Having spent countless hours reading, I have a few ideas I would like to run past people. Here is some quite background info about where I am at. I first started to follow price action and watched the videos of Al Brooks. Eventually I found Elite, and stumbled upon posts by DbPhoenix, the forever helpful NoDoji (your reputation on here is extraordinary!), then discovered the amazing journal of "geez" where he used a 1:2 risk/reward ratio to rake it in, which transformed into the posts from "puddles" who carried on in his footsteps. I have seen the power of risk management and completely trust the math/statistics of trading. I am humble to the importance of psychology and taking all emotion out of a trade. I am now doing research on developing a system/learning how to trade. As a side note, given all the continued importance placed on back testing, there are hardly any threads about how to do this. I see posts about various software programs out there for coding automated trading which seems to offer back testing, but these are quite complex. I also see that there are places you can buy historical data as well. But given how little information there is about this, it makes me think most people really aren't back testing and perhaps just learning from their own paper trading? Anyway, on to my question. So what exactly would be a good edge? Watching the markets for the past few months and studying a bit of price action, just throwing an idea out there, double bottoms seem to work great. But can it be this easy? If you bought every double bottom, applied a 1:2 risk/reward ratio, even if you were right 50% of the time you would be in the money (as geez has shown). This seems too easy, but has anyone done the calculations? Sure you need to define what your risk and reward is, but lets say you keep it well within the average daily swings of the particular stock and are looking for small movements that are attainable... call it 10 cents on a $20 stock as just a crude example. (I personally love watching DUST and NUGT... the volatility is awesome, and buying ETFs is free from my account (only pay to sell) so the comissions are well controlled is my thinking) How about gaps on open? Sometimes a stock gaps up, and keeps going up. But often I hear and see that it has to come back down to fill in the various levels where people are waiting to get out or in, so it comes down close to the previous close before it shoots up again. Once again though, what if you just found out that 50% of the time, a gap up does result in the stock slowly coming back down to the previous close, even if it does eventually take off again. You only need to be right 50% of the time anyway! So throwing these two ideas out there, can it be this easy? Perhaps mostly people don't stick to their plan, but if you can be right 50% about something within your parameters of how far away your risk and reward is, then isn't this system more than likely to work? Al Brooks states that when you are thinking you are right, you are only about 60% right, and vice versa. Nobody would take the other side of the trade if your chance of being right is 90%! I see that geez has 4 patterns that he mostly followed, and NoDoji I think loves to buy after a retracement with the trend (mentioned in the thread about why do people buy at the top), so is looking for an edge more easy than it looks? And how are people testing these ideas to begin with? I feel that most people just learn as they go so is all this back testing talk lost on most people? And lastly, other than in a choppy market, if I am going to find I am only 30%, can't I just always do the opposite of my original though and be 70% right? (I know that to be right you have to hit your target before you hit your stop, so doing the inverse of this might not lead to an inverse of your % correct ratio as your stops and targets would be different) Thanks so much everyone for reading and hopefully this will lead to a great discussion.
Seems like you've prepared well. But theory in this business can only take you so far. Nothing is as valuable as actually trading live on a small scale as a beginner to get a true feel for trading and the markets. An EDGE is paramount but I think its overrated for us retail traders. Besides, anything works..., anything - Fibs, Stochs, Price Action, Volume, Waves as long as you have tested it and have 100% CONFIDENCE in it. Because trading is about probabilities as you must know, if you can get a 60% win rate, and you learn how to manage a trade - which includes psychology you will be profitable. Once you start trading though and take your first couple of losses, how you react, and if you are able to trade your plan consistently and take the trade when you see your edge again will make you or break you as a trader.
Thanks ktmtrader... and its funny how you say anything works because I believe this as well, and its just a matter of finding what suits you. I have traded a bit live, and my gosh, did it ever get my heart pumping... which is obviously bad bad bad. I honestly want to start with just a $20 loss and looking for a $40 win. I would feel completely fine with losing 5 in a row, and hence $100, and it wouldn't phase me one bit but provide excellent learning experience. I am just very curious to find out if others have done the work to figure out what the probabilities are for extremely simple setups. Its like geez was saying, he wants to be able to teach an 8 year old what to do, and because they are so honest they would just put the trade on every time. Since I haven't come across any software I can use myself and have no data to work with, I wonder how well simple things like gaps turn out, or double bottoms or..... anything really! Just want the explore the power of the 1:2 risk/reward ratio.
That's true, backesting systems requires serious knowledge in computer programming. If you need 10-year's worth of historical Forex data (1 min time frame) you can download it here for free, for example: http://www.dailyfx.com/forex_forum/...967-free-strategy-trader-historical-data.html Happy testing.
Your edge begins with the knowledge you gain through your research and testing that a particular price pattern or market behavior offers a level of predictability and a profit to loss ratio that provides a consistently profitable outcome over time. Until you've completed this research and testing -- preferably manually -- and developed a trading plan that is consistently profitable, trading "live" is a waste of time and a waste of money. The attached may help.
