman, author of " Trading Systems That Work" and " Trading Systems and Money Management", had this to say:
"What is more important than how large your winning trade is when you win, or how many winners or losers you might have in a row. It is the mathematical expectancy of your strategy. That is, how much are you likely to win on average on all trades, winners and losers combined, and how much this value is likely to fluctuate in the short term.
"For an even further increase of peace of mind, you also probably are better off looking at profitable time periods, such as weeks or months, rather than profitable trades. Simply looking at a win/loss ratio is not enough."
If you think that doing either more or fewer trades would be a more successful strategy, think again. Kaufman demonstrates in "Short Course" that the more trades the trader performs, the lower his or her profits over the long haul. On average, longer-term trades generate more eventual profits. However, if you're a long-term trader, your risk of getting one or more big losses is increased, since you are in the markets longer and therefore exposed to risk for more prolonged periods. Bottom line, no matter what your trading style or preferred time in a trade, you will lose and lose big on more than one occasion.
Kaufman has the data to back up his claims. He has performed thousands of tests on various systems, and some of these are presented in Short Course. In one example, he tested Microsoft for 10 years ending Jan 2001 and covering a period when the stock moved from a pre- split price of $1.04 to a high of $60 in Dec 1999. It should be pretty easy to beat the odds following that kind of trend, right?
Using an 80-day moving average during the period to generate buys and sells the system, trading both long and short positions, generated a total of 88 trades. Of these, only 36 trades - or 41% - were profitable. And Kaufman comments in the book that "[t]hat's actually good for a trend system, which often has closer to 35% good trades".