Discussion of high-frequency trading on MT5 - page 8

 
papaklass:

HFT has no relationship to scalping or intraday. HFT is trading without risk as this tactic trades through market inefficiencies. For example, Ask below Bid (inefficiency). Stupidly and stupidly bought and immediately sold. The profit is in the pocket. But such inefficiency has a very short lifetime (10 - 100 ms). You need a very high speed to trade it, that's why they call it HFT trading.

This is where the speed of delivery of the trade order to the liquidity provider that your broker, terminal, etc. can provide comes into play.

White noise is from another line. :)

Are you saying that 10-100ms this inefficiency is there and at 100-1000ms it is not there already? Is there discreteness in this?

I'm not interested in the Martian technology on supercomputers, but in what limit of high-frequency trading is available to people like me on mt5 or mt4.

How many trades it makes per second, what algorithm it uses (in general terms), expected profit, risk, etc...

And how much to order it might cost.

 
gunia:... What type of account to open, on which symbol(s) to trade...

I have no other options for scalping except for ECN accounts. The instrument must be liquid (you can open the Level 2 market and see bids at current prices).
If it is as profitable as described in this thread. Although my FOREX teachers say otherwise. The most stable earnings are on average 1-2 trades a day, on price action, with a minimum of indicators and simple logic. But as a test it's quite possible to try out the noisemakers as well.

As for the minimum number of indicators and a simple (but working) logic - I fully support your teachers.

The most stable method of earning (trading) - any, as long as it brings profit, 1 - 2 transactions a day or 20-30-50, as some do. Your teachers, if it's no secret how much they earn (in percent of depo per month) since they say so?

 
vav990:
I have no other options for scalping except for ECN accounts. The instrument has to be liquid (you can open the Level 2 market and look at orders at current prices).

Regarding the minimum number of indicators and simple (but working) logic - I fully support your teachers.

The most stable method of earning (trading) is any, as long as it brings profit, 1-2 trades a day or 20-30-50 as some do. Your teachers, if it's no secret how much they earn (in percent of depo per month) since they say so?

When I attended one of my classes the real trader with 20 years experience, he claimed he was making 5-10% per month as a manager and that's super cool. And only very green beginners are interested in experiments with risk, anyway it is impossible to get more than 20% a month from the market stably. At some point you can get 1000%, but this is psychosis.
 
gunia:
During the courses that I attended the acting trader with 20 years of experience, the trader asserts that he earns 5-10% per month as a manager, it is relatively stable and it is super cool. And only very green beginners are interested in experiments with risk, you can't get more than 20% a month from the market anyway. They may get 1000% at a certain moment, but this is crazy.

That's what I thought at first. I thought that 10% a month was manna from heaven. Although it's true, if this 10%, for example, at least from 3 - 5 million and traded by my uncle (or robot), is stable and reliable.

But this is only a dream limit for stocks. But the futures market and especially the forex market can safely (or not safely) do much more. With a reasonable risk. The reality of this you understand only through time and practice.

That is not what we are talking about here.

 
papaklass:

MT5 : in asynchronous mode the sending speed is about 30 ms + internet channel speed, on MT4 much worse.

Omg, do you know what this figure means at all? What about the bullshit that this sent order has to reach the server and be processed there?
 

I have read your writing - theoretical reasoning, nothing more. I am no smarter than you, it just so happens that I deal with this topic closely in the practical aspect.

You are either reasoning or researching. There is no technological insight in FOREX. Even trading elementary arbitrage (Bid > Ask) is not a free-risk strategy.

HFT is full of pitfalls. From dumb "time to time" to methods of maximising your strategy from market maker banks and probabilistic models of price behaviour for the next second.

The whole technical HFT infrastructure is available for the first time to almost anyone willing to try it.

You look at the tick and Level2 history with volumes and millisecond sampling and see that here is the grail. When you try to trade, all is not so clear. There are redirects and phantom prices and much more. In general, it is interesting. You think over how to change the history to get the right picture. You wait for the cross-connection, etc. It's very fascinating. And you realize that a successful implementation will certainly not stop (even if it's the first time in your trading experience when you get a consistent profit), but you will keep moving in this direction, because you can identify a lot of inefficiencies.

 
vav990:

At first I thought the same. And I thought that 10% a month was manna from heaven. Although it is, if the 10%, for example, at least from 3 - 5 million, and traded by someone else (or a robot), stable and reliable.

But that's only a dream limit for stocks. But on the futures market and even more so on the forex market you can calmly (or not calmly) do much more. With a reasonable risk. The reality of this you understand only after time and practice.

That is not what we are talking about here.

I do not know, in a year 314% with reinvestment, and for 10 years 9270907%, that is from 10k$ for 10 years will be almost a million.
 
hrenfx:

I have read your writing - theoretical reasoning, nothing more. I am no smarter than you, it just so happens that I deal with this topic closely in the practical aspect.

You are either reasoning or researching. There is no technological insight in FOREX. Even trading elementary arbitrage (Bid > Ask) is not a free-risk strategy.

HFT is full of pitfalls. From dumb "time to time" to methods of maximising your strategy from market maker banks and probabilistic models of price behaviour for the next second.

The whole technical HFT infrastructure is available for the first time to almost anyone willing to try it.

You look at the tick and Level2 history with volumes and millisecond sampling and see that here is the grail. When you try to trade, all is not so clear. There are redirects and phantom prices and much more. In general, it is interesting. You think over how to change the history to get the right picture. You wait for the cross-connection, etc. It's very fascinating. And you realize that a successful implementation will certainly not stop (even if it's the first time in your trading experience when you get a stable profit), but you will keep moving in this direction, because you can identify a lot of inefficiencies.

What is the minimum amount of money you would be interested in introducing a person to this infrastructure?
 
gunia:
For what minimum amount would you be interested in initiating a person into this infrastructure?

I mean, how much would I be willing to pay to carefully read the information in the link I gave in THIS thread?

 
hrenfx:

I mean, how much would I be willing to pay to carefully read the information on the link I gave in THIS thread?

I didn't find the subject information in this thread not in the links.

I am offering you money for the FX-HFT robot and its integration and asking about the price threshold.

Specific, substantive, concise information. Paid, received, applied, took out the profit.

It's not that I'm lazy to read the memoirs, but it's often the case that it's of little use, so I'm asking about hashing out lost time for money.

Reason: