Why a system may continue to work after everyone knows about it After 3 or 4 consecutive losing trades, 90% of the traders using it will have decided that the system no longer works, or they don't like a system which may have losses and will quit using it IMHO. This is where a traders' mental state is important, more so than the usual fluff about winning attitude, etc. In the famous "turtle experiment", Richard Dennis and William Eckhardt of C&D commodities set about training new traders. The story goes that the course was for 1 year. About two weeks were spent explaining a simple channel breakout method and some position sizing and money management rules. The rest of the year was spent dealing with the problems the students had following the rules. Many of the turtles went on to exceed their mentor's success. Simple System September 2004 losing streak 13 573.10 572.50 65.00- Long 14 570.20 571.40 125.00- Short 15 568.10 567.80 35.00- Short 16 574.20 573.70 55.00- Long
Yeah, a lot of new traders believe that 90% number (actually, I believe it's closer to 80%) of traders that lose is because the market is to chaotic to trade. But, it's because 80% of traders can't follow their plan for more than a couple of weeks. Even this simple method you shared will make money. But as soon as it loses a couple of days, 80% of the traders will give up on it or start tinkering around (optimizing) with it. And so begins the endless cycle of optimizing...
So the holy grail is sticking to ur plan? Is that all? But anyone with a backtester can find out the max. drawdown and profitability right? Why would 80% abandon their plan if they already know the drawdown numbers? I read stories on ET about people trading a method for months/years and ultimately failing... Not trying to argue...me just another newbie trying to learn.
I think sticking to your plan is a big part of the whole. Of course, I'm not saying you can just flip a coin (although some believe this) and make money if you stick to it. You need a method that's sound, such as breakouts, retracements, or reversals to name just a few. But, I've noticed a lot of traders constantly jumping from one method to the next and back again, over and over. One week it's divergences, the next week breakouts, the next retracements, and then go back and start all over again. I think that's why traders fail. They can't pick a method (plan) and just stick with it. So, I believe if you can find a method you like and then stick with it, you'll at least have a chance. I've never been a big fan of back-tested "Systems". There's just to many other factors that come into play. Will you be able to make every trade? If someone dies, and you have to go to the funeral, I guarantee you that will be the day you would have cleaned up. Or, the day the power went out...etc. I'd rather just have a simple sound method, and do the same thing over and over every day. Anyway...that's my two cents.
True story. I just came to this realization about my own trading last Friday after analyzing my PnL statement for every stock I traded last month to determine where I went wrong. Turns out if I had been disciplined enough to stick strictly to my real-time screening for trades, I would have been net positive for the month. Instead, I was net negative because of only four bad trades. One was an after hours, pre-earnings trade on a stock that dropped almost $2/share on me the next day. Two were stocks that I picked the night before from an EOD screener. They went up initially at the open, then tanked the rest of the day, but since I had seen that the EOD chart looked strong, I convinced myself that they'd recover before the day was over, so I just sat and watched as my losses grew. The fourth was one where one of the other guys in the room shouted out a ticker symbol which had been dropping, but had a huge uptick right before he shouted it. I saw the huge uptick and thought there must have been news on it, so I bought instead of sold short without asking him which way he was going. Ended up that was the only uptick for the rest of the day. I bought at the top of a 15 cent uptick and it dropped almost 50 cents on me before I could get out. Rules learned/re-affirmed this past month: 1) Stick to what works, even if you get bored with it. 2) Don't try and day trade after hours. 3) Cut losses sooner, no matter what the charts said the night before (or any other time). Cut 'em and move on. 4) Don't use EOD screeners for day trading. EOD screeners can't take into account news or other influencing factors the next day. (Though all of my picks did well within two days of my picking them - there were others I only watched and didn't trade) 5) Ignore the shouters, especially if it's not clear which way they're going. 6) Stick to what works, even if you get bored with it. Another thing about these four bad trades was that I was convinced that they were going to recover and foolishly averaged down instead of cutting my losses, even though I knew better.
I can't tell you what will work for you. What works for me is using a real-time screener to alert me to stocks that are moving right now and not stocks that were moving yesterday or last month. If you choose to try this route, learn some charting basics so you can spot patterns. Specifically, I find candlesticks are useful. I have a 3-month chart with one day long candlesticks, a 10-day chart with 30-minute candlesticks, a one-day chart with 15-minute candlesticks and a 30 minute chart with 1-minute candlesticks. All four are linked to each other and to my screening program, so when I click on an alert, it loads that stock into all four charts, where I quickly scan and decide wheter or not to act on the alert.
you run your strategy 100% until it fails and hopefully you have built up enough money so that if it ever does fail, you can start again. but proper testing is absolutely necessary and important to be able to run your system on 100% faith. if your going to be a system trader, you just have to do it. its been said many times in this thread and is absolutely true that those who jump from strategy to strategy when they feel a little paid WILL FAIL. the way i figure this whole thing is if i don't have any rules or strategies, im going to fail so I might as well adopt the rules and stragegy and run them 100% and whats the worst that can happen. ill fail. but ive given myself now the odds to be on my side.
I doubt anyone will easily abandon a strategy they thoroughly backtested. Unless the historical drawdown is exceeded by a lot. My problem is will obtaining a strategy that works in the first place. Sticking to it is no problem since i'll just automated it and go find another one.