Stay with strategies that rarely trades?

There is a few very top strategies out there that rarely trades. Asking $150 to $250 sub fee and do 1 to 2 trades per month. Most of them just buys dip and gets out as soon as it turns green. Off course performance has been great this year, high return low draw down.

But what are you guys opinion on these type of strategy ? I feel like im not getting my money’s worth when 75% of the time its sitting in cash. But of course the gains each month is above the monthly fee, so in the end im still ahead. But its like a dilemma to cancel this strategy and reallocate the funds to a more active strategy.

Need some advise on what i should do going forward.

What do you think? Warren Buffet is one of the top 10 richest men in the world. He does hold for long. Are you looking for return or looking for trades? I can trade 100x a day, but profit is not guaranteed.

1 Like

In my opinion, It does not matter how many trades it does. If you are happy with statistics of the system on C2 and the statistics was created out of SIGNIFICANT sample size, you should try it. Issue with such systems is that it would take longer time to judge by its statistical significance

1 Like

If you don’t care, short UVXY or long XIV.

1 Like

Best situation possible is an acceptable return with minimal exposure. Cash is a position imho. The caveat being as pointed out by a previous poster is the lack of quantity in the sample size, Confidence testing usually demands very high trade counts
to get up into the 90 percentile + range. A Tos cert would help to offset my some of my concern, If it wasnt Tos i wouldnt even consider a low trade count system.

1 Like

Once you are above a total number of trades that provides you with statistical confidence in the model, I don’t really see what the problem is? I can find you many systems that over-trade, have terrible profit margins, and are one small change in market structure or behavior away from going down the drain.

Now I suspect the real origins of this discussion are with people just buying /ES dips, or just holding XIV, and I agree that over the short-term it is hard to tell if you are getting anything for your money because the market and volatility have gone in one direction.

So if the correlation to the underlying is very strong, I wouldn’t think it is a good idea to be in such a model until you see what happens with a change in market behavior.

The correlation to underlying ideally should be as close to 0 as possible.

I would think people would be looking for low correlation to underlying(s), and good risk/reward performance. Those are the hallmarks of something that actually has a statistical edge and should be robust in the future given a change in market dynamics.

1 Like

I agree with your (implied) concern and I am in the process of adjusting the strategies I follow because of this.

Everything else being equal a system that trades more frequently is better for subs. Consider this: if both systems make 7.2% / trade then you can double down after 10 trades. It is obviously to your advantage if 10 trades happen in 10 minutes than in 10 years. Again: this is for “everything being equal.”

Of course no two systems are equal, so in practice when I calculate my (totally subjective) rank of a system I weight in frequency of trades, length of trades, average p/l of trades, coefficient of variation of all this (as a measure of consistency), longevity of the system.

And this is only the start, ranking individual systems. When designing the portfolio, you need to mix in correlation of the underlyings (as others pointed out) and also correlation of trades. If you have two systems, requiring $10K margin each and one trades only on even days and the other only on odd days, then with $10K investment you can trade, in effect $20K “worth of system.”

Good luck!

Joseph

4 Likes