4 Criteria and examples of useful strategies

My goal is to create an adaptive, diversified portfolio of asset classes, factors and strategies that yields high risk-adjusted returns. 
I think of it as building a personal hedge fund to manage my own money.

 

There is no need to reinvent the world. A lot of original thought actually “only” recombines existing knowledge. I don´t mind using inputs and backtested results from different sources (preferably papers by practitioners with an academic background) and check these independently when I have developed my own detailed strategy rules. In fact, drawing from outside sources of other people´s previous experiences, gives my own ideas a solid foundation and often a healthy reality check.
Before implementing strategy rules with real money, I often backtest by hand going trade by trade. Although this low tech approach is cumbersome and it is hard to be exact, it has some distinct advantages: you can visualize the strategy unfolding and begin to get a feeling for the emotional rollercoaster of drawdowns and profits it brings with it. It boils down a selection of possible strategies to the most promising ones, as only those will be worthwhile testing like that. It keeps curve fitting in check, as it is not possible to run through countless variations of a strategy. And it is possible to do right away without having to clear high technological hurdles first.

 

Is there a simple test to decide which strategies deserve closer scrutiny and which approaches to rule out altogether (even if there are influential market participants and commentators in support of them)? That would be a powerful method to cut through the noise.

 

The base rate counts.*
The base rate is the prior probability distribution of outcomes. For me this has been the most eye opening heuristic in recent years. It´s a higher level test that serves as an Occam´s razor** for an investment´s reliability.
This is how you can use it in practice: all strategies have to pass the basic test of having a basic probability distribution that is strongly in your favour in their simplest form. A positive edge.
These strategies always tend to produce superior risk-adjusted returns in the long run, even when using only the most general rules and parameters or none at all. They are investment approaches that really work.
Collect information on the long term historical base rate (basic probability) of the risk-adjusted returns of different approaches: asset classes, factors, strategies etc.. Radically narrow your investment universe to strategies that pass this test. Those are the opportunities with a natural edge – you can select the ones with the best historical results and rule out the ones that do less well.
To be able to do this, it is necessary to find resources**** that provide systematic rules and the resulting historical statistics or to generate the statistics yourself.
A discretionary approach won´t be able to generate these guidelines before we actually implement the strategy.

 

Why go with the base rate?
There are enough approaches out there, that will give you a positive long-term expectation through historically persistent risk premia and strategies, that withstood the test of time and that have a compelling explanation. There simply is no need to fight an uphill battle against basic probabilities in search for an edge.
Going against the base rate makes it much harder for a strategy to succeed. Your skill must be consistently above average (and that includes all professional managers out there) and even then your portfolio will be less robust, as mean reversion always kicks in sooner or later. Chances are, that your portfolio will underperform or lose money. For me this has become the most powerful filter, it has completely transformed my investing approach and success. I don´t aim to generate alpha, but instead concentrate on combining many diverse sources of beta.

 

Interestingly enough there are many commonly touted investment approaches that do not pass this test.
A simple, but eye-opening, example is selecting individual stocks. About 65% of stocks underperform their index. The best performing stocks in a broad index perform much better than the other stocks in the index. That means average index returns depend heavily on a relatively small set of winners. No wonder, that studies tell us that almost no active fund manager consistently outperforms his benchmark over time.
Which implies that, using low cost index ETF to capture the market´s equity risk premium will be superior (with a probability of about 65%) to trying to pick the best stocks. Therefore, that´s what I concentrate on.
With very specific expertise and skill value can be found in individual equities. But it would make most sense to me, to concentrate on the most inefficient areas of the market (e.g. micro cap, deep value or emerging markets). Most deviations from a market cap index, that have shown long term outperformance, can be explained by different factors (e.g. value, momentum etc.) and can be accessed through a wide range of securities targeted by a specific ETF.


This particular screen has the valuable advantage of filtering out a lot of the confusing media noise and to bypass an incredible amount of intricate information about individual companies – it allows us to concentrate on more useful ideas.

 

A more obvious example is a whole universe of very short term (day) trading strategies, luring with the promise of instant wealth from a small capital base. Many people (including me) find that idea attractive and feel drawn to it, even though almost everyone knows intuitively that it is unrealistic: something that looks to be too good to be true probably is.
Different sources put the number of day traders that consistently lose money between 80% and 95%! Why would you try to go against such odds? Could you be falling prey to overconfidence bias or a grandiose marketing scam? How much more sophisticated than the average player are you realistically? Picture hedge fund quants with Ph.D.´s on the other side of this zero-sum game. But even if you are in the top 25% of the most skillful market participants, you are still a long way away from profitability given the basic odds. You would have to be in the top 5% to 20% and even then you would just start to be profitable – a far way from beating the market average.
Then consider the issue of regression to the mean and the power of arbitrage: how sustainable is your edge over time even if you experience profitable periods? How fast will the edge erode, because of other investors discovering a profitable opportunity and quickly arbitraging it away? Which part of the performance is due to luck, which part to actual skill?
As I can´t answer any of these questions with a convincing “yes” for myself, any short term strategies I look at have to derive their edge from a basic risk premium, that can be shown to have existed for decades and is likely to continue to do so. I want to see quantifiable rules and long term strategy statistics from different sources and of course those are hardest to find in the short term trading space – that in itself is a red flag.

