It is not an open source code. If you want to see it go to Marks site. It is reasonable so I bought it. Mark has a group at Yahoo that you could join to get a lot of info. I find JMA to be of limited value over T3 average or the like that are free open source code but that is just me. I do use some of his indicators and functions while day trading. I have found Mark to be helpful in every way when I had a question or issue.
Not true. Market isn't random at any timeframe. Otherwise, how would trading work at all? I've noticed: the longer time frame the more trends > noise. At intraday level: trends = noise. Trading can work well with longer time frames. In this kind of trading timeframe, however, it's more important to watch time & sales and market depth to get an idea what's going on, rather than just look at charts which is why you probably think it's "random". As far as I know trading systems don't use market depth, just prices. I don't know how can you code such a system. Such scalping method can be rather based on intuition, feeling. Time & sales is also the price action. Metooxx wrote that he uses scalping systems. I tend to say "When the charts lie, watch the order flow". What I really mean by that is that sometimes the depth/TS will tell you what's gonna happen before the charts do. Well, often I can state the opposite. Sometimes order flow lies. When I watch bid & offer sizes in relation to price action I frequently see fake signals. The "edge" you're asking for - let me try to summarize a few: -Larger cumulative profits -Higher trade frequency / consistency = no losing days -Smaller drawdowns / smoother equity curve Do you base these conclusions on hypothetical backtesting or on your actual scalping performance? Do you really have "no losing days"? BS alert flashes in my head right there! Metooxx, nitro, candletrader: do you have no losing days in scalping? Huh, let's trade on 8-tick chart, maybe we will have no losing HOURS But first of all, it isn't the edge I was asking for. I've asked about the source of an edge. In your opinion (or experience?) edge lies in using patterns in market depth. The most important factor which makes profitable scalping close to impossible are execution issues, IMO. You can use market or limit orders. With limit you can get better price but no guarantee of execution, as a result market can move against your position quickly, and you can try to "catch" it with limit orders for a long time without success. With market orders spread is your cost. And it's pretty high cost. Say you make 30 trades per day on ES. Spread is 0.25 pts x 30 = 7.5 pts ~ half of intraday range.
chasmann, how does the t3 average compare to jma? Can you post more info about t3 as google didn't find much? What other moving averages are there? thanks
And if you meant 30 round turns, that's 30 x .25 x 2 = 15pts. Factor in commissions and it starts to look pretty ugly, pretty quick. I'm not best the person to comment on this, but that looks pretty hard to beat.
For more info than you could read in a weekend on ma and T3 and many more technical aspects of market study go to Marks site. His graph comparing Jma and T3 is correct but on real time data you seldom see a spike of that multitude preceded and followed by flat line. The link just21 put up works or try http://www.jurikres.com/catalog/ms_ama.htm#top and check it out, lots of neat info. And for comparison to many other ma's http://www.jurikres.com/faq/faq_ama.htm#betterthan
alfonso, I think your method of calculating slippage (when using market orders) overestimates it. Last trade price is either at bid or ask. So, 1/2 of your trades should have zero slippage, while other half - slippage = spread. That's why my calculations assume 0.25 / round turn.
in scalping systems? how big of scalps are you going for? looks to me like this would chop a scalper to shreads
No, mechanical trading systems don't. They have a lot of downsides to them, anyway. That's why I tend to use a CPU that has several times more processing power than yours (in fact: 100,000,000*E^10,000 - larger than the number of known particles in the universe and cannot be written out fully on any medium!) - This CPU has amazing speed, abilities and runs on BrainOS '81 by GOD. I mentioned this earlier. I watch hundreds of indicators, but at the end of the day, I want my brain to be filter to my trading opportunities, not a machine. Intuition, feeling? In TRADING??? What are you talking about man? I have a fully thought-out trading system with rules (written down) that I follow, improve and update religiously. I don't need a computer to execute them. Jesse Livermore didn't, doing the same kind of scalping, ~120 years ago... Well, how smart. Actually, very often order flow lies. And this is where opportunities arise. The "fake signals" should be your friends. If not, you're on the opposite side to the pro's! I'm talking about finding these opportunities and "observing" depth and T&S, not necessarily to follow it as "confirmation". Fernando Gonzalez said (I quote!) during his chat on ET a few days ago: <i> "Since we are in a two-sided market (where participants are allowed to go LONG and SHORT any time) Momentum is established by the number of Bulls versus the number of Bears If there are more BULLS ... the market tanks straight DOWN BULLS = Selling Energy If there are more BEARS ... the market Skyrockets steeper than the Space Shuttle BEARS = Buying Energy do you see that my brain is inverted? It's the only reason why I'm alive still. So ... Momentum .... the number of Bulls vs Bears ..... is king of direction"</i> Do You know that this is probably the most well-kept secret in the Trading world? Amazing that he just said it there openly like that - But you have not the faintest idea yet how fundamentally valuable this is! More bears=bullish (selling energy) More bulls=bearish (buying energy Until you get this, you will in fact always be on the OPPOSITE side of the trade - Like all losers. You should also read the chat transcript and study it well! Once you understand this concept, watch market depth again! It now can become a lethal weapon to you! I wish you good luck. Actual scalping performance, smartie. First, you said yourself that scalping systems cannot be backtested. Second, nobody gives a flying fart about "hypothetical performance". Yes, I can have "no losing days" If I'm scalping very short-term. If I do 10-20 or more trades (a realistic daily scalping frequency on ES/NQ), then I expect to be net profitable at the end of the day. Otherwise, how good am I if I'm not net profitable after 10-20 trades??? Regarding my "edge" is not patterns in market depth. What the edge is, I clearly defined and detailed in my post. Read again. Again, You have no idea what you're talking about man. Any professional scalper knows that you don't "buy at market". What you try to do is X-M "cross the market", that is, so even you understand it, to buy at the bid and sell at the ask. With the filling fluctuations in the ES, this is indeed possible, and you're already 1 tick ahead of the game. If not, pass. If you're afraid that the market may "move against you quickly", such as in a breakdown promising high velocity, then you put in a market order and i.e. sell 1/2 of your position as soon as you're up 4-5 ticks and trail your stop initially 3-4, then 3-6 ticks behind. No noise, brother. You just can't read it. If I showed you an EEG, you wouldn't know what it means, and nothing makes sense and seems all fuzzy and random to you. Well, if I showed that EEG to a trained doctor, he would be able to tell me whole stories and their meaning out of that EEG... Trust me. It's all real, true market action, no matter what timeframe. It all starts somewhere - small, that is. ~The Scientist
Yeah, maybe. I guess it depends a lot on your style, whether you're willing to be filled or you want in/out right then and there. For someone that always pays the market then he's paying the spread on every turn. Still, however you add it up, it still looks like one tough gig.
"Can have"? Well, since your'e comments are based on "actual scalping performance, smartie", then why not just look at your results and say whether you do or not, rather than just "I can"s and "if I do"s and "I expect"s.