How to trade the Unholy Grails system?

[Trembling Hand]

Look at the equity curve at post 28.

Yeah sorry cannot really see what it starts at. But starting Capital was 30.

If he blew up does he lose 30k or 90k?

That depends on technicalities but it would be 30. Of course if it was a disaster where all your holding if using all margin dropped lower than 30% overnight then you start losing more than your capital. But come guys really?

 

[craft]

To stop this going round in circles what I’m digging at is

What return is attributable to the system and what return is attributable to the leverage.

You shouldn’t confuse the two.

 

[Trembling Hand]

Actually this is a very interesting discussion and dare I say it gets to the underlying dreaded argument, one that I never want to really get involved in but here goes. TA vs FA!!

System traders use leverage, or can use leverage, with simple portfolio management. Thats why we use stops and cannot for the life of use understand hanging onto something thats the market is pushing against us.

I bet I regret this post.............

 

[Trembling Hand]

To stop this going round in circles what I’m digging at is

What return is attributable to the system and what return is attributable to the leverage.

You shouldn’t confuse the two.

Most systems NEED leverage to hold position.

 

[McLovin]

Actually this is a very interesting discussion and dare I say it gets to the underlying dreaded argument, one that I never want to really get involved in but here goes. TA vs FA!!

System traders use leverage, or can use leverage, with simple portfolio management. Thats why we use stops and cannot for the life of use understand hanging onto something thats the market is pushing against us.

I bet I regret this post.............

Would it be fair to say a system, unlevered, will return below average returns because it is designed (from a risk perspective) to be levered?

 

[CanOz]

Would it be fair to say a system, unlevered, will return below average returns because it is designed (from a risk perspective) to be levered?

No, not necessarily. In fact most system design and testing that I've done with my own systems and the flipper were unleveraged. No leverage at all was taken into account. The results are compounded, that's all.

CanOz

 

[JimBob]

Without leverage, you wouldn't be able to take up as many positions as you would with leverage so you either need a much larger starting base or your equity curve would be reduced due to taking less positions. In this case, the end point should still be achieved, just at a later date as long as your system remains valid.

 

[McLovin]

No, not necessarily. In fact most system design and testing that I've done with my own systems and the flipper were unleveraged. No leverage at all was taken into account. The results are compounded, that's all.

CanOz

Thanks Can. So if you were assessing a system you'd want to know about its unlevered returns?

 

[craft]

Actually this is a very interesting discussion and dare I say it gets to the underlying dreaded argument, one that I never want to really get involved in but here goes. TA vs FA!!

System traders use leverage, or can use leverage, with simple portfolio management. Thats why we use stops and cannot for the life of use understand hanging onto something thats the market is pushing against us.

I bet I regret this post.............

FA is also a system.

It could also be leveraged and I’m sure some do.

Portfolio management is just as relevant to FA.

I would be the first to admit that volatility is higher for many successful FA systems and this in turn would be a reason to keep gearing levels more conservative then for a smooth technical system equity curves.

But no matter how smooth the equity curve the higher you leverage the more at risk to an outlier you are, so leverage and its risk/return needs to be analysed separate to the underlying system return.

You shouldn't regret your post - I see no difference in the most important aspects between the two.

 

[Trembling Hand]

Would it be fair to say a system, unlevered, will return below average returns because it is designed (from a risk perspective) to be levered?

Well I'm sure we are now getting into the how long is a piece of string. BUT

Once you have a profitable system you soon find that you are taking a position worth $100,000 but only risking $2000 to hopefully make $6000. Considering your risk is $2000 but the position is worth $100,000 what should your capital be? Any reasonable trader would be mad to alot $100,000 to something only risking 2 g. Its just a tool to get your capital working.

Why not use leverage take 3 positions risk 6 g against $100,000 and aim for say 18 win? Thats why techies struggle with FA. I imagine a FA trader wouldn't stress over their stocks going 6% against them?

 

[Trembling Hand]

FA is also a system.

It could also be leveraged and I’m sure some do.

Portfolio management is just as relevant to FA.

I would be the first to admit that volatility is higher for many successful FA systems and this in turn would be a reason to keep gearing levels more conservative then for a smooth technical system equity curves.

But no matter how smooth the equity curve the higher you leverage the more at risk to an outlier you are, so leverage and its risk/return needs to be analysed separate to the underlying system return.

You shouldn't regret your post - I see no difference in the most important aspects between the two.

Yeah I agree. I hate when I see people saying I made 90% in 3 months. Sure you can do it. I have statements laying around this place turning $1,000 into $50,000 in 7 days. But its BS to a large degree. The turnover was something like $8 mil. I **** you not. Whats the risk adjusted returns. 0.6% for a weeks work..... ho hum.

 

[skc]

Craft,

When I used to pair trades CFDs I might be, say, holding $20k long and $20k short. The margin required to support the trade is probably $8k maximum. But with my own risk management in mind dealing with lots of pairs over a long time, the trading account size would be ~$100k.

So if I made $500 on that trade, I measured my return in two different ways:
1. Return on position size which was $500/$20k = 2.5%
2. Return on account size which was $500/$100k = 0.5%

But when I consider my annual return and CAGR etc, it's simply how much cash is in the account over how much I've put in. If I had $150k at the end, it's 50% return. It doesn't matter that at some stage during the year I held $400k long and $400k short.