 

What makes these thoughts interesting, is to turn them around – akin to Charlie Munger´s famous inverse thinking. How is such a consistent loss rate or underperformance possible (that seems very inefficient)? And most importantly: where does the money flow to?  After all the sum of the game is zero (after commissions, spreads and taxes alas). Because we want to be where the money goes.
A consistently bad strategy or consistently losing asset is very useful, if it is possible to find ways to take the opposite side.

 

Where can an edge be found? 
It is not that hard – being able to beat the average retail investor would be quite a good edge, wouldn’t´t it? This would already do it (as will become clear a bit later):
A simple example of a favorable base rate is buying and holding broad equitiy indices: the expected return above the risk free rate is the equity risk premium and that has been positive on average over long time horizons: around 5% real return per year with a long term Sharpe ratio*** of 0,3-0,4.

 

Here is an overview of long term positive risk premia that can potentially be harvested.**** The data is from different resources and these are rough numbers only – which is good, because the past can only provide a rough idea of what the future may bring anyway:

 

  • Risk free rate – the baseline against which to measure returns
  • Asset Classes (stocks, bonds, alternative assets): Sharpe ratio: 0,2-0,6
  • Factors (value, momentum, size, illiquidity etc.): Sharpe ratio: 0,5-0,8
  • Strategies (trend following, carry, volatility selling etc.): Sharpe ratio: 0,4-1

 

 

But where are these higher factor and strategy risk-adjusted returns, compared to asset class returns, coming from?
Outperformance comes from capturing money flows created by consistently underperforming strategies. For us to be able to make bets with market beating expectations, someone else needs to make consistent bets with an expectation below average.
Logically an outperforming trading strategy should target other trader´s mistakes. Another possibility is to find areas where other market participants lose money willingly, because they have different incentives (e.g. hedging a portfolio) – that makes for a very persistent edge as players don´t aim to improve to avoid losing.

 

Her are some examples of strategies that exploit consistent underperformance by other market participants:
  • Performance chasing leads investors to increased buying at market tops, combined with panic selling at bottoms. This leads to a behavioral performance gap for individual investors of more than 4% underperformance compared to the index according to the Dalbar study***** (Which is why simply holding an equity index would beat the average retail investor). That means an over-performance of nearly 4% (that is 50% higher than the historical average for equities!) is theoretically possible for investors who do the opposite. That money flows mostly into value strategies that have a Sharpe ratio, that is more than 50% higher than the equity index Sharpe ratio.
  • Loss aversion leads investors to the losing strategy of cutting winners and letting losers ride: money flows into trend following (based on the principle of cutting losses and letting winners ride) and momentum strategies.
  • Short term trading creates consistent losses: money flows into long-term strategies (patience is a strong edge in todays market) or strategies that can define those losses and take the opposite side.
  • Investors systematically overpay for both insurance (hedges) and lottery tickets. This can be utilized by providing tail risk insurance by selling index options, and selling far-out-of-the-money call options as lottery tickets. Commodity trend following takes the opposite side of commercial hedgers.
  • Forced selling (margin calls) and buying (fund inflows) identify very good opportunities especially at the extremes of crisis and bubbles.
  • Many institutional investors must follow certain rules. For example they often are obliged to sell corporate bonds that are downgraded from investment grade (BBB to BB); these subsequently often perform well.
  • Professional investors avoiding career risk, provide opportunities in unconventional strategies, unfashionable ideas and over different time horizons, if investors are willing to tolerate intermediate, prolonged underperformance at different times than the market.
  • Multinational corporations hedge foreign currency exposure. This gives an edge for FX carry strategies.

 

From all these possibilities (and there are many more that I research and collect in a spreadsheet), I continue with the task of selecting the most suitable and building a portfolio that fits my goal and preferences.

 


 

*Great ideas on such useful concepts and on higher level thinking in general can be found in „Think Twice“ and More than you Know“ by Michael Mauboussin.

 

** Occam´s razor: “Among competing hypotheses, the one with the fewest assumptions should be selected“.

 

*** I use the Sharpe ratio as a measure of risk-adjusted return as it is the most widely used measure. More about my interpretations and the limitations of the Sharpe ratio and why I concentrate on risk-adjusted returns rather than absolute returns in section 7: Realistic goals and expectations.

 

**** A great resource on risk premia is the book „Expected Returns“ by Antti Ilmanen, it´s a tough read, but so worth it.

 

 

4 Comments Add yours

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s