I want to start this post with a statement that I believe to be true. The key for success for a retail trader using the tools available to us is being able to properly identify current market conditions. Current market conditions could be, trending, ranging, retracing or whatever at varying degrees of direction and momentum.
I think we would agree that all trading systems/approaches are profitable under the right market conditions. So i propose that we work together to develop a system to properly classify market conditions. A system with which we could go on to approach the market in the appropriate profitable way.
The idea would be to:
Collectively agree on a way to categorise the market
Undertake research to accurately categorise each market - with probabilistic duration and recurrence times (i.e. understand that our market categorisation comes with lag and understand the probable duration of those conditions being maintained based on historic data)
Build an Indicator to do those categorisations with all probability values included.
If we can crack that I am sure there will be a significant uptick in the performance of the community, particularly newbies who may not have settled on an approach that works for them.
I would be interested to know your thoughts on this. Bearing in mind that this is a custom build job, not an off the shelf product which is well understood and explainable.
I think I look at it a bit differently: the attraction of forex, in a sense, with its 28 pairs (just the Majors) is that you can develop a method that suits your own style (whether thatâs âtrendingâ, ârangingâ, âretracingâ or whatever - even ânews tradingâ if you have a suicide-wish), and stick to it, looking around and waiting for a pair whose price action suits you?
Not that this necessarily detracts from your suggestion, of course.
I canât help thinking that in whatâs a beginnersâ forum, the huge majority of members are going to be looking for either trend-following or range break-out (similar?) methods, though?
Iâd say the attraction of forex trading is that you donât have to do that at all. Thereâs so much going on that you can have your style, that suits your temperament and your availability and your skills, and find a currency-pair that fits it, precisely because there are so many different ones available. So why hugely limit yourself unnecessarily?
Personally, I think that every pair needs a lot of attention. I imagine that if you want to be successful, you must use technical and fundamental analysis. So there need time to concentrate on each pair. Especially if you donât have to much time itâs better to trade with 1-3 pairs. Right?
Yes. Maybe the scope in the OP is too wide. We could limit it to the major pairs and also to trend trading and build it out from there where possible. That would be more achievable more quickly.
But then trends are usually easy to spot from the screen.
Hi Ropunzel.
I see where youâre coming from.
For how hard this business is, (especially for the new comers) one definitely needs many resources, and people, to rely upon for help. Then again, thatâs what B.P.'s is for. But, I do see that you want to hone in on the particulars. Oh, Iâve thought about this very thing myself.
But, this is the other side of the coin.
How many times have you heard of the notion that all traders need to find their own way? Meaning, each person needs to find a strategy that fits their own personality. So therefore, doesnât that mean we would all have a different system, since we all are different?
Yeah.
But, then again, how much of a difference would that really entail? There is much commonality between many different systems. It seems like many, many people choose the route of trading WITH the trend. Thatâs one. Everyone inadvertently needs to know what the market sentiment is. Thatâs two. But then, itâll start to deviate regarding which indicator to use, cause you know how many there are out there.
So, my point is, this would be a good thing to explore, but only to a point.
I think a good idea would be to start with the (momentum gaining) Strong/Weak Analysis thread. I mean, doesnât that only make sense? Their finding what works, in the short/medium term time frame. But, everyone still has to pick their own method of getting in and out at a particular point.
I donât know. These are just some of my thoughts.
I do hear what youâre saying. But, I think it can only go so far.
You know what I mean?
I donât want to use his formulation but rather develop one. But the idea would be to do something along the same lines, assess the direction and momentum strength of a market in a given horizon and to use that information the classify the market further.
Then to introduce probability values for how long the market will continue in that state etcâŚIdeally you would be able to test the performance of a trading system in the different states and would therefore be able to identify the market conditions under which the system is profitable.
I hear what you say, and in order to get any info at all about âSQNâ apparently you have to fork out your first $250 to âBuy the bookâ - which if anything like his other book " Trade your way to finacial freedom" - promises much but delivers little of any real use.
He has a course on offer for what $80,000, which says he will âTeach youâ over a period of 3 years or so to be a successfull trader, provided YOU put in a great deal of effort and pass his âexamsâ - Oh andyou have to âGet your psychlogy rightâ - else he has an excuse to chuck you off the course for âNot taking it seriouslyâ
Much of his âTeachingâ revolves around his new skill NLP - which says âThink positive and train your subconscious to be successfulâŚâ - first identified I believe by Napoleon Hill and furthered by many others since.
He then goes on to âPosition sizingâ - which I seem to remember he treated more logically than the much hackneyed âonly risk 2%â, but again gave no real information except as I remember he did say âUsing proper position sizing, it is possible to make money using Random entriesâŚâ
An interesting thought, which one skilled in programming and interested in probability could easily write a programme to backtest - I would have thought ?
The principle that we can assess the probability of a certain move, at a certain time, I cannot even begin to concieve of, since we know a couple of things:
a) Prices go up and down - Hence it is attractive to feel that we can form strategies based on âReversion to the meanâ - one can see the possibility of forming some proabalistic calculation to predict the approximate value of a return to the mean based on the distance and time from said mean. (This is an important area where Forex appears to differ from Indices)
b) SOmetimes prices trend This is an area where some seem to make good profits, but how could it be possible to calculate said probability. Trade selection seems to be a huge issue here with as many as 90% of bets being unsuccessful if you go âFull Turtleâ
@MikeWolski links to @Dennis3450 s thread, which gives valuable insights into current strength / weakness of various pairs. Probably best used as a background tool rather than a trading system, but valuable nonetheless.
That is the question which the answer to would provide us with an income for life - or possibly kill the market stone dead !
What factors could we quantify, to give an indication of this ?
Essentially you are looking to build a better mousetrap, Traders have been doing this for years, and just like people who buy those high tech mousetraps, only to go back to the original Victor spring loaded board style trap, Traders will eventually abandon the bells and whistles of complicated platforms, indicators, and robots, and come back to naked charts and pure price action trading. My Strong Weak ranking that I post daily is simply taking price action and trend, then converting that into a simple strength ranking. We are currently on a 19 trade win streak, ( a win is +100 pips from entry, most go much higher) using this simple approach, here is a link if you would like to add this to your trading arsenal
It would be pretty straight forward to do this if we model it as a stochastic process and quantify the transition between states to find the transition probabilities, duration and mean recurrence time etc⌠But thatâs jumping the gun because we havenât even decided on how to classify the market types or what time horizon to look at.
Van Tharpâs work is interesting. The main take home from the citation should be to create a template for the different variables we might look at in building a classification system. And it would probably be useful to include his work in a review of relevant literature or research on the subject before deciding how to proceed with ours.
Oh okay. Well there is a blank canvass to make the âmousetrapâ as complicated or simple as possible. My ideal would be a simple market type classification system as a backdrop for back testing multiple trading strategies. You would then know what market types those strategies are profitable in and be able to pick the appropriate strategy from the shelf depending on what the market is currently doing. (and likely to continue to do)
Yes. There is almost nothing new under the sun. a computer is a better mouse trap for a type writer or pen and pad which itself is a better mousetrap for a stone tablet.
