Rol, these charts are very informative. It's been a long time since I've seen a thread this candid and interesting on ET. I do find it interesting that your losing days, although they occur less, are typically larger in magnitude than your winnings days. It seems like lots of small winners vs few large losers (although there are also a few large winning days). Is there some aspect of your strat that causes this behaviour? And how does it affect you psychologically?
Hi Kohanz and thanks for the continued interest in this journal! I have not been getting a whole lot of questions so I was wondering if it was interesting. I am big on data and charts because I can see better what is going on with the strat, like how you noticed the fewer but larger down days. There is not a whole lot of emotional attachment when I let the strategy perform without intervention. Sometime I kid myself by saying that is just a savings account with a highly variable interest rate! On down days, I realize that draw downs are an aspect of my strat that I have accepted. The $1581 down day on Mar 3 was about the only one that gave me pause for the quarter. But if you look at the market during that period, it continued even lower, so it would have been more devastating to my equity had I not exited my long only strat on the 3rd. Being in cash then allowed me to capitalize on the lows a few days later, making back most of the losses and gaining percentage points on the market. I discussed this earlier in my journal when it was happening. Even though I donât use stop losses in the usual sense, I am using back tested exit rules that will exit a loss without hesitation, when a person might hesitate. I think it is the draw downs that most who have attempted rtm trading have a hard time dealing with. I think this is where automation can help. When I look at the results of my system, I donât question that it is a valid system for short term trading. I noticed that questions and interest had picked up when the market was falling during the Japan earthquake. I guess that is just human nature. This weekend I worked out the details of my new strategy that shorts gap downs at the open of companies that reported that day before markets. My rationale is that if it had gapped down, then the earnings report disappointed investors. I will report how it works out.
Rest assured I read every new post as it's written (I am subscribed), but my day job has been quite busy lately which has relegated me to being an observer more than a participant. You're basically accomplishing what I would like to do within the next couple of years. I also have a day job that I plan to keep, but feel that operating a modestly profitable automated trading system as a source of additional income in within reach for someone with the technical knowledge who is diligent with backtesting and money management. I love the chart and I like the variable interest rate perspective, hadn't heard that before. Out of curiosity, your strategy is long-only - have you ever tested a short-only strategy with similar logic? For example, it seems like you buy gap downs, or something like that - have you tried short gap ups in backtesting?
Thanks for subscribing! That means a lot. I think having a âday jobâ helps to avoid stress from having to generate income from trading. One reason I got into auto trading is to lighten the effort needed for trading. Now I am more of a âcoachâ watching my strat execute my game plan. The markets open at 0830 here in Texas, and my alarm clock is often the words âOrder Filledâ. I then check to make sure everything is operating as it should. When my exits fill at the market close, I am at work away from the trading platform. I know many will say it is risky not to monitor your system constantly, which would definitely be true for scalping type systems. If for some reason I wasnât connected to the trade server at the close, the positions that should have exited would simple exit at the next day open. I know there is a risk of a gap down overnight, but so far I have not been bitten. It has been my observation both in backtesting and just watching stocks, that fighting the trend to the upside does not work . Here are a couple of recent links to some current popular journals on how it has worked out for them: http://www.elitetrader.com/vb/showthread.php?s=&threadid=213286&postid=3141988#post3141988 http://www.elitetrader.com/vb/showthread.php?s=&threadid=216290&postid=3141163#post3141163 Besides, I get too emotionally attached to stocks when rtm trading the short side, likely because it has not back tested consistently. The next few days I will try to reverse my code logic to a short only strat and post an equity curve for the past 4 years from Amibroker. I do have my new "gap down on earnings miss trend following strat" in play, so we will see how that works out come earnings season.
I saw a little buying action around the open, with EW and RBN turning out to be nice day trades. I am appreciating how crucial it is to be well capitalized. Being under funded forces you to take on size and leverage right from the start, so you better hope you are right about direction. Well, I looked at reversing my strategy buy/sell rules to short/cover and I was surprised with the results. A few years ago, when I tried it the results did not look good. I would get stuck in some trends going against me too often. Since then, I have refined by rtm long only strategy. Now the short side results are looking much more promising. In 2008 it had a 250% return, when the market was tanking. It only makes sense to incorporate both long and short strategies to the markets if possible. I just need to come up with one that back tests well to build my confidence in it. I will start out small to be sure the strat is functioning as it should. I think by using proper position sizing, it will help with psychology. As with many things, there is a right way and wrong way of doing something. So I think there can be a right and wrong way to swing trade. Hopefully, I can get it going by next week. I will put up some equity curves in the next couple of days of long only, short only, and long/short equity curves over the past 4 yrs paper trading for comparison. The past 4 years will be good for testing because they have seen some extremes both ways, as well as sideways movement. What I want to see is a smoothing of the equity curve and less overall system draw down Real-time Account Net Worth $62,772.43 Beginning Day Account Net Worth $62,327.88 Real-time Buying Power $111,686.02 Real-time Cost of Positions $13,743.00 Real-time Unrealized P/L $49.42 Real-time Realized P/L (4/4/11) $295.13 Entries: 10 Exits: 5
Hi Rol. Thanks for the journal. What you said about decreasing risk by diversifying across a lot of small positions has inspired some recent work on my ATS. Also, I'm intrigued by the idea of not using stops or profit targets. My system currently uses both and the results, as you probably know, are both good and bad (less bad when, I think, when you have an equal number of long and short positions, but I have been having a hard time coming up with a short strategy that works in this market). I haven't quite figured out how a strategy without stops would work. What kind of exit rules does your system follow?
