As Albert Einstein reminded us, “In order to win, you first you need to understand the rules of the game. Then, you need to play better than everyone else”. A corollary applied to the financial markets game could read: ‘Playing by the wrong rules will most likely lose you money’, and there is no rule more important than the objective of the game.
Let’s look at a simple numerical example to explain the difference between ‘investing to make money, or to be correct about our predictions’. For simplicity, let’s assume a market scenario with two binary outcomes (‘up’ or ‘down’) with a 90 per cent probability of making $1 if the market is ‘up’, and a 10 per cent probability of losing $100 if the market is ‘down’.
A very asymmetric outlook of crossed probabilities and payouts which could correspond to a bullish event being almost fully priced-in (not far from today’s complacent global markets).
What is the right investment decision?
Well, if we are being judged by our ability to predict the market (as it is the case for many strategists and analysts), we would probably bet for ‘up’ as it will win 90 per cent of the time.
On the other hand, if we are being judged by our ability to make money (as it is the case for traders and investors), we should bet for ‘down’ as the expected gain is $10 = 10% x $100 if the market is down, much greater than the expected loss of $0.90 = 90% x $1 if the market is up.
From the above it is clear that as investors, the correct decision and objective is to make money (not to be accurate in our predictions), which must take into consideration the asymmetry of both probabilities and outcomes.
But making the right decision is hard. Let me review some of the common obstacles and considerations.
First, it is difficult to have a true assessment of the forward-looking probabilities. We have historical probabilities (what happened in the past), implied probabilities (what the market is expecting based on the pricing), and conviction (about what forward-looking probability may look like), but no one really knows what those probabilities actually are.
On many occasions, our conviction can be very different from the true forward-looking probabilities due to a range of biases and imperfect information. As one of my old bosses used to say, “Ninety per cent of the drivers think they are above the average”, which is obviously wrong, as by construction only 50 per cent of the drivers can be above the average. From there, we conclude that at least 40% of drivers have an inflated view of their driving – same thing for our conviction about the markets.
Our over-rated conviction may lead to larger position sizing and higher concentration of risk than what is warranted, and can be particularly dangerous when our views are in line with the consensus in the market, as the expected gains tend to be largely priced-in, exposing us to large asymmetric risks to changes in fundamentals and positioning. As Heraclitus reminded us, “The only constant in life is change”.
Second, it is difficult to make the right decision because the true risk-reward and asymmetry of payout can be blurred by misconceptions and speculative positioning, among many other known unknowns and unknown unknowns.
Fortunately, we have tools to control the asymmetry of payout. Stop losses are designed to limit the damage from our mistakes, for example. Profit-taking levels are the mirror image of stop losses, and are designed to create asymmetry of payout in our favour. As George Soros reminded us, “Once we realize that imperfect understanding is the human condition, there is no shame in being wrong, only in failing to correct our mistakes”.
Third, it is difficult to make the right decision because of psychological aspects. In our example, positioning for a ‘down market’ (good decision) will lose money (bad outcome) most of the time (on average 90 per cent of the time).
Worse, whilst we should expect to get one win for every nine losses, probabilistically, it is very possible that we will get a longer streak of losses without a single win. Based on the assumed probabilities, for example, the probability of 20 consecutive losses is 12%, which is non-negligible.
Long streaks of losses can be exhausting from a psychological perspective, and dangerous from a financial perspective. In order to remove luck from the equation, we must adjust the size of our bets needs and ensure they are small enough to sustain the most unlucky of streaks. Position sizing is in fact the single most important factor in trading success, a topic that requires and deserves much more detailed discussion, perhaps in a future article.
It is also worth noting that in our example, positioning for an “up market” (bad decision) will make money (good outcome) most of the time (90% of the time). Rewarding the wrong behaviour keeps bad players playing, but eventually normalizes in the form of loses.
Another important consideration is the polarization of views implied by highly asymmetric outcomes. Polarized markets often lead to significant volatility from small changes in fundamentals or sentiment; in a dynamic I call “today I am right, tomorrow you are right”.
As Keynes reminded us: “The market can remain irrational for longer than I can remain solvent”, which is why we need to be very careful with leverage as it has the potential to take us out of winning positions and lead to the painful outcome of “I was right, but wrong”.
As a result of the considerations above, I would recommend that we de-emphasize conviction (our ability to predict the market) and emphasize disciplined trading and position sizing (our ability to make money), and why I believe in the aphorism “bulls make money, bears make money, and pigs get slaughtered”.
Following up from my article “Gold’s Perfect Storm” in the Financial Times earlier this year, I just finished writing a book called The Anti-Bubbles: Opportunities From Lehman Squared and Gold’s Perfect Storm, a contrarian framework that presents the risks and unintended consequences of monetary policy without limits and credit expansion without limits.
From my analysis, I have a high level of conviction that the gold market is at the beginning of a multi-year bull run. A bullish super-cycle that offers investors a highly asymmetric risk-reward with ‘a few hundred dollars of downside and a few thousand dollars of upside’. It is worth noting that many respected analysts and investors hold the exact opposite view; a high level of conviction that gold prices are going through a quasi-permanent bear market and will continue to go down ‘forever’.
Time will tell who is right and who is wrong, but, frankly, it is secondary. Focusing on being correct about our predictions misses the point and true objective of the investment game: to make money, and why I find it helpful to present humble investments ‘theses’ that need to be proven right or wrong as new information and evidence arise.
As Karl Popper taught us, “One million positive observations do not prove a theory right, but one single negative observation proves a theory wrong”. As of today, my investment theses of Lehman Squared and Gold’s Perfect Storm are yet to be proven wrong.
The investment then thesis needs to be implemented efficiently, which is “an art, not a science, but it helps to know a lot of science” – an art that looks for positive asymmetry, as much as possible, always with the primary objective of making money.
Diego Parrilla is regular contributing editor at Portfolio. This is the first of a series of his articles for the publication. Diego is a macro commodity portfolio manager and co-author of The Energy World is Flat. Follow him on Twitter @ParrillaDiego.