GG, You have to start thinking more mathematically. All you are doing is applying a kind of filter on price bars, which are themselves a filter on ticks. There is nothing "wrong" with it, but you better understand the lag that ANY filter introduces into price movement. Another thing to understand well is what others have hinted at - do things work because lots of others are looking at it - maybe they don't work because others are looking at it - or it is in the blurred line between the two that we find truth? I think that this question is similar to asking whether an electron is a particle or a wave... If you know how to do statistical analysis, fire up Tradestation or Wealth-Lab and start testing (doing this correctly is another set of skills) and see what you get. Even if the results are discouraging, it is THE JOURNEY that will make you stronger, as the PROCESS of thinking about the markets in this way may [will] lead you to the nuggets. Brother Candle, I agree with you 97% - I don't want to go into the 3%. nitro
Software. Remember that software that switches your trades from buys to sells and vise versa? Well guess what, it inverts your charts too! So a perfect long set up is so obvious a short that you take the trade and the software switches it to beat the market on a consistant basis! Next year, the developer is going to invent "see Tuesdays chart on Monday" software, but it too will be in Bassackwards style, and only paying members will get the decoder ring!
******this is not me talking here********** Let me ask you what would you think of a chart, which takes the same parameters as a standard, bar chart but your daily high, low, close and open data had a time frame different than daily? You would laugh. I like that laugh! It tells me I have no competition with the idea. I will give you an idea and an example here. Let us say you take and make a chart, which starts one hour and fifteen minutes prior to the close of a market. We will justify this by saying that the most important trading data for a day is in that last part of the trading day. We will call this the opening for your next day's chart. One hour and fifteen minutes of today's trading is already on tomorrow's chart. We will call this tomorrow's support and resistance. We continue to chart tomorrow until one hour and fifteen minutes prior to the close. This closes out our day's trading chart. Ok I think you see what I am getting you to think about. Now keep in mind I am not giving you my way of charting but using this as an example of how to change your behavior when it comes to charting. Most traders will never chart this way for several reasons and that to me is good. They can't get the data this way as they may only get it out of a newspaper, delayed or through a broker. Other reasons exist which prevent them from getting a different outlook chart. I feel you need a jump on tomorrow's trading to get the edge. The edge to me is important but not as important as execution. It is just that with the edge, you can get better execution. You are ahead of the game because you are in front of the day traders, funds, scalpers and position traders because you are not using their data to follow them. Instead you are using your data to look beyond their view. By using rules one and two, you can establish a plan which is a little more remarkable than anyone would think. You will have to do back testing and research and most traders can't even come up with that data yet even today. I guarantee in the future there will be those who look at what I have done lately and say it is time to make the computers earn their keep. You see the frontier is just now opening! The sharpest trader with the most intuition will win here. I want you to remember where the idea of "Different Outlook Chart" came into play. It started with me when I was thirteen. It can start with you today. Do your research! Do it again! Learn what different outlook charting can do for your trading plan. I have given ideas of what my criteria is in trading certain situations so that an understanding of where I am coming from shows up. Vary your data times. Use fifteen minutes, half hour, half day, first hour and other time frames. I am giving you a gift here and someday it will hit you. Just don't take too long to see what is behind all three doors. I still want to point out that we will see different outlook charts as the years go by and they will get better and better. There will be a day when they are followed closely enough that they no longer have the same value. In trading you need to change the odds to your favor. By using rules one and two you are moving in the right direction. By using your own mind, you are doing what that computer programmer did many years ago. You are looking at a different view. An artist will view his subject material from all angles. Shouldn't you?
The fallacy in all of this is thinking that your unique time frame will give you a "better" view. What you are guaranteed to get is a "different" view, but not always better. While you may be trying to get on the train before everyone else does, your "different" view can just as easily miss the train altogether. I suggest your results will be about 50/50. I suggest that if you have a positive expectancy strategy that works on 10 mins bars, that same strategy will produce almost identical results on 10.333 minute bars, and almost identical results on 10.45345345 minute bars. P.S. Tick-charts are used by many to see a slightly different view. The same strategies that work well on tick charts work well on their approx. minute equivalent chart.
I'd like to see 'calculations' that a single, medium class CPU could not handle standing on its head, with room left over to play a vid, an mp3, and download email in the backround.
Lobster, If your trading strategy bogs down a Pentium chip, I suggest you move to something a tad bit simpler -- unless of course it is working well for you, but personally I've never seen a system more complex than some of the differential equations I did in high-school. In fact, the more mathematical the trading system, the more it generally breaks down. aphie
GG and others, an interesting chart type I played with a few years ago are what I call "rolling" n-minute charts. Say you consider 5 min your core time frame. You then construct 5 min bars at 1 min intervals: in each bar you include the past 5 one-minute bars. In the following example I use the time of the first tick as the time stamp of the bar. So e.g the first bar of the day would be 9:30 EST Example: 10:00 bar includes data from 10:00, 10:01 ... 10:04, The 5-min rolling bar at 10:01 includes the following 1-min bars: 10:01 .. 10:05 and so on. Looks interesting and has built-in "volaility" to some extent. I'm sure others have come up with similar ideas and may even have a name for it.