In the case of tech/a return... it is neither here nor there. I would have accepted the fact that he started with $30k and was able to withdraw $360k (or whatever) at the end of 7 years and call the CAGR that way. How much face value was held during the journey is irrelevant.

Returns do matter, the whole point of investing/trading is to achieve excess returns. Everybody who visits ASF is trying to achieve excess returns (and protect principal). My bullsh!t detector starts ringing when I see anybody claiming a CAGR>15% over timeframe of 5 years or more.

You probably need to recalibrate your BS detector. The market itself did better than that at times. Even a term deposit back in the 90s could do that...

 

[craft]

Craft,

When I used to pair trades CFDs I might be, say, holding $20k long and $20k short. The margin required to support the trade is probably $8k maximum. But with my own risk management in mind dealing with lots of pairs over a long time, the trading account size would be ~$100k.

So if I made $500 on that trade, I measured my return in two different ways:
1. Return on position size which was $500/$20k = 2.5%
2. Return on account size which was $500/$100k = 0.5%

But when I consider my annual return and CAGR etc, it's simply how much cash is in the account over how much I've put in. If I had $150k at the end, it's 50% return. It doesn't matter that at some stage during the year I held $400k long and $400k short.

Is a small return on a large (undercapitalised) exposure that results in a 50% return on equity equivalent to your 50% return in a professionally approached pairs trading method?

Which is more repeatable?

Leverage has a different risk profile to system return and I want to know what’s responsible for what return before I can evaluate the risk/reward.

For pairs trading I assume net leverage is relatively small even if exposure is large. That’s not the case for a leveraged long only system.

 

[Trembling Hand]

Leverage has a different risk profile to system return and I want to know what’s responsible for what return before I can evaluate the risk/reward.

Is it a simple matter of picking some sort of risk metric? Or how do you actually evaluate the risk of a two or more vastly different system?

 

[sinner]

A couple of trend following systems leveraged 3:1. There is definitely alpha in portfolio management, you are simply not holding assets which aren't in a bull market, which is more than you can say for at least 10-20% of any given stock index. The green curve is actually techtrader run against the XSO.

(against the XSO to represent techtrader universe better)


(as you can see even in the most potent of bull runs there will be 10-20% of any index which isn't participating)


I'd also point out that the traditional use of technicals has been to reduce volatility. Leveraging low volatility is a technique which is now being recognised as the major driver in returns for fund managers like Buffett.

 

[craft]

Is it a simple matter of picking some sort of risk metric? Or how do you actually evaluate the risk of a two or more vastly different system?

I’m sure you don’t want to side track into distribution of returns and how truly random or correlated they are etc. The point I am making is that I wouldn’t even try to evaluate a system until after I had isolated its returns from any effect of leverage.

The figures Tech is flinging around clearly includes some effect from leverage but how much – can’t get a straight answer to that so can’t get to the underlying system performance. It appears to me the system equity curve ran from 90 to 360 or 21.9% CAGR which is nothing extrodinary for a trend following system in a big bull market, the rest of the result is leverage. But I’m guessing because my questions are dodged with insinuation of my lack of understanding.

As well as the bull market period I would want to know how the system performs in a choppy sideways market before I could even think of assigning safe leverage levels.

 

[Trembling Hand]

Craft yeah I agree. Its a good result not spectacular but its still a long way from what most traders manage - even in bull markets. Sad but seriously true!!

By the way they are not designed to work in sideways markets. Thats why they have triggers that turn them on and off.

 

[tech/a]

Ive made a simple point.
Pedantics aside.

Sure the headline got discussion going.
Yes there are ways to maximize return.
Which includes the use of leverage.
If anyone wants to read through the pages of info
they will see how Risk is minimized and how draw down
is also kept well in control (With tradesim you can test including
leverage--dont know about amibroker).

But an average system Techtrader turned $30k into $360k

If you had a SMSF and got on that with say 200K or more
then you'd be pretty damned happy with the results.

You only need one prolonged move in your direction with
enough in it to be life altering.

Food for thought---digest it or let it Rot--up to you.

 

[craft]

Craft yeah I agree. Its a good result not spectacular but its still a long way from what most traders manage - even in bull markets. Sad but seriously true!!

By the way they are not designed to work in sideways markets. Thats why they have triggers that turn them on and off.

But no trigger is perfect and getting whipsawed in sideways markets is the norm.

I here lots of discussion on Monte Carlo but how many chunk the actual market equity curve in varying block sizes to preserve the serial correlation?

As soon as you Monte Carlo resample using a period of one you are making an implied assumption of no correlation or full randomness and that’s’ just not how the market works. It’s quite feasible that a trend system will turn on multiple times to trends that fizz out early – and you can’t know in advance which one will be the real deal so you have to take every signal. Under these false starts its easy to get multiple losers across a correlated portfolio and have it happen two, three, four or more times before a sustained trend gets under way. The reality of what correlated instruments do is almost inconceivable under a random distribution.

A system has to be evaluated across all market conditions – cherry picking the return from a trend system in a bull market gives you a false sense of reality as does Monte Carlo simulations if you destroy a serial correlation that actually exists.

 

[Trembling Hand]

Craft, again no argument from me. Thats why I take the discretionary way.