When we look back at 2016, some very strange and counter-intuitive things happened in sequence. We believe this phenomenon will repeat in the years to come, and market participants just have to get used to investing with decreasing certainty in the background.

First, both with Brexit and the U.S. elections it became clear that forecasting political events is hard, if not outright impossible. Second, even if we had been able to forecast the political outcomes correctly, we would not have been able to forecast market reactions to the events with a high degree of confidence. And finally, even if we had been able to get the first two right, we would not have been able to forecast the speed and magnitude of the market reaction.

To me, this trifecta of failures to anticipate and properly position is a sign of the times. Probabilistic computations are everywhere, and they get updated in real-time. We see survey results and “odds” through many election websites and on “markets” where we can bet on outcomes. We can also extract probabilities of expected market outcomes from option prices, which are available in real-time. Given all this information, participants rationally discount the probability weighted information, position portfolios based on expectations from this information, and when the facts are actually revealed, rapidly rebalance their portfolios in light of this new information. Actual (yet unknown) probabilities probably do not change much, but information and perception can change a lot. And in a fine-tuned and tightly coupled world primed to react to new information with speed, counter-intuitive market reactions are going to be the norm.

A simple thought experiment will make this clearer. Suppose we have a bucket with two balls, one black and one white. We remove one ball, but do not look at it. What is the probability that this ball is black? Clearly, given the information so far, the probability is 50%. Now suppose we remove the other ball, and find that its color is white. Now we ask the same question, what is the probability that the first ball we took out is black? Clearly, the answer now is 100%. The only thing that has changed is the amount of information. The actual, physical world has not changed at all! But in an instant, the probability estimates have changed in light of new information.

This simple example highlights the subtle nature of probability as it applies to the markets. If probability is a measure of the actual state of affairs in the world, then the probability of the first ball being black should not change when the second ball is revealed. On the other hand, if probability is information, then it should change as new information becomes available. Given the plethora of data we are inundated with, our estimates of probability (and hence pricing) in the markets is basically a summary of the information available to us at any given time. There is no guarantee that this information is accurate, or in any way a true reflection of the underlying state of affairs – it could possibly be accurate, but until new information is revealed, we just don’t know. In other words, probability when viewed in this way is something attributed by the participant rather than being a property of the real world. And because it is a perception, it can change quickly.

So short of turning off our news sources and market data feeds, what can we do?

Given the difficulty in forecasting, we can, instead, focus on building a portfolio with structure that is more resilient to information shocks. In simple terms, this means three things: (1) try to be on the right side of the market, (2) try to earn yield while we are at it, and (3) try to build sufficient downside protection for the portfolio. In more technical terms, this means to generally follow the trend (or at least not to fight it unless there is a really good reason), if possible earn “carry” (or at least not pay too much negative “carry”), and spend a small sum to protect the portfolio to stay invested (as long as the cost of the protection is reasonably small).

As an example, there is one investment today that has all three of these elements – investing in the U.S. dollar and dollar assets.