No registration required! (Why?)

Spoos just for Daytrader


lets see if you think this is ambiguous.

Click image for original size
No Description Entered


Click image for original size
No Description Entered

Apologies for not seeing this yesterday. So let's take the comments "Maybe we have to sell off into Easter and then on Tuesday the buying starts in earnest again."

So we didn't see any selling today and the longer term profile has not changed shape so I assume that your analysis continues through to tomorrow and your statement translates into: High probability for sell off on Thursday (and Monday?) and then high probability for rally on Tuesday right?
it takes time for a super tanker to halt then turn around. You say no change in the longer term profile. yes I agree but then we have not spent hardly any time for that to happen yet. I have given you a clear concise gameplan and yet your comment/statement followed by a one word question...Right?... leaves me thinking that you personally have little flexibility and maybe that is because you want an entry that you can program eg: when the macd is at...and the price is kissing the ema and bollingers are in a pacman then I enter long/short. Who cares about any of those lagging indicators when you can find out in advance the who and where and when and why "they" (the dominant trader) will be operating. I used to suffer from precision and perfection but eventually you learn that there is no such thing but MP comes a pretty darn close 2nd.

So to answer your question which was....Right?.... the answer is...Yep
I agree with your comments about MP - I too consider it the holy grail (or as close to) and trade almost only with it. We differ in that I like to get the computer to answer questions for me and give me undisputed facts about the market - using MP of course. That is why (I think) that you think that I'm at loggerheads with you - which I'm not. I'm trying to understand what you see so that I can get my software to confirm the facts. I am not looking for precise entry points either. I'm looking for probabilities - but I'm looking for a lot of them - in real time - and the answers to a lot of MP questions - again in real time - and so I need software assistance to come up with these answers.

Say for example the MP chart creates this pattern:

A
A
AB
AB
BC
BC
BC
C
C

Can you tell me what the range and relationship of the D bracket is going to be? And with what sort of probability? The software that I'm working on tells me this sort of stuff in real time. It means that I don't have to rely on a subjective memory that can be influenced by other current market events.

We are both using the same methodology but just approaching it from different angles. I think that yours is subjective and you think that mine is too rigid - that's normal that we would think of each others approaches that way because of where we're coming from. However, we can both learn from each other in this process. I can give you some hard probability facts about what the market has done in the past when presented with this situation etc.
There are no 100% definite hard facts in trading because of the many elements that can influence. However it is possible to get to 90%
Consider the elements Asia vs Europe vs US vs bonds vs commodities vs fx vs interest rates vs company earnings vs upgrades/downgrades vs event risk vs margin calls vs arbitrage vs need vs day vs overnight vs swing vs hedge vs well it is endless as a list.
A trade first put on is an anxious trade and only becomes less anxious as trade location evolves. Once trade location is established then the degree of anxiousness and need to trade is lessened until the next setuo occurs

You want a statement IF x then DO y. Well I give many examples of that ...go back over the charts. You want a 100% objective trade. Well even Pete says there is subjectiveness involved. Sujectiveness in my humble opinion is approx 10% of the trade.
I quote Pete ............."The ability to adjust comes from thinking for myself. In other words, using sound market logic rather than following formulae by rote or automatically accepting the conclusions of others.
You can’t respond mechanically and expect to succeed on a regular basis.

In other words, while you might be able to beat the standard probabilities in the short run, but the odds in favor of that happening in the long run are infinitesimal. There is a reasonable chance, for example, that a nickel tossed in the air five times might land heads up five times in a row. However, as you increase the sample size and toss the same coin 1,000 times, the probability is that the nickel will land heads up closer to 50% of the time, not 100% as it did in the smaller sample.

By developing a logical approach, it is possible to identify data parameters that work 60 times out of 100 – sometimes 90 times out of 100. The key to using these parameters effectively is thinking about the conditions in which they will succeed. You can separate the winning situations from the losers more often than not when you don’t react automatically. By way of explanation, the characteristics that classify daily market activity – Normal, Trend and Non-Trend – are defined data parameters. Once you recognize the characteristics of a Trend Day, for instance, the appropriate response is to go with the market. This strategy won’t work on normal days because normal day’s activity creates a different kind of opportunity which requires a different approach or response.

