This is a really excellent piece by Joe Duffy from another list. He's a futures trader but this stuff applies to stocks too:
*********************
Here is part of a copy of an article I wrote for another publication a few weeks ago. My success with the methodology described continues to be quite satisfactory in real time, and matches my historical test results. It represents a very major shift in my approach to trading. I think it took me so long to get to this point because almost everything about trading is the exact opposite of human nature, and for me this was an area where it was very hard to embrace change.
======================================================================================
......My natural disposition toward trading had always been to maximize my percentage of winning trades, rather then the dollar size of winning trades. I liked the constant reinforcement that winning trades, even smallish ones, provided. In 1999 my percentage of winning trades was 77%. However my average winning trade in terms of dollar size, was only .68 as big as my average losing trade. (These figures are from my trades using my intra day entry method).
Overall those figures still add up to a profitable trading methodology. However when the really violent volatility crept into the market this year in the S&P, it really exposed some of the flaws in my trading method. I believe that my trade entry method was, (and is) excellent. In fact, it had been so good, that for a long while I really had no set plan to exit my trades. I just got out when I was happy with the profit, the market hit some potential support or resistance level, or when I couldn?t stand the pain, or just thought the trade was wrong. In calmer markets I was still able to avoid excessive trouble like this, while booking a lot of gains.
However when things got wild in the SP, I found I could be offside 10 points in a very few minutes, and still not sure whether the trade was right or wrong. By the time I was sure it was wrong, I had a few whopper losers staring at me. On the flip side, the volatility instilled a fear that if I did not take a profit, that it would quickly evaporate with the violence of the moves in both directions. Needless to say, George Soros wasn?t the only one finding things tough in this new paradigm.
After far worse months in January and early March then anything I had seen for quite a while, I put the SP on hold for a bit. I knew I needed to do some things differently, because I was getting beat up. the way I was currently doing them.
During this time I remembered a discussion I had seen on one of the trading lists I am on. It was about the importance of exits and exit strategies. On one hand were the proponents of trading multiple contracts and booking some profits quite soon, with say the first third of a position. Then taking some more profit a little later, and then trailing the stop on the final third.
This general philosophy had always been one I intuitively agreed with, even though I never formally tested it--- probably because it fit my natural disposition to book a high percentage of winners, which is what taking small profits will definitely allow you to do. On the other side of this discussion was only one voice--- a gentlemen who really seemed very non adversarial?like he really did not want to argue this, but was doing so for the benefit of those who might be helped by it.
One of the points he made was that trading by taking small profits early, it was really like trading completely separate systems, and each system should be tested with its own exit to see if it is really profitable and/or is the most efficient way to trade. The undertone was taking small profits with high accuracy was NOT an efficient way to trade. I remember the reply to this from the major party on the other side, who stated that it "did not need to be tested" because one "just knows" it is correct. Oh Boy! In my experience anyone who makes this kind of inane statement about trading, is likely going to be wrong in that statement about 95% of the time. In other words, each exit likely DID need be evaluated as part of totally separate system, and that taking the small profits probably was NOT the best method. But I did not do anything about it at the time. After all, why change a winning system? But it did effect me to the extent that it was stored in my subconscious, just waiting to be heard again.
And heard again it was, as the $ size of my winning trades compared to losing trades became very negatively skewed by the S&P volatility, and my accuracy also went down a bit. The bottom line was I was finding I was on the wrong side of the P&L ledger.
So I stopped trading the SP and went to work on my exit strategies. My entry strategy was already well defined. The basis of it was (and is) to buy a dip in an established up trend, or sell a rally in an established downtrend, according to a clear set of rules. I had certainly noticed though, that since I was always trading with the larger trend, I had very often exited with smallish profits, only to see the size of the bigger trend move, dwarf my profit very quickly. So this is where I went to work.
In the end, I devised a set of exit rules that looked more like my entry rules, in that they were now very clearly defined. These rules took my expected accuracy rate down to around 50% winning trades, which I would never have considered acceptable before. However under these rules my average winning trade could now be expected to be about twice as big as my average loser. Not being a statistician, I was surprised to see that these exit rules actually produced DOUBLE the total profit on the exact same entries I used in 1999.
There is the caveat of the limitations of back testing on these exits, but I don?t think that will be an issue, as they really are not optimized. Thus far in real time, they are producing a profit since I adopted them. Not a huge one, but still a decent profit, and that is with a couple of really bad luck stop outs at top tick, on trades that would have been big winners.
The basic things I take from my work in this area are these: 1) The initial profit target should be 1.5 to 3 times the size of the initial stop loss, and must average at least 2 times the size of the initial stop. 2) Move the stop tighter when reasonably possible as the trade progresses, as a function of time and/ or price movement. 3) Move the stop to protect a small gain when a market shows a profit equal to one times (or a little more) the initial stop $ risk. 4) Profit targets are still preferable to trailing stops (at least for my entry method, which will re enter on a counter trend move again). 5) It is better to enter trades in times of lower volatility where a volatility expansion toward the mean can work to your advantage if the trend resumes, and also afford the opportunity for more modest size initial stops. 6) The percentage of winning trades is less important (far less) then the $ size of the winners versus losers. 7) All of these principles can be applied equally to intra day or end of day trading.
Trading is a constantly evolving business. We are always learning new things, and like any business, if you can?t change, you wont survive.
Below is a quote from William Eckhardt, who gained fame with Richard Dennis and the Turtles (the Trader Turtles, not the ones who sang "Happy Together"). It is topical to this discussion.
"One common adage on this subject that is completely wrongheaded is: You can't go broke taking profits. That's precisely how many traders do go broke. While amateurs go broke by taking large losses, professionals go broke by taking small profits. The problem in a nutshell is that human nature does not operate to maximize gain but rather to maximize the chance of a gain. The desire to maximize the number of winning trades (or minimize the number of losing trades) works against the trader. The success rate of trades is the least important performance statistic and may even be inversely related to performance." |
