In partnership with:
CME Group
Eurex

“The Most Repeatable Method of Trading Ever Invented” | Scot Billington, Covenant Capital | #26

“Trend following is the most repeatable method of trading ever invented.” – Scot Billington (Tweet)

Welcome back to the second part of our interview with Scot Billington of Covenant Capital Management.

In this episode, Scot delves into the practice of trend following and why it is a great model to follow. He also discusses tips for beginning traders and investors, and advice for those who wants to start a firm. Thank you for listening to Part 2 of our conversation with Scot Billington.

Subscribe on iTunesStitcher Radio or TuneIn

In This Episode, You’ll Learn:

  • The story of Scot’s childhood obsession with mathematically based, but unknown future outcomes.
  • The discipline required to follow a mathematical model.
  • The advantages of trend-following and why the media reports that “Trend Following is Dead” every few years.

“Your largest drawdown is always in front of you. But there will always be a winning period bigger than we’ve seen because it is over a longer period” – Scot Billington (Tweet)

  • Scot’s philosophy on position sizing.

“I would rather have a small edge with proper position sizing then vice versa” – Scot Billington (Tweet)

  • How every sector in the market is highly correlated.
  • Getting comfortable with taking risks: they exist and it is impossible to avoid them.
  • The Barbell strategy: have very little money at risk, but the money that is at risk is in the most aggressive things that it can be.
  • The biggest mistake that allocators make.

“I’m a big believer that the handling of the large moves is sort of all that matters.” – Scot Billington (Tweet)

  • How Scot conducts research for Covenant Capital.
  • Why he is based in Nashville and how he overcomes the challenge of asking investors to buy into the long time-frame.
  • Finding your niche as a boutique firm.

“As a boutique firm you are going to attract some people and some people will never be attracted to you.” – Scot Billington (Tweet)

  • Advice for managers wanting to start firms today.

Resources & Links Mentioned in this Episode:

This episode was sponsored by Saxo Bank:

Connect with Covenant Capital Management:

Visit the Website: www.CovenantCap.com

Call Covenant Capital: +1 (615) 678-6742

E-Mail Covenant Capital: info@covenantcap.com

Follow Scot Billington on Twitter.

“Any new business is like a small tree: it cannot take a tire swing yet” – Scot Billington (Tweet)

13 Comments

  1. praxeologue on 09/14/2014 at 10:22 AM

    I would like to thank both of you for a wonderful interview. I am an investor via Attain capital so am probably predisposed to like trend following and specifically Scott’s long term style but nonetheless, you do yourself a disservice Scott in saying you are no salesman. Just speak honestly and openly as you seem to do and if it doesnt attract people, well, fair enough.

    I was a bond dealer for many years and if you start changing what you really think to win business it does not leave your origjnal views, or you imtegrity, untainted.

    My only worry in this space is the high level smell test. When the best investors over multi decades like Munger make barely twenty per cent anyone offering 15 after hedge fund fees is saying they can match gross the best investors of all time. This seems an immodest claim. what are your thoughts?

    • Niels Kaastrup-Larsen on 09/14/2014 at 11:18 AM

      Hi Praxeologue,
      First of all thank you so much for your kind words, it really does help when I get feedback from the listeners as they are the reason I do this – to help them get the clarity, courage and commitment to invest with these amazing managers that I have the privilege to speak with.

      I’m not 100% sure of your question – but what I personally find about these managers and strategies is that investors don’t seem to appreciate the returns they have produced over long periods of time. Instead they often get criticized for having done it with a 25% drawdown at some point.

      People forget that Buffet and Munger have had much bigger drawdown along the way plus I’m sure they do benefit from their network of policy makers and shakers of all sorts!

      CTA’s are the most consistent managers you can find (both in long term results and volatility). I wonder what many traditional investment strategies will look like when we finally start the bear market in bonds after this 30 year plus bull market that has taken interest rates to zero.

      • praxeologue on 09/15/2014 at 9:01 AM

        I didnt word my question clearly, my apologies.

        When a hedge fund manager charging 2 and 20 says they can deliver 15pc to investors after fees, they are in effect saying they can achieve gross returns of 20pc plus.

