by Joe T » Wed Jul 08, 2009 4:29 pm
Yes, Edward. I have two notebooks keeping track of different aspects of the trades (one more of a strict buy/sell journal, and the other a less strict/observational set of notes.)
So far, the bad trades fall into a few categories. To understand the comments, you need to know that this is a trend-trading system.
The first category is (1) Dumb trade. A closer review of the limit orders on this category leaves me realizing that I didn't give proper consideration to weaker trend lines and trades too aggressively. While frustrating, it is a good learning experience that should help choose proper entry points on future trades.
Category (2): Good trade with bad results. The trend line looks solid according to my "rules" and there are no issues with either the take-profit or the trailing stop. It's just a trade that goes in the wrong direction. Oh, well. I've successfully gotten past concerning myself with losing trades as long as I am confident that, overall, at the end of a month I'm a winner.
Category 2(a): Impact on money management due to correlation means that if a particular currency goes off in a direction that sets off orders, only to get stopped out, it means I can take a beating in the short term due to multiple trades going bad all at once. That is mainly what happened so far this week with JPY. The upside to this is that if/when the currency "does what it is supposed to do" I get the benefit. So, theoretically, I am limiting losses and taking a short-term hit, but at the same time setting up future opportunities when things go right. On the other hand, this can create a problem with having too much money at risk in a single currency, so I need to weight that.
Category (1) is beneficial in the long run, I think. It keeps you from getting too cocky! Early on, this occurred more often than it does now. I've adjusted my selection of trading trend lines and have become more selective about what I consider to be a tradable line. It may not seem that way with the flurry of activity this week, but that had more to do with JPY's movement through support on many charts.
As for Category 2 - with my strategy, one has to recognize that trends do change and that at some point price movement changes. This strategy presupposes that there will be times when the market changes where you will have setbacks. You try to limit the damage during those periods. Hopefully, this is just one of those weeks. So we accept these losses for what they are: an establishment of new lines from which to trade for future profits that should exceed these short-term losses. (Since I'm only into week 4 of this testing, it's kind of theory at this point, but I think it will happen. I expected these situations when starting this method of trading.)
As for Category 2a, I am evaluating the necessity of this approach. Do I want these large swings up and down? In my mind, despite the correlation, there is still a spread of risk in the approach that lessens the impact, and that my money-management of 2-3% per trade still protects me in these times. On the other hand, a series of setbacks from a single currency going haywire may lead to some issues, and I do need to strongly weigh the benefits versus what we actuaries would call "an exposure to catastrophic risk." Sound strategy on a per-trade or per-transaction basis is great, but if you put all your eggs in one or two baskets and something happens (say, for example, unrealized losses in otherwise profitable long-term mortgage-back derivatives that need to get marked to market sending your company into a liquidity crunch during a downturn in the real estate market, sending your company into a 98% freefall in stock price, and only avoiding bankruptcy because the government buys you. Hypothetically, of course...) then those otherwise "good investments" only need one or two downturns to kill you. For now, I am going to continue as is and try and monitor this effect.
One potential option I'm looking at is to eliminate redundant pairs. For example, the trheesome of EURUSD, EURJPY, and USDJPY has a redundancy. Arbitrage demands that these three pairs are in sync. Each pair can be "created" by trading the other two, within spread differences. So, I probably should only trade two of those pairs. However, I can also create each of those pairs by trading two others. EURUSD can also be part of the USDCAD and EURCAD trifecta. So, I need to play around with all these pairs to see how I can get the best spread of risk across currencies and at them same time eliminate some work without eliminating opportunities. I'll judge it by volume first, but not solely on volume. Otherwise I'd probably end up with all the USD pairs, which concentrates the risk in the USD. So that will take some thinking. Hopefully, as I continue to test, the best pairs reveal themselves and it sorts itself out.
Finally, while I am trading live, this is still a large experiment. There may be certain pairs where this just is not a good strategy, and by simply eliminating these pairs it eliminates higher-probability bad trades. But that will take months to work out, because I don't want to make a judgment on any particular pair based on a month. I'm guessing I'll probably go 6 months minimum (assuming the strategy overall continues to keep me afloat).
These are my thoughts so far. Further comments:
I realize going in I am breaking a major rule: Don't go trading every currency! Learn the ebb and flow of a single currency. Focus on it. So, why am I tacking them all? The reason is not because I believe this style works equally well across the board, or because I think I can necessarily learn ALL the currency patterns. It's because I don't know . I learn best by doing. So, my thinking here - right or wrong - is to throw crap against a wall and see what sticks. Money management alone probably saves me here. I'm not going overboard with risk per trade. I fully expect to weed out certain currencies over time for the following reasons: (1) redundancy - as stated above; (2) it's a loser with this style; and (3) I just plain don't get a comfort level with it, for whatever reason, and don't think I ever will. In the end, then, I realize I will have lost some money on certain currencies and I'm willing to make that investment for the future. When the time comes, I will unemotionally cut it from my portfolio and say good-bye. Just as a business division may need to be shut down, the USDMXN division of my business may be dropped at some point. But maybe not. It's a relatively cheap investment to give it a shot for a while and doesn't cost anything to shut it down. The second reason is, I want to be very selective in my trading. In a given currency pair, there may be only one trade per week, or even less on average. Perhaps it's counter-intuitive, but being selective on multiple pairs, in my mind anyway, keeps me from overtrading or getting impatient. I am almost always in a position or more, but on a single currency pair I may go days or weeks without a single trade. Now, this may be good or bad, but it will help me weed out the winners and losers even more because I will know that I traded all pairs on the best entry selections. So if a pair still overtrades and loses money, it's time to drop it. Time will tell. The third reason is the spread of risk. I'm big on that. These pairs are correlated, no question. But there are two currencies in each pair. So even though the USD part may be correlated on all those pairs, the other currencies will offset this, providing some benefit to diversity of trading more pairs.
All this could be B.S. and it's not a recommendation. It's my way of whittling down where I eventually want to be. Again, I consider initial money lost as an investment in future gains. And why not just demo it? Good question. I probably should. I just find that I end up focusing and working through things a lot better if I have skin in the game.
With all this, keep in mind an important fact: I've been at this less than 6 months. My account is up with this strategy, but overall it is not. Full disclosure so that you take any thoughts I have with a grain of salt.
I do plan on providing more insight into why I am choosing my entry points in the future. However, as I alluded in my opening post, I will refrain from that until this strategy is more proven - if proven at all. No sense clouding minds with gobbledy-gook that yields little benefit.