I have read this article in Sep2003 issue of stocks and commodities about day trading emini with p&f. Strategy looks ridiculously simple and execution chart has shown profit of 55 points by trading only one contract. It is stop and reverse strategy. Article shows 32 trades on 1 min chart. I have used the same settings as mentioned in the article, but never came close to 32 trades, it always goes to 60+ tardes a day for the past two weeks which is not humanly possible, but have to admit it incurs no losses and you catch the bottom of the trend if you are fast in executions and fills. You need automatic execution system. Is there anybody who may be trying this strategy, or knows about automated strategy trading.
I think I read the article (seems like it was before Sept though) - it didn't really explain what the method was, other than P&F. The trades didn't take into account bid/ask spread and also didn't take into account realistic fills based on trading without hindsight because of the built-in delay of P&F charts. In other words, the fills were unrealistic. And like you I couldn't match up the trades to the timeframe. I think the published sample results are bogus at best, and that actual implementation of a short term SAR P&F system might even lose money because of the bid/ask and the delay inherent in P&F. I hope someone can prove me wrong.
This is a link to an excerpt from the article: http://www.traders.com/Documentation/FEEDbk_docs/Archive/092003/Abstracts_new/Steele/steele.html
Hi TriPack, I can't get the above link to work. Any other way of getting a copy? How are you doing with your "Equal Range Charts". I programmed this and find it quite interesting. I have not done any simulations on it though? nononsense
This article was about emini, so there is no question of spread as there is only tick difference between ask and bid, and metodology was to trade whenever row changes color. Since author only waited for one box and he was using .33 box size, the loss is limited to a tick or two, which can be overcome by sell on ask and buy on bid, and obviously fast market would not let you do that, therefore maximum loss is .50 which are two ticks. I have tried this in realtime with only one box for reversal and it seems to work ok except insane amount of trades which are hard to execute. Increase in box size to .66.or 63 would result in 10-15 trades but it would increase the losses to around 1.5 to 2 points and it would also not predict change in trend, but would take the lot of noise out especially during lunch hours. I am using esignal which let you do all kind of settings as mentioned in article, and author has also mentioned that you need to observe p&f charting before you determine box size suited for your trading style.
The thing that bugged me was that Figure 3 (from the article) was supposed to be .33 box size with a 3 box reversal. Then for the results he used .33 box size with a 1 box reversal. However, if you compare his trades to Figure3, they seem to be an exact match. Thus I say his fills are unrealistic because they base the trades on the 3 box reversal yet the fills are based on a 1 box reversal.
The site was down earlier, looks like it's up now. The article on the website is just the first part of the article. Basically he says, look at your bar chart for the timeframe you wish to trade in. Plot a point and figure chart, tweaking the values until your desired patterns show up. He was using a .33 with a 3 box as an example. I'm still studying the Equal range charts here, trying to get an understanding of the pros and cons. Not much additional to report, however.
I do agree with tripac on this that author has used only one box reversal, yet he shows chart with 3 boxes reversal. Fills are unrealistic as well, its like lightning fast which I think no human can do.