What is the maximum portfolio heat you take on?

[qwertyuiop1]

Im asking myself what is the maximum portfolio heat should I take on at any one time ?

Surprisingly there is very little guidance to this issue when you google it.

Does anyone have advice as to what they use?

I have just started trading live my backtested system which shows a nice positive expectancy.

I think I'm going to go with 13.5% (9 positions at 1.5%) - but I would be curious as to how this compares to others.

 

[tomorton]

Its hard to be categoric and still sensible about total risk, it depends greatly on what your positions are in, e.g. 9 long FTSE100 trades is basically one big 13.5% position. But if 4 are long and 5 are short, one half acts as a hedge to the other half.

Don't forget a lot of the "golden rules" for traders were written by US authors for a long-only US readership during a bull market, so question sources and details.

Many people say just run one position at a time, or very very few, otherwise, position management workload and stress can cause paralysis by analysis or panicky reactions to what might be just price "noise".

Do you want to give more detail about what you do / are doing?

 

[qwertyuiop1]

Hi. Thanks for response.

All my individual positions are completely uncorrelated. (Well....at least as uncorrrkated as they can be). So that aspect is not a concern.
This system does not have the frequency of tradesvyo run just one position at a time. It has an expectancy of 0.3R. (This is just one of a few systems I run)

I know when to add to positions and manage my risk etc.

The only thing I'm unsure of is how much maximum risk to take on at any one time. Hence this post.

How does 13.5% sound?
Is there a rule of thumb for total portfolio heat out there like there is a rule of thumb of 2% for an individual trade?

 

[tomorton]

I seem to recall a rule of thumb for 5% total portfolio risk, with each trade having an individual risk of max 2%. But the degree of correlation is key. To a point, all financial trades are correlated since they're all in the financial market. Also, all commodity trades are correlated since settlement currency is always USD. And all oversees equity investments are correlated as your account will be in GBP. Its hard to think of 9 trades that are either uncorrelated or which cancel each other out in the event of a black swan event. And 9 open positions incurs 9 sets of costs and you have to do 9 times the management oversight than on just one: so risk of loss is increased even if your trades are well founded.

I'm still not going to say 13.5% is too high. Assuming you can sleep at night.

PS: If one of your trades has achieved break-even and you have set a stop at b/e, its individual risk is reduced to 0%, so discount that from your total risk assessment.

So the thing is, at some point you must have had just the first trade open, then 2, then 3, then 4 etc. Why do you keep on obeying your system when it tells you to keep adding trades when the predecessors it recommended aren't achieving b/e?

 

[qwertyuiop1]

Ok. So to clarify, I opened all 9 trades together when my system went live this week.

And yes - when one trade eventually gets to breakeven my overall risk is reduced. But I use this opportunity to add a position. Essentially I try to keep my overall portfolio heat at max 13.5% at any one time.

While I agree everything is correlated to a degree, this system however trades every type of instrument imaginable bothlong and short. I.e. Currencies, commodities (diversified across metals, softs and energies), bonds, ETFs, indices as well as individual stocks among others. So the basket really is well diversified.

As far as risk goes with a basket of instruments, the way I see it, there are 2 risks to consider.
1)The individual trade. This risk size is concerned with limiting drawdowns.
2) the oversll heat. This risk is concerned with black swan events that wipe out your entire open portfolio. Other ways to reduce risk on a black swan event is using guaranteed stops as well as good diversification. I use both.

(My worst drawdown on my backtests was 6R (Risk per trade) - or 9% in my case as R for me is 1.5%.

In short - both considerations above are there for different types of risk.

So If you think about it, overall heat is only a factor in case a blank swan event occurs whereby everything gets stopped out. Since all mine are very well diversified both long and short, a market crash will not result in this (I also use guaranteed stops by the way to avoid habit slippage)

So the black swan event I am concerned with really is if something particularly weird happens and everything goes the wrong way all at once by just sheer bad luck. I have to assess the likelihood of that happening and take on my portfolio heat accordingly.

So on that basis 13.5% doesn't seem too much to me. Given the diversification as well as guaranteed stops I use it seems nigh in impossible they will all turn against me at the same time.

But that's just my view. The reason I'm here is to get the view points of others

I haven't heard the rule of thumb of nax 5% heat. Interesting to hear that. It does seem very small though does it not?

 

[tomorton]

Well it all sounds OK qwerty though I would definitely have scaled into a new-starting system. After that, if the system generates 10 buy signals on the same day, how can you be so confident there is no correlation? Appreciate if the system is looking at say 1,000 instruments, then trading 9 of these on any day, especially the first go-live day, doesn't seem an excessive sample, and yes they might all be totally uncorrelated. But none of this makes me feel easy - why do you need a system that views so many markets plus gets you into so many? What do you gain from that (when two hedged positions will make your portfolio diverse)? (you must know there are many people who just buy and sell the EUR/USD and haven't any need to get into pork bellies or Apple or oil).

 

[Pat494]

I understand your point about diversification but the complexity to follow and make decisions on must be take a lot of hours. May I suggest looking at forex. They are world's currencies, so unless the whole planet goes into freefall you should be pretty safe.
I am risk averse sort of person and 5% seems OK to me. If you are desperate to generate lots of income be careful not to let this consideration push you too high imho.

 

[qwertyuiop1]

Tomorton - in answer to your query, I guess I cannot be sure if there is no correlation.
But if I have a soft commodity going long and an energy commodity short and say German bonds as another product along with some exotic currency pair for example then it's reasonable to assume the correlation is pretty limited.

Also - I'm not sure I see the point in loading up the 9 positions gradually one by one. Any potential drawdown isnt reduced by doing so. All it means is the drawdown is exoerienced over a longer period of time. If you're prepared to eventually load up to 9 positions anyway then you may as well do it all in one go.

And yes - I am aware that people only trade single instruments like eurusd. I also do that with another strategy. I trade 3 other strategies in fact. I'm looking for a completely new strategy that isn't correlated to the results of the other strategies. Again, for diversification purposes. Hence I trade this new one which trades multiple instruments over a timeframe I don't already trade.