Thank you for your kind words dbphoenix's post reflects my own thoughts on this. I saw something that appeared to work more often than not and I meticulously backtested it manually, keeping stats in spreadsheets so I could come up with a plan that involved filters (such as context, survivability of stops, pending news schedules...) and trade management. The double bottom and its two variations (failed break of the previous low and higher low) are basically 1-2-3 formations. Even with a 1:1 risk:reward they have a high rate of success; however, I advise doing a statistical analysis of as many failures of this setup as possible so can identify potential contextual filters. Some areas I found useful for developing contextual filters were: - Moving average slope or lack thereof - Inside bars and outside bars - Back to back opposing breaks (price breaks a previous bar in the opposite direction of a directional price move, then breaks the opposite way again on the next bar; this is Al Brooks' barb wire/chop action) - Trend/trend channel, wide range, narrow range/flag/triangle - Proximity to key S/R levels using the chart of a longer bar interval (as a 5-min chart trader, I reference a 60-min chart, and a daily chart)
the last true edge was seen sometime in 2003 before the bots and Jack Hershey students grabbed them all. Artistic License That thing with the teeth, that's the market. The thing IN the teeth, guess who.
You remind me of myself when I was first starting out. It took me years to master price action. I cant even say I've mastered it yet, but I can extract money from the markets fairly consistantly. . Choose an instrument you want to trade. Stocks behave differently than futures. Al Brooks book was geared mostly for futures I believe. Also, you need to get a read on PA (price action) to understand when double bottoms / double tops happen if this is what u want to focus on. Simply trading the double bottom without understanding why it has occurred will do your trading no justice. When I trade the ES, I look for certain things to be occuring to trade double bottoms / tops. The move that creates the first bottom usually occurs on strong momentum or reversal signs. The subsequent move up normally 'should' have stronger momentum, but you witness the price action is weaker than it should be instead. This gives you a 'hmmmm' moment that the master of price action picks up on. This means there's too much supply in the market. I like to call the move back down to create the double bottom a 'sell side flush'. The sell-side flush will smack that double bottom area, and now you hope for an entry signal. The sell-side flush is different from the minor pullback in a trend. You must recognize this. Its a complicated art to understand. That just barely scratches the surface. Get a read on price action, then start looking to develop edges. Be a master at reading price, and then the edges will just jump out at you. Most of us that trade, don't have this magical setup we use that nobody else has. Many of them are fairly obvious to someone who has stared at a 5 min chart of ES for 4 or 5 years straight (in my experience at least). so, 1) choose an instrument you want to trade 2) choose a short term and longer term time frame (longer term to provide reference. I prefer using a few moving averages for further reference 3) stare at how price moves. look at historical charts. There's an order to it. Its not all random, but a lot of it is. 4) Many will say the edge is not in the entry, but in the exit. Certain setups of mine, I know exactly how much heat to take. Years ago, I'd have panicked and sold. Thats the benefit of knowing how to read price. 5) You will start to see things repeat over and over. Now you need to backtest. Whether it be with software, or just going over past charts. It must be done. The level of detail used is up to you. 6) Begin writing a plan on paper. If you have no plan written in detail, do not trade. I wasted almost $30k trading without a plan as a newbie. Discipline is hard to come by when you dont have a plan.
Hi K P, Backtesting can be done by computer program or it can be done manually. I didn't become consistently profitable until I got serious about doing manual backtesting of my strategies, which are chart-based and very visual, and I was as objective as I could possibly be about my entries and exits. I'm a swing trader who uses daily charts and my approach has been to put January 1st of a past year on the right edge of a chart. Then I begin moving the chart forward one day at a time, pausing at each bar to assess whether there is an entry signal. I record the entry price, initially assuming no slippage occurred (not a problem in my markets), and then move forward until my exit plan gets me out of the trade and I record that too. I continue on, recording entries and exits until I get to December 31st of a later year. After that I calculate several things, including total net profit, net profit per trade, winning percentage, profit factor, maximum drawdown, net profit/max drawdown, maximum consecutive losing trades and sometimes avg win/avg loss (payoff ratio). For promising methods, I start making logical adjustments to the entry/exit parameters and try to make them better. Lately, I have worked hardest to increase the profit factor, which seems to improve several of the other parameters too, but not all of them. I don't use chart patterns and so don't know the best ways to trade them or even identify them consistently. If you want to know if double bottoms can be traded, you could define them as objectively as you can, decide how and where to enter (limit or stop order?) and define an initial stop-loss plus an exit plan for winning trades... then test it over a period of time that gives at least ~ 50 trades and see if it's worth pursuing further. During testing, you should be consistent in the taking of entry and exit signals, and also "ruthlessly empirical", an expression that I like very much. Basically, go where your testing leads you, dumping the bad and keeping the good.