Diversified Trend Trading Approach

Thank you for the series, it was both informative and inspiring .

Trade well and prosper .
B.R

{ I forgot to annually revisit the strategy and undertake this process to project the performance results for the next year.}
{ By rotating systems each year to replace ‘blunt systems’ with new ones, the process lifts, as opposed to lowers, the equity curve. As we progressively compound the results, the impact of this annual ‘sharpening of the portfolio’ lifts the curve over annual intervals and we no longer see the obvious ‘recency’ acceleration in the curve as depicted in Exhibit A above. Rather we see a far stabler equity curve over its lifetime. }

I know some traders, 2 in particular who trade very short term, they revisit their strategies every weekend and optimize them to then trade the newly optimized permutations or settings the next week, on and on like an endless walk forward Analysis, keeping their strategies razor sharp and adapting rigorously to changing market regimes .

I agree that in long term Trend Following models 6-12 months is a reasonable timeframe to adjust unless one of these strategies shows a severe degradation of performance than a circuit breaker is needed and that have to be determined prior to real live trading and during the creation of the trading model .

Great stuff, thank you again .

Cheers Bass:-)

It makes sense. Short term methods require a very fast response. :-)

You can't afford to wait till they get blunt and must sharpen without falling into the trap of 'timing' I think.

Bacteria and viruses don't think and attempt to time change. They just have some mutated options that are always ready to go.

https://atstradingsolutions.com/is-t...n-or-gambling/

The Many Paths of Uncertainty

Attached Image (click to enlarge)

https://atstradingsolutions.com/the-...f-uncertainty/

Simple & brilliant explanation. This is why it makes sense to play with random price series to get a better understanding how large this effect can be and how much it can distort the true picture. Stone cold distinction between luck and skill is essential and a very humbling experience.

Spot on H :-)

As you and I know, the problem is further exacerbated by traders with a naive understanding of a 'supposed edge' who then apply leverage to these random equity curves to compound their demise:-(

Sometimes I feel the results of many of the forum's participants (perhaps even my own!) are derived from sheer luck and excessive risk.
And once that batch has gone over the cliff, a new and equally zealous batch is ready and waiting to take up the mantle.

How can I be sure I have an edge? Is there a mathematical formula?

There are no guarantees in being able to determine an edge with 100% confidence unfortunately but some methods are better than others.

Unfortunately the Expectancy Formula is not the way and only applies across a stationary market condition. In this case E = (Avge win x Pwin)- (Avge Loss x PLoss). This is a rough metric that does not take into account changing market regimes where a strategy varies in their performance but are not considered 'broken'.

The most reliable guide for me is the degree to which an equity curve from real market data departs from a random equity curve from random market data across a broad array of different market conditions.

So the way to achieve this is to randomly shuffle market data derived from a very large data sample to break the auto-correlation that resides in the 'unshuffled series'. You then test your strategy against this random shuffled data series to generate a random equity curve....and then compare this against the equity curve achieved from the 'unshuffled' series. There is a bit that needs to be done to achieve this but if you are interested here is a great book on the subject.

https://www.amazon.com/dp/B084QLXFKW?pldnSite=1

This was a great video on Better Systems Trader that provides an overview of some of the techniques you could deploy.

Let's Consider Skew


https://atstradingsolutions.com/lets-consider-skew/

Profit factor and expectancy are popular metrics, I was hoping it would be that easy.
Randomly shuffling market data scares me somewhat. I think I will watch the video.

That failing, simple folks like me will have to rely on the old fashioned method to reveal if an edge exists or not…..TIME.

Thank you for your reply.

Copernicus, what is a realistic return pa for "skilled" retail forex traders in your view?

at one point I was sure it was 100%, today I think 20 to 30% is realistic
interestingly, a survey of retailers revealed 96.4% of them disagree with me, which parallels the failure rate of forex retailers

Hi Le Metier

The way I consider it is like this. Most retail traders do not trade with an edge, but that doesn't mean that they cannot be profitable over the short term.

For example have a close look at the random series of equity curves below.




Have a look at the Normal Distribution plot at the end of the chart that plots the frequency of where these equity curves are distributed. The bulk of them plot in the centre of the distribution but as you go away from the mean of the distribution, the frequency tapers off with a bell curve distribution.

