Very thoughtful, honest genuine post Shodson. Thanks. So nice to know that other people go through the same stuff! Looks like you are working with all the issues - most of which are personal - quite thoroughly even though this is a difficult phase.
After looking at several of your trade pictures the past couple of weeks, I have several times had the following thoughts/ideas:
1. I wonder if it might not be good to investigate just taking one pullback per swing?
2. I wonder if it might not help to have a longer term cyclic element in there to confirm short term trend since this is a trend-based approach. For example if I was using the 20 bar SMA as the pullback line, I might want to have a 60 bar SMA up as well to get the longer term trend perspective and then take the first ST pullback once both trends have lined up. Something like that.
3. How often do losing PB signals presage a reversal and/or how often are they getting stopped out prematurely, so to speak, i.e. they are stopped out but then later turn out to be winners. If the latter is not happening, perhaps failed pullbacks can be looked at as early heads-up reversal indicators.
Just random, non-researched thoughts. If not helpful, ignore them immediately! Everyone has to make their own methods and decisions. Seems that's the way of it.
Thanks for your kind words. I'm sure I'll get through it and come out a better trader. I took a nice gap trade today, which I'll post later, so that helped.
Thanks for the feedback, I'll address each one here.
1. That's one of the backtesting scenarios I want to do: once a new push is apparent, which pullbacks are more successful? The first? The second? I already manually have decided that most pushes don't have more than 3 good pullbacks in them, so I usually avoid anything after the 3rd pullback. My sense is that the 2nd pullback is the most successful, because the first one could be the beginning of chop or a deeper pullback in the previous trend, but the 2nd one is confirmation, but I will let the testing tell me. I haven't tested it much on the 600 tick charts I use because I hate how slow NT 6.5 is at backtesting tick charts. I'm anxious to get my hands on NT7 so I can fly through these sorts of tests.
2. I have this. The green diamonds that appear means a HMA(50) has crossed above a HMA(200), and a pink diamond means it has crossed below. This is a crude but effective way of determining larger-perspective trend. The indicator won't fire a long pullback signal (up arrows) unless the HMA(50) is above the HMA(200) (a green diamond appeared previously). I'm not sure these are the best numbers or indicators to use, but my backtesting showed this improved results, but I've only backtested on minute charts due to NT6's slow backtesting of tick data.
3. I haven't kept record of that. I believe that a large percentage of the time when I get stopped out it's because the trend is reversing. My mental memory tells me that if I get stopped out it usually doesn't turn around in my initially expected direction, but that I was just in front of a trend reversal and not a pullback. But that's something I should look into and measure in my testing to see if I want to adjust my stop size.
These are the sort of things I want to research more before trading these more often. I also find that I leave a lot on the table targeting 4 or 5 ticks with many pullback trades, so I should incorporate a small runner for double expectancy or more. That's something else I want to work on.
I also recently read an article in TAS&C about pullbacks and how pullbacks from bull moves don't pullback as far as pullbacks on bear moves. I've seen this be the case as I often have a harder time getting filled on a pullback when the price is running up, but getting filled on short pullbacks is easier because the pullbacks and deeper. This makes pullbacks on short moves a bit scarier because it does pull back so much, but I think it's the nature of the difference between selling and buying. It was an interesting piece. It's made me wonder if my pullback line for up moves should be tighter (like a smaller period EMA) than my pullback line for down moves.
12/18 - Well, I know that I said I wasn't going to trade anymore, at least until next week, but I later thought that if there was a gap play that I would take it, but that I wouldn't trade any other signals/entries, especially since it was a quad witching Friday. Well, MTG signaled the first gap play of the month, so I took it and got 3.25pts from it, essentially wiping out Thursday's losses (except commissions).
As you'll see in the screenshot they've updated their gap guides to include probabilities for the other factors they look at. These just looked too juicy to pass up. I considered increasing, even doubling, my position size because of my confidence in their system, but I decided not to on that day and not try to avenge my losses by taking larger risks than I'm willing to suffer. I've lost a lot of money in the past with "revenge" trades and I don't want to go through that again.
12/22 - There was a nice gap setup with very good odds so I took it. I considered reducing my position size a bit because of the bullish Santa Claus rally we are in but didn't.
Unfortunately I was in the trade all day long as it traded between my SL and PT all day, and ended up closing out the trade right before the cash market's close for a 2pt loss. I've never had a gap trade take all day long but I guess sometimes they do.
There were many times I wanted to get out of the trade, especially since I went in with a full position. Even though it was a loser I'm pretty proud of how I stuck with the plan for the most part, for better or worse.
12/23 - Picked up a cold yesterday and slept through the open. Of course, I slept through a winning gap trade. The setup was very similar to yesterday's so I probably would have been more cautious/skeptical but it filled quite nicely and quickly.
12/24 - Gap play odds were poor today so didn't fade the gap. It was a good call as it opened above yesterday's highs and stayed there all day.
Shortened market session, holiday, I'm not trading today.
MTG is taking the week off next week so I won't be taking any gap trades. I'm still not ready to start trading other discretionary systems that I have yet either. Plus, it'll be a low volume week, probably bullish, so maybe I'll buy some calls on the QQQQ but I may have already missed the party. Other than that I'm not going to be doing any trading next week.
I coded a reversal strat using cclsys's ATR bands today (available to Elite Circle members only). Win rate is about 50/50 but profit factor is high because my targets are always twice the size of my stops. If it's going to reverse it usually will reverse big, or it wasn't meant to be so I can have smaller stop losses. I tend to favor reversal strategies: they are easy to identify programatically and you can usually get large profit factors even with mediocre win rates. They don't work as well in trending markets (forex) but ES is so noisy and heavily arbitraged that it works there. Might work well in commodities, I'll have to try it.
It avoids the first 30mins of the session and doesn't enter any new trades during the last hour (blue shaded background). I found this improves performance. Knowing when to trade (the yellow shaded background) and not to trade (white background) is important.