I was just re-listening to Mike Covel’s interview with Marty Bergin but now in Youtube format……*well looky looky here…..very flash Mr Covel*…….. as having a second take sometimes allows you to flesh out some more of the meat off the bone……and came across the following statements which for the untrained eye need a second take.
Mike C to Marty: “I wonder whether there has been a bit too much emphasis placed on volatility associated with Trend Following”.
Marty to Mike C: “It is what it is. To actually do trend following, you have to absorb losses. You have to ride out the bad times to be profitable over the long term. We don’t focus on volatility…but rather settle on what is an acceptable risk to absorb to make money?”
Let’s break that little nut down. For most casual observers, they see risk as being represented by the ups and downs of an equity curve over a period of time. They like upward sloping straight lines in other words. Of course that upside vol is favorable risk and none of us are going to complain about that so let’s just eliminate that part….so then we are left with all the unfavourable down bits of the equity curve. There are lot’s of little down bits with trend following as we are always cutting losses short and perhaps over the course of a very long time, those down bits accumulate into a bit of an eyesore we call the maximum drawdown.
Well that maximum drawdown is not just a Trend Following symptom. It is present in every equity curve of every investment even the fictitious risk free asset which we approximate to the T-bill. That of course is taken to mean that all civilisations end, even the very civilized world of the Fed……but you just might not be around to see it. So in our assessment of risk, we need to consider our interval of time in which we make that assessment.
Now this is where we need to understand the notion of what another trend follower Bill Dreiss of Dreiss Research calls “strategies that release risk versus strategies that warehouse risk”. What do we mean by that? Is it really important? Perhaps this picture tells a better story than what I could ever muster in text.
You see, when it comes to predictive models in this market (which are a different beast to price following trend models), when the predictions are correct you get that nice straight line. But there is a price to be paid in predictive certainty which you try to ignore called intrinsic risk, namely that to convince the crowd of believers in your predictive amazingness…., when market conditions aren’t quite as favorable to your predictive certainty you turn off the risk release valve…..’cause your amazingness is always going to be right in the long term.
So you get a straight line of realised equity with a different unrealized story but you start to believe in the fiction…the ponzi scheme of ponzi schemes….because you actually start to believe your own cleverness …but in that straight line of ascent, lies the boiling steam that is about to put spaghetti all over the kitchen. You may have heard that story before….RIP LTCM.
So what really is the volatility of predictive models. You just gotta go out a bit in time to see them….but that inevitably is too late as they just ain’t around long enough to laugh at this insiders joke.
Trend following folk are just more transparent than the ‘market wizards of predictive certainty’. They reveal all in their equity curve. They simply don’t just warehouse risk. They release risk continuously after each trade event. Don’t sit behind a trend follower especially after curry night, but they are safer in the kitchen than those other ‘more clever folk’.
Mike C to Marty: “You mention that you specifically target a 20% max drawdown *to deliver peak performance to your clients for acceptable risk over the long run* and you have hit that target consistently 4 times. How do you know this in advance?”.
Marty to Mike C: “It’s just numbers. If you look at the data and you design the system and have the capability of knowing that history provides you with all the data you need but is never exactly like the past, then you can always work to a target….but you just don’t know that you will get it exactly…..but in our history of 340 months we have only had 4 losses greater than 20%…..about 1.2% of the time so we are pretty good at targeting….but in today’s world, not so amazing and we should expect this from everybody “
Now this is where it gets really interesting because Marty then starts talking about how Dunn Capital adapts to changing market conditions. Trend followers aren’t those single recipe guys….as their recipes are very diverse and continuously evolving and adapting but not in terms of adapting in a predictive way, but rather adapting their risk management models. They do this through portfolio management.
Adaption is applied at the ‘global’ portfolio management level. As time progresses, new trend following models that navigate more recent market regimes are continuously added to the portfolio. The risks of the past are continuously updated in the portfolio to plug possible weaknesses in future uncertainty and the dials are adjusted at the global portfolio level on a continuous basis to operate within desired adverse volatility targets yet still offer uncompromised superlative returns.
Now if there is any form of ‘black box secret sauce’ in trend following, it is not in the trend following models themselves as these are all very simple solutions that all can apply……but the secret sauce to the CTA’s that stand the test of time….is in their particular form of managing risk in the context of global portfolio management. Things go a bit quiet here as they do just gloss over this nugget….but if I was an interrogator…..and had Marty Bergin under the spotlight in manacles……this is what I would like to focus on.
You just don’t hear much about the trend followers approach to portfolio management…..and I am hoping that in future podcasts on the subject…that some serious time is spent cracking open those nuggets.
Here is Mike Covels new format on Youtube. Finally I even like his intro music. Crumbs…that was always an ear tester on his podcast channel that used to frighten the kiddies.
PS Nice curve Marty. I love the trajectory post 2008. Can I have some binoculars as I need to compare this against the S&P500 TR Index ?
Trade well and prosper