You will see that the most profitable equity curves over this 500 trade sample are represented by approximately 5% or more of the population of random equity curves. In this case a trader in the top 5% through luck alone can achieve a 100% return over 500 trades starting with an initial balance of $1000. Some traders using leverage can ramp this up considerably to beyond 100% but you need to remember that this is just the natural result from a total game of chance. Just like participating in a game of roulette.

So 5% is a pretty good result that seems to accord with many of the statistics you find 'stating the supposed success of retail traders' over perhaps a 1-2 year track record where 500 trades are likely.....but this is not success at all. This is just the natural result of luck alone.

However now look at how the random equity curves plot against that prior distribution over 10,000 trades. You will see that all of them fail over such a large trade sample size attributed to the frictional costs of trading that apply a slight negative drag to the compounded results.



So in an assessment of determining realistic expectations for a retail trader with an edge, you need to look to a validated track record of the best Professional Fund Managers in the world with a sufficient trade history of thousands of trades to validate that they possess an actual edge.

So here is a graph of the top 11 Funds of the world. You will see that Berkshire Hathaway is in the list.



Let's have a look at the annual returns of the top 10 funds of the world. Now most of these funds have experienced drawdowns approaching 40% over their track record so they cannot really be leveraged any higher than they currently are without causing critical failure.


The average annual return over a 21 year track record is 12.37%. This represents a CAGR (compound annual growth rate) of 12.04% per annum. However some years produced fairly good results of say 30%-40% per annum whereas other years even produced small negative returns. This is what happens over many different market regimes.

So for the best retail traders in the world with a sustainable track record and a demonstrated edge, you would not expect to see any of them exceed this track record of the best FM's in the world....so I would assume the result would be somewhere south of this result....but only for a very small number of retail traders that really know what they are doing.

So I would be even more conservative than your estimate over the long term.

I have been fooled by randomness and didn’t realise. I had thought the stars at FF had much better strategies than me and therefore I must abandon what I have been doing and find a better strategy. But in light of the random equity curves I am only chasing the end of the rainbow, which goes on forever. So I can stop that.

The random equity curves also explain why these people, their signal services and websites generate 404 errors within 2 to 5 years. And woe upon those who post an image of a random series of equity curves in their threads as a warning.

Regarding measuring edges, the bigger the sample the better, but if your strategy only generates 100 trades a year, can an edge still be determined?

As for returns generated by funds, is that a fair comparison given the size of their capital? Retailers are playing a much different game with far less capital. Perhaps the returns of prop traders over many years is a better measure. Would you agree?

Hanover’s fund manager perhaps is a realistic benchmark for the retailer in the long term:
28% 2016; 28% 2017; 26% 2018; 14% 2019

Hi Le Metier

Detecting an edge is a very hard exercise without having a significant trade record at your disposal. In fact I would say that it is impossible. That is why I only adopt practices in my trading strategies that are backed up by strategies adopted by FM's who can clearly demonstrate that validated long term track record.

I have come to accept over time that there is only a very slight edge made available by these very efficient markets......but you only need a slight edge when you can apply compounding to magnify that edge....but the compounded series only bears fruit over very large sample sizes. Most techniques can never survive long enough for compounding to take effect and 'lift' the equity curve.....but a few techniques that are robust can.

Survival is therefore a prerequisite for a long term trading career.....hence leverage is something that needs to be avoided as it acts as a 'two edged sword'.

You need to be extremely patient and manage risk exposure at all times with a technique that simply leaves yourself open to the occasional bounty that the market delivers from time to time. I am merciless with my losses :-)

Yep Hanover really knows his stuff and this is a very good example of a sustainable return series. :-)

Professional FM's with large AUM actually have clearer advantages to a retail trader in that they can achieve far wider diversification which makes them more robust than a retail trader. They therefore have the ability to 'survive longer' than the retail trader....and this is where you actually benefit from the compounding effect over time.

You actually find that compounding over the long term with a strategy having a slight edge does the majority of heavy lifting to net wealth. You strategy is simply a means to allow you to achieve the best compounded returns. This is referred to as the best 'geometric returns'. That is the ambition of professional FM's that know their stuff :-)

I don’t have the trade record to establish if an edge exists or not; which means I may very well be wasting my time. Strange that we happily spend years on strategies that are not statistically proven.

It would be sensible to “only adopt practices in my trading strategies that are backed up by strategies adopted by FM's who can clearly demonstrate that validated long term track record”. And therefore wise to migrate to EDTT1, but if I told my wife I am switching strategy again I am confident she would strangle me.

Thanks for your replies. They are a breath of fresh air at this forum.