Hi Rol, Interesting thread, though I only read the first few pages. Seems your trading has a few similarities to my own in that you are automated, trade equities, use Amibroker and you trade intraday. Seems you are much more frequent than me though. I starting trading automated through Amibroker and IB around 6 months ago. Have only done about 130 trades. My exposure tends to be very low as my system is quite opportunistic (haven't made a trade for 2 weeks) but trades frequently enough to at least make money. My results are on my blog. thetrendfollower.blogspot.com Keep up the good work brother I will be following this thread with interest.
Hi In2Deep and thanks for checking in and enjoy your new house! Iâve been reading your journal, as well as ElecEquityâs, PropTraderâs, and archived oneâs. I think we all can learn my catching themes of what to do and not do from people with skin in the game. When diversifying across symbols one must still be sure they have a back tested system, otherwise what does not work on a few symbols will still not work on many. I believe portfolio level trading allows you to âplay the oddsâ better, while automation keeps the mmâs off balance ( if only they knew they were trying to play off the emotions of a Pentium 4). It seemed to me while reading your journal that you were trading news related stocks discretionary and not with an ATS. I am poor at discretionary trading long term. Sure I have had my windfall trades, but the false confidence generally caused me to give it all back. I made $50,000 overnight in early 2008 on a small biotech (Encysive Pharmaceuticals) buyout by Pfizer. I suspected it was going to be bought out because the CEO was saying that he was in talks with an investment bank in order to âIncrease shareholder valueâ. I then proceeded to give it all back later that year by counter trending the down market through ETFs discretionary. What is unreal is that my auto strategy would have come out ahead had I had my code in place at that time. Now, I have no desire to trade discretionary because I know in myself that I cannot control the risk. Besides as my balance grows, it will become necessary to automate to be able to put on the number of trades required to diversify. I have read how others have had a hard time shorting, like during the recent earthquake crisis. What the market was doing was gapping down overnight and then trending up during the trading day. That was when I was using heavy margin to buy the dips at the open and it saved me. That is another reason I donât like day trading because you totally miss out on the overnight gaps in your favor, while trying to scalp 2 or 3 ES points. I am looking for the bigger moves that work even after slippage is accounted for. Naturally I canât go into details of my entry and exit rules, otherwise everybody would be trying to get in and out at the same time. I occasionally post charts of my trades so it should not be too difficult to see what I am doing. I can say that I am buying into weakness and selling into strength. The trick is to discover what constitutes weakness and strength. This is where back testing and optimization come into play. I work in a clinical laboratory and occasionally use microscopes. Anyone familiar with microscopes knows that they have 10X, 50X, and 100X objectives; or low, med, and high power. In analogy, the universe of symbols one tracks is the low power, the 50X is the scan you perform the night before, based on your scanning criteria before the markets open, while the 100X is where your strat zooms in on the individual symbols, based on additional criteria, that results in a hit, or not, of your buy trigger. I may be proven wrong over time with my strat, but it survived a recent major earthquake, and nuclear accident, so perhaps I am onto something. There is nothing wrong with learning to paper trade a portfolio until the required $25,000 in funds are available, even if it takes years of saving.
Great GekkoToBe, a fellow autotrader! Does Amibroker execute your trades? I have heard that it is unstable for real life order execution through IB. I only use it for portfolio backtesting. Once Tradestation releases their Portfolio Maestro backtester, I will probably switch to that fully. I do hold overnight, so I am not a pure intraday trader. When trades turn out to be day trades it is because they reversed the same day I entered. In back testing I noticed when I became too choosy about my entries, it resulted in limited market exposure. I may have nice expectancy, but long term returns were still reduced significantly. For me being in the market about 66% of the time seems to work out about right. Although, if I go with my short strat in addition to long strat, I may find myself in the market all the time!
Thanks. I actually do ok day trading in simulation mode--mostly playing the volatility off the open (my day is often over after the first half-hour) but as I've painfully documented in my journal, emotional issues were too overwhelming when I started doing it for real. I plan to get back into it again when I have a little bit of a cash cushion. The type of day trading I seem to be good at is more akin to playing a video game than anything. And I've always been a pretty good gamer, so maybe I have an edge there. Anyway, it is a very risky style of trading so I would never advise doing it with money you can't afford to lose. And I can't afford to lose that money anymore since my wife and I have decided to buy a new house. Don't you worry about having too much long exposure? The market could have easily continued downward on those days. I have been searching diligently for an automated short strategy that will work--even if it only performs a little better than break even. It seems just having some sort of short exposure would act as a hedge against big gap-down days. Was worth a shot I think you're definitely on to something although the amount of long exposure would worry me. Not that I know any better but maybe if you can figure out a way to hedge that risk then I think you're in perfect shape. On that note, do you gradually increase your exposure over time or do you go all-in when your signals start going off?