Okay Fifo, here is what I think you are talking about Price touched the longer MA and so a possible trade sets up fading the test. But it goes wrong immediately. Entry was at 1230.25 with a 2 point stop...normally I would just let it run until stopped out, however what you are asking is "is there a way to simply reverse and take a long?" Yes if you wanted to, you could make note of the following 1. price does not take out the 200 in a strong way, instead it simply consolidates. To me this means a battle is going on to determine where to take this market. 2. As price continues to consolidate notice the slope of the 80 period MA (blue line). It starts to turn up indicating a possible change in momentum. 3. Notice the "tests" of the 200MA. For some reason, when you see three (3) tests of a price point, it usually signifies a pending move (either way). 4. We see a small "peekaboo" move up, a retrace and then the move up begins. Its always easy to see in retrospect of course but I have placed my indication for a long entry as seen on the chart. I hesitate to suggest this to "struggling traders". If you have skills and the ability to think clearly under constraints of time, well that is another case. If you have that kind of skill, you probably don't need my advice That long entry at 1232.50 would have gone to a high of 1241.50 for a profit of 9 ES points. Scale outs at 2,3,5,7, would have got filled. I would have got stopped out at B/E on the balance (I wouldn't have got the full 9).
Thanks for the explanation. So if I have this right, you would still not go long the breakout but wait for the tests from the opposite direction then enter. Makes more sense as the stop could be tighter. If you miss the move, so be it. A lot to think about in the middle of a trade. What are your thoughts on automating. I guess it would be nice to run some simple backtests but how could you go about defining "approximate" tests of MA. I no longer trust myself at discretionary trading after a few large screw ups. Even partial computer assisted trading is better for my mentality.
I understand. Discretionary trading isn't for everyone. Your first challenge is writing a rule set that accurately reflects the system. It isn't easy. I have had to hire folks to do mine so I can't claim to have that expertise. I wish I could be of more help Good luck Steve
No worries. You've been a tremendous help. I will keep tinkering with the concepts and trade sim discretionary until I am confident. In the meantime, I will see if I can come up with a rule set from my observations.
Some additional thoughts Fifo One of the things that could cause problems is the way I use the data. For instance, in evaluating the viability of a setup I often look at the "slope" of the Moving Averages as well as the relationship (distance) between the longer (200/White line) and shorter MA's (80/Blue line). I could be wrong, but I am uncertain as to how that data could be programmed. Also I sometimes take trades based on my performance. If for instance I am having a good day, I might take a chance on a marginal or lower percentage setup. In contrast, if I am not doing so well, I will pullback a bit, waiting for setups that tend to deliver bigger winners. Other issues that you may want to test include the possible differences between Simple, Exponential, Weighted and Volume Weighted Moving averages. My 2401V chart uses weighted MA's. My 377tick chart uses a single weighted 80 period MA.
Yeah, slope and distance are pretty straight forward from a programming perspective. I would imagine a test of a rising slope from above is a higher prob long and vice versa.. Distance, not too sure how to interpret that but the real problem comes in when using MAs of X length. Using 200 and 80 could work but if testing shows 50 and 250 to be better then we get into curve fitting. So picking the MA periods would have to have some sort of reasoning behind it. As for trading based on performance, I would put a daily profit target and either stop and avoid overtrading or perhaps reduce size. But I could be completely wrong about this until I can see the stats. Of course if the system stays simple and has high win rates then maybe trading might as well continue. Good thing to remember about different types of moving averages. When using volume charts the dynamics of averages changes as volume is constant per bar, it tells a different story.
While I am not a programmer, I am a researcher. One thing that I know to be true is that looking for one single best MA or pair of moving average will result in a curve fitted system....If the system is going to work (and I have done this much work) it will work with a range of Moving Averages (I am talking about the absolute number not the type).
here is today's open folks I really can't do this in realtime and manage my trade, so I post it here after the fact. I am still short at the time of this post. I used Market Profile to get short above the longer MA (White line) but had the opportunity to re-load (add to shorts) on the re-test of the longer MA as seen in the chart I had a preliminary first target of 1248 at the pivot and you can see price hit that twice. My initial entry short was at 54.50 and I added just below the re-test at 1252.50 I may be forced out at 10:00 due to news (I often get out if there is a news event or report at 10:00 EST).
and I am out of this one with 10 ES points I anotated the chart so that others can see how it worked out If any of you are interested in the MP side of it I can talk about that (a little bit of it anyway). Except for the event at 10:00 (fed speaker) I would still be short looking for a test of the last session's value area high