Hey KS96, I check out elitetrader every once in a while, usually to take a look at the blotters in the p&l thread, and often I seen your thread in the newest post section. I personally trade spotfx, so I cant totally relate to your trading style, however, I traded the same time frames as you. The 5M/1H I must have traded for almost 3 months, in the very beginning of my trading growth. I found that the 5M/30M increased my profitable trades by 17% overall. That figure is actually specific to my system, so Im not saying it will do it for you, im just showing you the amount of increase was substantial. Alexander Elders gives a very good reasoning behind this. I decided to go ahead and find the research for you. It is attached. My hat is off to you. Your strength to look the drawdown in the face and keep trading shows a great sign of trading skills. Keep up the great work. Please do not interpret this as negative criticism, I was just making an observation. Sincerly, - secXces "Copy from Alexander Elders Book "Come Into My Trading Room" ------------------------------------------------------------------------------- TimeâThe Factor of Five A computer screen can comfortably show about 120 bars in an open-high-low-close format. What if you display a monthly chart, each of whose bars represents one month? Youâll see 10 years worth of history at a glance, your stockâs big picture. You can display a weekly chart and review its rallies and declines for the past two years. A daily chart will show you the action for the past few months. How about an hourly chart, each of whose bars represents one hour of trading? It will let you zoom in on the past few days and pick up short-term trends. Want to get even closer? How about a 10-minute chart, each of whose bars represents 10 minutes of market action? Looking at all these charts, you quickly notice that markets can move in different directions at the same time. You may see an upmove on the weekly chart, while the dailies are breaking down. An hourly chart may be sagging, while a 10-minute chart is rallying. Which trend to follow? Most beginners look at only one timeframe, usually daily. The trouble is that a new trend, erupting from another timeframe, often hurts traders who do not look beyond their noses. Another serious problem is that looking at the daily chart puts you on par with thousands of other traders who also look at it. Whatâs your advantage, whatâs your edge? Markets are so complex that we must always analyze them in more than one timeframe. The Factor of Five, first described in Trading for a Living, links all timeframes. Every timeframe is related to the next higher and the next lower by the factor of five. There are almost five (4.3 to be exact) weeks to a month, five days to a week, and close to five hours in many trading days. We can break an hour into 10 minute segments and those into 2-minute bars. The key principle of Triple Screen, which we will review later, is to choose your favorite timeframe and then immediately go up to the timeframe one order of magnitude higher. There we make a strategic decision to go long or short. We return to our favorite timeframe to make tactical decisions about where to enter, exit, place a profit target and a stop. Adding the dimension of time to our analysis gives us an edge over the competition. Use at least two, but not more than three, timeframes because adding more only clutters up the decision-making process. If you are day-trading with 30- and five-minute charts, then a weekly chart is essentially irrelevant. If you are trading market swings using a weekly and a daily, then the wiggles of a five-minute chart are no more than noise. Choose your favorite timeframe, add the timeframe one order of magnitude higher, and start your analysis at that point. -------------------------------------------------------------------------------
You may be right. I am thinking of getting the 1h down to 30mins so I can have more granularity. However, I am not really expecting a great improvement because the decisions I take based on the hourly charts depend also on price & volume patterns on daily charts. I may get better entries/exits though... I will see what I will do... Thanks for your supportive words. I am not worried about drawdowns. Actually, I am around $800 up (+6.1%) since I started this journal (7 trading days) thanks to the 2 winning trades I am still holding. My trendfollowing is around 70% losing trades. When some trend sticks, I recover. I am targeting to a 5% return on equity per week. Since March 2006, I manage to average to something around +4.2%/week. I believe I am on track. In summary, to make $1000 bucks in today's markets, you have to make $5000 and then lose the $4000. If you know what I mean
Rather than move 1hr down to 30mins, would you consider moving 5mins up to 10mins? I remember reading an Acrary post on the intraday trendiness of the S&P. He was of the opinion that the market didn't start trending until one looked above a 30mins timeframe. For interest, here's a trend following system he provided that used channel touch triggers, as opposed to breakouts. http://www.elitetrader.com/vb/showt...hlight=acrary trend following sp&pagenumber=1 He mentioned he was planning to "morph" this character-based system into one with an edge and share the results. Sadly, he never did. However, if we think about his edge test and abstract away from it, maybe we can find a clue. I believe I'm right in saying the test examined a strategy's ability to beat a basket of random trades of the same direction/length via a Monte Carlo procedure. Plainly stated, he was looking for a system that performs better than random. This infers he was trying to find non-random moments to trade. So, what market conditions mean a trend following system is operating in a non-random fashion? Trend is established? Volatility has contracted? Previous bar tells us something? Add to an existing winning position? Use the system as an entry and exit methodology on top of another insight, ala a Crabel ORB + NR4 set-up? Something else?
Currently I need the 5min timeframe, because it gives me entries with a risk of <2% of my equity. My account size is still too small for 10mins. (That's the main reason I left the DAX Future out. I had to go down to 1min, and still couldn't find enough entries with acceptable risk for my equity.) Anyway, I will try 10min in the future, if all goes well. Concerning Acrary's thread, I find the distinction between edge and character rather questionable.
OK, understood. Regarding edge vs character, I take it as non-random fat tails vs observable but unpredictable fat tails. Acrary said he'd been trained to trade with pseudo-random entries in the begining of his career. According to him, these trades broke even on the chop and paid him on the fat tails. However, after time, he decided to trade exclusively with edges as defined by his test. Why? I don't know for sure but I think a lot had to do with consistency of expectation. His story is there to take or ignore. I have to say, personally, I like it.