Consistent performance depends on LOGICAL DEDUCTION (my emphasis)–
About WHAT is ACTUALLY happening in front of you on the screen. That’s why the only effective tools in today’s uncertain investment climate, in my opinion, are those that DON’T treat all process and opportunities as equal or alike. To evaluate the situation developing in front of you on the screen realistically, you need experience gained in the marketplace and tools that DON’T rely on happenstance.".................


based off probability of historicals it is possible to generate what you are looking for but there are multiple elements. your risk parameter is different from the next guy which is also different for the time horizon and also different from the need to trade.
How do you adjust for size of lot to trade is another example. how do you adjust for the liqudity mirage that I refer to. In minutea it is the same as slippage. How do you react to the non fills despite being clipped at your price etc etc
What you are clearly trying to get to is using the power of volume ie many people involved vs degree of movement which in reality is determined by speed vs ? and this is the key. The ultimate key is knowing whether to be responsive and bid/offer or whether to be initiative and sell/buy. Understanding what the market is telling you. ie is it trying to go higher or lower. is it doing a good job of that (facilitating) where is it in relation to previouly known reference points. is the value higher lower etc etc.


what is the basis of your programming
How far advanced are you
Are you working with a programmer and telling him what to program or doing it yourself
How are you collating and storing the historical data
How are you extracting that data for future probability trades
What data vendor are you using
What API are you writing to

I suggest you also add to your arsenal of reading Chapter 17 that i supplied to John Carter for his book
http://www.tradingclinic.com/downloads/bio/chapter17_jc_mp.pdf
quote:
Originally posted by alleyb

..."The ability to adjust comes from thinking for myself. In other words, using sound market logic rather than following formulae by rote or automatically accepting the conclusions of others.
You can’t respond mechanically and expect to succeed on a regular basis...

You probably could successfully respond mechanically but the mechanically system would need to respond to the changing environment. I agree that a simple mechanical response would not work.

quote:
To evaluate the situation developing in front of you on the screen realistically, you need experience gained in the marketplace and tools that DON’T rely on happenstance."

Experience gained in the marketplace: How does a new trader gain that experience without spending years in front of the screen? Taking courses and reading books? What if you could short circuit this and see on the screen what the experienced trader has stored in his memory banks?

quote:
...your risk parameter is different from the next guy which is also different for the time horizon and also different from the need to trade...

That's important and that's why you would only use the probabilities from the data and not the strategy of another trader.

quote:
what is the basis of your programming... How far advanced are you... Are you working with a programmer and telling him what to program or doing it yourself... How are you collating and storing the historical data... How are you extracting that data for future probability trades... What data vendor are you using... What API are you writing to...

The software is written in C# and targets the ASP.NET 2.0 platform - I'm writing it myself. The data is stored in both Access and SQL Server and updated in real time. The stats are calculated on the fly and parameters allow you to zero in on specific dates. (i.e. Stats are not stored once calculated - not at the moment but may in the future if they are complicated and require extensive processor juice.) The project is about 20% complete but 4 traders are already using it to assist their trading decisions with real money.

As I mentioned in another post. I'll take you through it once it's on a public server.
"system would need to respond to the changing environment".
yes that is good but when I ask about different time horizons and needs etc you say "you would only use the probabilities from the data"
? to me your answer creates flexibility that is then taken straight back

Robots were created in excess of 40 years ago but still we are a long way from creating "Hal". AI has been around in financial markets for 25 years and still noone has come up with a program that beats the human. One US Investment bank when Chaos theory was the rage a few years ago spent $50 million bucks (when 50 was actually a huge sum of money) at a super cray computer to try to establish whether chaos or random walk or fractals or anything else consistently performed and they walked away with no results.

Re Experience: consider applying for any profession.Lawyer,brain surgeon accountant. There is on average a 7 year study period with qualifying exams and a further 3-7 years of application etc.....
Financial markets are in reality no different. There are no shortcuts that's why people so voraciously devour all written works by Benjamin Graham, David Dodd, Richard Wyckoff, Jessie Livermore, Richard Russell etc
Re Risk params: "you say you would only use the probabilities from the data and not the strategy of another trader".you missed the point which is about risk something that in reality only the wizards apply consistently.

