PTF, I want to thank you for taking the time to post here and provide so much detail. I have to say that in some ways, the entire thread can essentially be distilled down to this statement, and the most important part of it is after you believe. I guess I was wondering if there was a more strictly codified statistical basis for your technique (given the earlier discussion of the evolution of pockets of normally disctributed price events within the larger context of an F-distributed day). I think, however, that your perception of the statistical realities was your starting point, and that the development of your actual trading rules involves the discretionary evaluation of the end of the move. This is not to say that you could not or have not developed a very sophisticated sense of when a move has ended; if you are trading this technique successfully, it is the one thing you must have. The fact that your system did not cause huge losses on the heavily trending day I posted earlier means that you can somehow 'sense' when to invoke your fading system and when to wait. I truly believe that for anyone entering the intraday markets, an understanding of what you and f9 are talking about re: the conceptualizing price action in terms of the areas that the most action is occuring and then playing moves that stretch out to the tails of that distribution will go as far and probably further than looking at MACD, RSI, or any of the other canned indicators (which I have found to be of limited use).
contrary to common perception indicators are viable tools to systematic trading. the obvious lag should not be much of a concern since for how much behind price, signals mostly generate within a tolerable delay, often trifle to achieve acceptable exposure. one indicator negligibly behind price is Bbands: price bars touch or violate bands as you see them printing on real time charts. signals generating off Bb's hatch quality trades, often initiated in much attractive price areas: tops and bottoms. i for once do not rely exclusively on indies but i grasp the tool's potentials thus often use it to aid the decision making process. alone the constant bashing of indies by the majority should make you pause for thought. as seen in this ES chart among 6 or more buy signals just half may translate into profitable trades and the other half into losses. yet hardly any losses come from all sell signals that would potentially generate 5-6 profitable trades: <img src=http://www.elitetrader.com/vb/attachment.php?s=&postid=1733083>
Not surprisingly, the only indicator I take trades from is a version of BBands called Aberration (at least something I devised which is based on Aberration). However, gnome, in his long thread tonight answering questions, said that he uses stochastics, so there you go. It's not the tools, it's the workman, right?
To the OP: Obviously, you were trading the short time pattern. But if you would take a glance at the whole picture, clealr the downward rally was getting to much and was due for a correction/reversal. But you can always try quant methods, I heard they work fine.
Bitstream, This comes about because the longside behaviour or the ES is quite different to the shortside. You need two distinct approaches as the "one size fits all" will lower your winning % regards f9
thx for the tip mate. by any chance, have you sampled it? if so and it's not too much of a chore can you show us your findings? i don't have a standard size valid for each position as i tend to average down over different price levels with a flexible quantity for every trade... still, interesting to know about it. i was guessing that it may be due to the recent downtrend yet the fact one of my system buys after 4th down bar and sells after 3rd up bar lends support to what you say.