        Given that 20pc is what the best managers of all time have delivered over multi decade periods it seems unlikely to take seriously managers promising to beat Buffett.

        I am an investor in these trend followers as I think they will do well especially after a period of underperformance for the industry and foolishly I am making subjective call that in general markets are ripe for at least vulnerable to major corrections when trend followers tend to excel. I am just doubtful that 20pc returns compound are realistic.

        • Niels Kaastrup-Larsen on 09/15/2014 at 10:13 AM

          I agree that it’s difficult to make promises about future performance – I think it’s fair to say that a strategy has been designed to make say 15%…because managers control the leverage of the strategy and thus can calibrate the strategy to a certain return/risk level. But in this business, like in most businesses, it’s hard to make promises when you are dealing with a lot of unknown factors. In fact I think it’s hard to even compare a Buffet strategy to a hedge fund as he gets his returns from a very different source. Thanks for making this point.

          • praxeologue on 09/16/2014 at 7:08 AM

            A great post on 20pc returns, or more and wha that would mean long term. http://mebfaber.com/2011/10/20/40-returns-over-20-years/



          • Scot Billington on 10/20/2014 at 3:25 AM

            This assumes that investors stay with the investment. Our Aggressive program has compounded at 18.92% since inception in 2004. We have only a single client who has been with us since then with an original fixed allocation and no subsequent withdrawals.



          • Niels Kaastrup-Larsen on 10/20/2014 at 7:22 AM

            Investors are really loosing out be not being able to invest with a long term horizon



        • Scot Billington on 09/21/2014 at 3:41 AM

          i think that is a very fair question, and one I ask every aspiring trader. I would simply say that I think trend following offers about 20%/year in edge that the best managers of this style can capture, particularly those not burdened by their own size. Just as the best value equity investors have made about 20 a year; I think the best trend followers can do the same. I do not think any or an average trend follower can do so. Particularly if you risk adjust returns to match Munger’s worst draw downs, I suspect there are multiple trend followers with 15 after fees compounded over 15+ years.

  2. Rob D on 10/04/2014 at 11:36 PM

    Thank you both for the excellent interview! I have a few questions specifically for Scott: In your discussion of position sizing you mention the optimal level (using the kelly criterion it sounds like). With an event where the exact odds of winning and payoff ratio can’t be known precisely (such as trading) you have to estimate them as accurately as possible. Using a back test would be the best starting point but will always over-estimate your edge. How do you get a handle on what type of haircut you should put on your edge when calculating the optimal position size?

    For full disclosure I run a small systematic fund and this has been an area of much research for us. The method we have settled on using ourselves is to run a 10,000 trial simulation where we vary our trading parameters randomly and then use the edge calculated from the worst trial. Any thoughts on this approach?

    • Niels Kaastrup-Larsen on 10/05/2014 at 7:00 AM

      Great question Rob….Let’s see what Scott comes back with!
      Thanks
      Niels

    • Scot Billington on 10/20/2014 at 3:22 AM

      As you know the Kelly won’t work with dynamic outcomes. We use a monte carlo type engine to best estimate return/risk. The curve losses slope throughout finally turning flat and then lower. usually, look for a spot at the beginning of the plateau. We then compare this point relative to the expected edge going forward.

      I like your method and think that the best part is that you can feel relatively confident that you are not risking too much.

      Overshooting your optimal level is obviously tragic; however, undershooting can be expensive as well. The optimal level is a critical piece to our loss management, so a large underestimate would not be useful. You might consider a walk forward in which you learn by how much your previous test has over-estimated your realized edge. Adjust your future edge by that amount, and use the plateau inception rather than peak.

      As a side note our realized edge has not been much worse than our original back-tested edge suggested. If you are seeing too much negative variance, you might be able to run better tests. If your backtests are so unreliable that you only feel confident using the worst of 10k runs, perhaps there are some more robust principles to use in the model. You might spend some time testing your test. ie “if this were 1995 and i had run my test up to now, i would have expected X. What did I realize?”

      It seems to me that the question here is, “How much can i trust my back test?” Maybe the title of Niels’ next series.

      • Niels Kaastrup-Larsen on 10/20/2014 at 7:20 AM

        I like the title Scott…back tests is to many a misunderstood tool! Thank you!

Leave a Comment