There are times to trade and there are times to not trade there are times in the active market to trade a 1 lot and there are times to leverage up. There are times that there is slippage there are times you do not get fills. there are times to pyramid. there are times to consider a day trade as being a swing trade. There are times to cut the profits (and losses) and times to let the profits run. there is much more and all are conundrums in terms of creating an app. an app that is meant to take away the thinking ...an app that does the anlysis objectively..Oh and by the way what about backup, what about being locked out of the market, what about liquidity mirage, what about event risk, and all that before we have even got to diversification.
quote:
Originally posted by alleyb

..."you would only use the probabilities from the data"
? to me your answer creates flexibility that is then taken straight back

I don't understand that part...

quote:
Robots were created in excess of 40 years ago but still we are a long way from creating "Hal". AI has been around in financial markets for 25 years and still noone has come up with a program that beats the human. One US Investment bank when Chaos theory was the rage a few years ago spent $50 million bucks (when 50 was actually a huge sum of money) at a super cray computer to try to establish whether chaos or random walk or fractals or anything else consistently performed and they walked away with no results.

It doesn't matter how much money you throw at it if you're not using the right people.

quote:
Re Experience: consider applying for any profession. Lawyer, brain surgeon accountant. There is on average a 7 year study period with qualifying exams and a further 3-7 years of application etc.....

The system I referred to in the other posting was modified from a system used in the Desert Storm war. Generals could ask the system "what would Montgomery etc. do in this situation" and would in effect have the electronic minds of the greatest war strategists sitting next to them. This is the same sort of quantification we're looking for here. We want to ask the question: "What would George Soros have done now?"

quote:
...there is much more and all are conundrums in terms of creating an app. an app that is meant to take away the thinking ...an app that does the analysis objectively..Oh and by the way what about backup, what about being locked out of the market, what about liquidity mirage, what about event risk, and all that before we have even got to diversification.

The app gives you the mind of the experienced trader. As for backup etc. What do you do when your star trader leaves or is run over by the ubiquitous bus?
lets deal with History then. Bernard Montgomery just before El Alamein was aided by inside information. The code breakers had got hold of Rommel's battle plan. Consequently Monty not only knew Rommel's planned route but his supply lines. Apart from which the Allies still controlled Suez and the Mediterranean. Rommel was receiving 30% of supplies and this battle eventually was about 1000 superior Allied tanks and 200k men vs 500 aged Italian ones and 100k men. Despite the superiority in fire power (Remind you of anything in modern warfare?) It was actually a very close run thing. The Allies very nearly failed and about 13k allies were lost and 25k Germans as well. It was most certainly not a walkover and in reality was purely won on overwhelming numbers of allies and tanks. Apply that logic to the markets and it becomes a very dangerous game. (Remind you of LTCM? or perhaps one other example where the jury is out. The Bk of Japan's 800 billion dollar FX "folly")
Any attempt to fully automate the markets and operate entirely off a rule based algorithm will fall at the first hurdle when the rules get changed. For example the CME and CBOT are proposing no longer publishing CTi information. For me this is not a problem as I have a long time ago created an alternative LDB that allows me to breakdown the market activity and determine the dominant trader and their positions. Look at all previous manipulations and the exchanges always move the goalposts whether it be on contract specifications or margining. Whether it be who can play and their percentages ETC ETC.
The chart attached to this thread was posted April 13th and gave a clear guideline . Scroll to the bottom right hand side and look at the pink dotted arrows
A rule based system such as that one is still operated by humans. If I said to you that you could have Peter Steidlmayer and James Dalton (and anybody else for that matter) sit next to you and answer your questions while you're trading, would you accept my offer? I'm guessing yes.

Once their minds have been captured by a Knowledge Engineer you can interrogate the system and find out what they would do in each situation. If the markets change then more rules are added to the system.
I have sat next to Pete on several occassions over the years and I look forward to doing it again. The rule based system you propose may very well work in the short term however when it gets too big for its boots then the human will take over and crush it. You mistake the power of automation for trading whereas its power comes from mining the data warehouse. "Hal" had to be turned off. "Kit" eventually dropped into the ocean. Logan escaped etc. I repeat again your system will fall victim to the liquidity mirage for you are a user and abuser of markets as opposed to a facilitator. or another way is that you follow rather than lead. or to put it even simpler you are not a market maker and the power of a market maker is to quote what you wish to quote not what the customer (market) forces you to quote.
Nevertheless I look forward to seeing what you create.