Looking for a scalping strategy...

[NetTecture]

...out of pure interest. Must not even work profitable.

Basically I do system development etc., but I simply do not really understand how scalping should work from a strategy point of view.

For me, extreme short term prices so far are governed by chaos (or: noise, if you prefer that).

Anyone knows of a published scalping strategy (like in a blog or something) that could give me a hint how scalpers attack the market? A book would also be on my list.

Again, not for trading - something about concepts would be fine, too. I just dont understand conceptually how to make money from... noise.

 

[pivosh]

hi, if you are listening something...if its noisy you will listen it better. Thats the basic concept.
I use binary options above DAX 30 or Footsie 100 and using chart and level 2 index futures...you can understand the next movements. You just read the book and the "book" never fails.

SEE u

 

[NetTecture]

Not exactly helpfull. Seriously As a first post.... I am tending to think you set up a spam (and want some othe posts first).

Anyone any strategy?

I was doing some simulated scaps the last days with puretick and I think I get the idea. Basicall trades in the range of let's say 15 minutes max (as a rule - not always), and trading fluctuations..... Hm.... Interesting.

 

[Trendy Trader]

I often watch and find opportunities in 1 minute charts. I know most people say it's just noise and it's best to trade from longer terms, and I want to do that, I really do, but I keep finding myself working with the 1 minute charts. I kinda have a feel for what's happening and I feel most comfortable with the shorter term, maybe it's my personality, maybe I'm just impatient, I know I probably have to work on that.

I mainly use support and resistance, trend lines and moving averages.

Anyway, here are a few 1 minute charts that I drew some lines on the other day, so you can see the way I work with them.

 

[dcraig1]

For a start, you really need to not think about short term price movement as "noise". It is the interplay of supply and demand in the auction market that determines price. Orders are not placed because traders are randomly tapping on keyboards.

It may help to think about supply and demand in futures markets as having four components - supply of longs (contracts visible in the book on the ask), supply of shorts (contracts on the bid in the book) demand for longs (represented by actual trades at the ask) and demand for shorts (trades at the bid).

You can divide market participants into two broad groups - the more passive traders who place limit orders and wait for their price to be hit and the more agressive traders who place marketable orders because they want instantly in to a position.

My own opinion (which I cannot substantiate) is that the best scalpers have a very good understanding of this market micro structure.

Use a scientific approach - find a way of visualising this process. Perhaps chart it. Look at aspects of the book and the time and sales.

Consider these in the broader context - are we in a trend or a choppy side ways market and so on. Over what time frame ? and so on. Where is S/R ?

You also need to know the specifics of your market(s). Not all futures markets are the same, and there is no good reason to assume they should be. Choose a market that is cheap to trade. By cheap I mean commish is low compared to potential move on each trade. eg DAX is much cheaper than ES.

Forget about full automation. It is really hard. But you might be able to get partial automation which might be very helpful as quick reactions are obviously essential.

http://www.trade2win.com/boards/us-stocks/10573-evening-trading-us-shares-124.html#post419199

 

[NetTecture]

Thanks.

Actually I am not sure I am interested in scalping - I really simply did not really get the idea what scalping is practically. Your two posts helped

For the automatism, I am working on that. I am slowly giving up on our friendly Ninja Trader, and starting to work on my own platform. Nothing I see so far i what I want.

One idea I have is to get away from "orders" towards "trades". The main problem I have with "orders" is that I can not play multiple strategies in a market with them - a "automatic long term long" position gets killed when I say "close all". If I can divide the orders and positions into trades... I can then kill a trade without affecting other posible trades. THis is great for some automatic, some manual parts. For full manual it wont make a difference, as one basically trades in a "default manual" trade. One good thing with this, though, is that I plan to have a semi-automated mechanism where a "strategy" can not EXECUTE, but PREPARE trades. The complete trade (entry, exit, targets) is set up and controlled by the strategy, but it opens in a window for me to say "ok, hit it" or "go away". Plus, at any poins, I can "dispatch" the strategy and take over manual control.

That said, I am kind of a good computer guy. I believe full automatism MUST be achievable. Maybe not as good as a human brain can, but good enough to make money

That said, thanks a lot Very nice information so far I am slowly getting a hack of some approaches.

 

[dcraig1]

That said, I am kind of a good computer guy. I believe full automatism MUST be achievable.

It's not the software side of automation that is really hard. A capable developer should be able to manage that, given time and determination. It's developing a strategy that will work day after day in a variety of market conditions that is the real challenge.

By partial automation, I mean something like automating entry or exit (or both), but being able to turn it on/off on the fly depending on whether in your judgement market conditions are suitable at the time.

 

[NetTecture]

That is correct. But then, whatever rules a trader has (and gut feeling imho is not a good one) should be at least semi-automatable.

"I suspend the strategy for now" may also be ok. But as long as trades are done for specific rules... automatism just clears a traders head My 2 cents. PLatforms suck, though