Trend-following CTAs performing poorly in 2017 and 2018. Switch to mean reversion?

[helpme_please]

This may be way over my league. When I explored trading strategy in 2015, I ran thousands of optimizations to narrow down to a few strategies (both trend following and mean reversion) that has worked 3 years running. I figure if it worked every year, then something must be right. Those strategies went live in 2015 with mixed result so I shut them down. I still run forward tests occassionally and I found out that 2016 and 2017 are both losers. I imagine those funds got funding because their strategies were right for some length of time. I would be confident too if my strategies worked that well for so long. But something in the market chemistry just change, so much so that even HFT is waning.

Perhaps the only thing that doesn't change is the fact that market always changes.

You mentioned you had both trend following and mean reversion strategies. Both were losers in 2016 and 2017? I would expect mean reversion to be inversely correlated with trend-following since they sound like polar opposites in strategy. I am surprised both were losers for 2 consecutive years.

 

[winnertakesall]

Yes, both trend following and mean reversion strategies got more difficult. Mean reversion would show slight positive return in backtest but not making enough to cover slippages. What changed is the nature of volatility itself.

 

[helpme_please]

Yes, both trend following and mean reversion strategies got more difficult. Mean reversion would show slight positive return in backtest but not making enough to cover slippages. What changed is the nature of volatility itself.

Surprising results. I would expect one to go zig while the other go zag. I think you're not alone in facing declining performance. There have been several reports about CTAs facing declining performance since the 2008 financial crisis. I wonder what has changed in the market.

 

[globalarbtrader]

To the excellent points already made I'd add this: you can't make these kinds of decisions based on 15 months of returns as you don't get anywhere near enough statistical evidence. It's very difficult to predict the future performance of a given risk factor like momentum, and if anything there is weak evidence that it's performance tends to mean revert (bad years for momentum are usually followed by good ones).

GAT

 

[tomorton]

All goes to show, if you want to make money don't put your money in a fund And if you're trading your own money, don't do what they do.

 

[ironchef]

It's been a much better investment for diversification vs the long tail funds or the dedicated short bias funds.

What are long tail funds?

 

[Maverick74]

What are long tail funds?

The so called "black swan funds" that are marketed to provide positive returns during a time of crisis while their long risk portfolios are getting killed.

 

[ironchef]

All goes to show, if you want to make money don't put your money in a fund And if you're trading your own money, don't do what they do.

What exactly did they do that I should not do, may I ask?

It seems counter intuitive to a non professional like me that in a raging bull market that is trending up up and up, trend following is not profitable?

 

[ironchef]

The so called "black swan funds" that are marketed to provide positive returns during a time of crisis while their long risk portfolios are getting killed.

Thanks.

How do they normally do that? By buying DOTM puts?

 

[Maverick74]

Thanks.

How do they normally do that? By buying DOTM puts?

Usually they are looking for underpriced puts or tail structures in general that are undervalued. Could be gold options, bond options, could be long treasuries/short credit, long yen calls, swiss franc calls, short the 2/10 yield curve, etc.