Would you trade in a group?

andrewstevens

Junior member
Messages
15
Likes
0
One way to look at financial markets is in terms of voting, more buyers, market goes up, more sellers market goes down.

Do you think by asking a large number of traders if they would buy or sell an asset at its current price would be a way to approximate global market sentiment?

What if you asked a large group of traders to which price level do you think an asset will reach next, wouldn't this information give you a better view of market sentiment and provide you with a statistical edge (law of large numbers).

I'm a computer scientist and part time trader and am interested in your feedback on a beta service structured like a managed fund setup to let traders trade together.

http://hivetrading.com/HiveTrading/topbusiness/trading.php

This is not intended as advertising I'm interested in your feedback, please let me know your thoughts on this trading service and what problems/benefits you can see?
 
99% of retail traders fail, just fade them all


How many times do you hear this?

It's no wonder there's such a high failure rate - the guidance is so poor.

The new trader quite rightly starts with price indicators and backtesting, but most of the guidance out there can now be seen to be off the mark, principally because it was devised in the days when computers were the size of rooms and only a few had access to them. This is no criticism of the then authors, as they had limited resources compared to today, and if they had the opportunity again today they would be shouting from the hilltops to update the knowledge.

Google INDE benchmark if you need to find a steady ship. Still in its formative stages but looks like it's got legs.

It uses 50 years of SP500 data to set a benchmark, which essentially will keep you afloat while your learning. Don't even bother trading a market or using an indicator unless it beats the benchmark, or you're in trouble - and you'll become one of the 99%.
 
One way to look at financial markets is in terms of voting, more buyers, market goes up, more sellers market goes down.

Do you think by asking a large number of traders if they would buy or sell an asset at its current price would be a way to approximate global market sentiment?

What if you asked a large group of traders to which price level do you think an asset will reach next, wouldn't this information give you a better view of market sentiment and provide you with a statistical edge (law of large numbers).

I'm a computer scientist and part time trader and am interested in your feedback on a beta service structured like a managed fund setup to let traders trade together.

http://hivetrading.com/HiveTrading/topbusiness/trading.php

This is not intended as advertising I'm interested in your feedback, please let me know your thoughts on this trading service and what problems/benefits you can see?


It is much better to follow the market, than to predict.

If you are computer scientist then you should have the calculating power to reduce the lag of a price indicator down to a minimum.

If you choose predict, prophesying over price movement, you will eventually come a cropper, even if this is coded as an algorithm, because you can only base your prophecy on that which has gone before. If you read up on Hurst cyclical analysis it will give you a clue as to why predictive actions are ultimately fatal - I'm not recommending Hurst's analysis as he also sought to predict - but his dissection of fractal movement into sine waves highlights that there are simply some deep wavelengths which are beyond detection because not enough time has passed to know them.

I think I have somewhere a basic wave generator in excel which will help explain - oh, here it is.

Adjust the figure in cell N3 and different fractal patterns will result. This model contains 9 wavelengths, but you just know that there are some bigger ones waiting in the wings - just like Jaws the movie.
 

Attachments

  • Wave Generator v1.0.xls
    1 MB · Views: 246
I would say that it will work! Will it be able to back test?

You can backtest variations of grid and martingale where support and resistance have been added by human (so stop when the series hits support or resistance). There are plenty of automated strategies that you would never use as they would eventually destroy your account. But it might be possible to consider variations of these strategies if you have human input controlling them 24/7. Professional trading houses do this already, a lot of there strategies are not completely automated they require human input to look for certain market conditions.

I want to do this on a larger scale, take a known strategy -> modify it and apply a large network of human input, to provide the best of both worlds.
 
It is much better to follow the market, than to predict.

If you are computer scientist then you should have the calculating power to reduce the lag of a price indicator down to a minimum.

If you choose predict, prophesying over price movement, you will eventually come a cropper, even if this is coded as an algorithm, because you can only base your prophecy on that which has gone before. If you read up on Hurst cyclical analysis it will give you a clue as to why predictive actions are ultimately fatal - I'm not recommending Hurst's analysis as he also sought to predict - but his dissection of fractal movement into sine waves highlights that there are simply some deep wavelengths which are beyond detection because not enough time has passed to know them.

I think I have somewhere a basic wave generator in excel which will help explain - oh, here it is.

Adjust the figure in cell N3 and different fractal patterns will result. This model contains 9 wavelengths, but you just know that there are some bigger ones waiting in the wings - just like Jaws the movie.

I certainly don't believe just throwing a lot of uniformed traders at the problem will necessarily improve prediction. However they can provide other useful information such as support and resistance levels which the computer can use to trade range bound strategies and so on. I've backtested many strategies where human input regarding possible (distribution) of support and resistance levels improved their performance. Prediction not as important as a distribution which might describe possible market movements. This is what is useful in automated strategies trying to maximize reward to risk.
 
If you follow there is no risk - the market is the truest expression of beliefs you can find. In consulting with a group you are actually reducing the group size down.

You also have to ask, if you are able to find a system which works as an individual then why consult with others. The market is the true line, not the opinion of a relatively select few.

Going back to the wavelength model, if you are bound in a space/time framework, as humans are, you can only base your opinion on that which has gone before. No matter how 'successful' a trader or investor has been in the past is no indication of how successful they will be in the future - unless of course they are larger than nature itself. Trust in yourself - be a shepherd to yourself. Why seek out other shepherds when you no doubt have the capability yourself.
 
It is much better to follow the market, than to predict.

Small sentence, big meaning, why make life difficult for yourself by guessing what might happen at a certain point!. Following and trading PA momentum will ultimately get you on the right side of trades more than the wrong side of trades. :whistling
 
This is basically what spread betting companies do .. you have a large number of traders making decisions - and 1 spread betting company as the counter-party- a tried and tested model - the wisdom of crowds doesn't work in leveraged financial markets.
 
This is basically what spread betting companies do .. you have a large number of traders making decisions - and 1 spread betting company as the counter-party- a tried and tested model - the wisdom of crowds doesn't work in leveraged financial markets.

Agreed, best to fly under the radar so to speak.
 
By the way, I understand that the INDE benchmark is also planning to compare the spread betting markets with the main markets through use of its comparables, just so you can see how well the spread betting prices are 'shadowing' the root markets - basically, to highlight which ones are closest to the mark and not flying all over the place like the looney tunes tasmanian devil - what's up doc?!
 
Well actually all the trades are with potential risks so this job is like that you can never know :/
 
Top