Yep, the blowoff top was like a rocket ship blasting off! Nice to see trade-after-trade being brought to successful fruition using this method.
ES 5 minute Note - GLOBEX locked limit down for a few hours resulted in bad charts - first few trades are invalidated. Also, bad price at 12:45 caused stop loss on a profitable trade. With a momentum trigger and filter this would be a very proftiable system. Also, the Keltner and momentum exits are quite good. Another refinement can be buy and sell only when price is above or below "long term" MA, e.g. a 50 MA for 5 m ES. Another possible refinement, together with momentum, is Stochastics. e.g. a fast stoch 7,3,3 and a slow at 21,3,3 Will explore these the days and weeks ahead. With the markets exhibiting such volatility I am occupied in other trading so no time for system development as yet. These are rare market conditions - some don't like the volatility, I love it, a daytrader's delight. <img src=http://elitetrader.com/vb/attachment.php?s=&postid=2144361 width=800>
2m ES Same Globex Limit down caused spurious trades at beginning; also bad price at 12:42 pm stopped out short. Look at an example of how an entry would not be done if momentum fiilter is used: Take that long trade at 15:56. That's a rank bad trade! Firstly, I would not enter at 4 minutes to 4pm! 15:50 might be the latest entry in day trading. Next, look at the the momentum in the second panel. It has already turned DOWN, from an 'overbought' green. You wouldn't buy in such a case. Even a simple filter such as "buy only if close > previous close" etc would prevent these types of trades. See attached chart. Another visual price action tool - look at 9:50 and onwards .. see those green dots at the top of those bars? They indicate a breakout of x-bar high (set to 10). Note how these dots are drawn as long as HIGHER HIGHS are made - several consecutive dots until 10:02 .. then no dot as a lower HIGH is made. You don't need the dots to see this but it is visually much better with the breakout dots. As soon as dots stop ... you can exit a long, maybe with a simple trigger such as close < previous close or low < previous low etc. See the opposite starting at 13:32 , with red dots for breakout lows. That SELL trade at 13:46 can be exited at 13:52 or 14:02 The big blue dots are PIVOTS - highs and lows for x bars; again, a visual aid to determine where the price is in relation to previous swing high or low; of course, the pivots don't appear as soon as price turns! May take a few bars to do so but it gives a macro view of price movements. <img src=http://elitetrader.com/vb/attachment.php?s=&postid=2144386 width=800>
Jahajee you are perhaps too kind. Thanks. I am also going to look at trading this with certain stocks on certain days, they can have larger trend moves then any index.
Yes, he is demonstrating the exact same logic in his charts that I am using to create my results. First one must understand the SPM method. Yep, all three paragraphs. Then you must use a momentum filter to keep you on the right side of price action as the MACD establishes the trend. *** I'm going to be getting much more serious in my trading now that I have a system that probably delivers a 85-90% win-rate. This is the piece of the puzzle I was missing, so it's time to get serious (at least for a minute) and build my account so that I can cut-out the life of a working stiff (nothing wrong with that, btw). I'll definitely be around, as I enjoy the company, comaradship, and the feeling that comes from being around winners, - but I won't be taking such an active part in the Journal as showing my wins/loss, calling trades, PnL etc. But there's no need to, I did that to prove a point ... and I believe that is what I have done. Good Trading, Jimmy Jam
I'm working on my code for this in ninjatrader tonight. This is a good way for me to try to improve my abilities on that front. Thanks to Jahajees suggestions I brought the raw system from losing to being profitable. The win rate I have is still low tho. It seems it should be above %60 for the kind of tactic it is. I have coded using stochastic as a way to not take signals in a direction that is extended (tho sometimes the stochastic as a trendometer works, like the stochastic pop signal for instance). I coded a bar count at which a trade exits, which is something God mentioned. This seems to be a good way to increase the win loss ratio since I can exit a non-performing trade perhaps closer to breakeven. With the super efficient way ES is this seems better than taking a hard stop loss since that hard stop loss will likely be at some small spike and the worst possible exit. Other things I will look at is using a higher time frame to key off of. Another is if the trade is x profitable then I don't use the bar count exit. This way if I do get on one of those larger moves I can stay until a profit target. It does appear that the best trades hit fast and if it hasn't hit, that the time based bar count exit gives me a better price than waiting for a stop to get hit by one of those lovely spikes against the position.