A guide to forecasting fortunes

January 16, 2016 08:00 AM

It’s prediction season and pundits, gurus, sell-side strategists and technical analysts are making their 2016 calls. Predictions will be based on everything from 48-factor models to lines on a price chart. 

They will be confidently delivered, highly specific and, very likely, wrong.

Markets are difficult to predict. Too many economic, geopolitical and corporate events determine performance, and most of them are unknown and unforeseeable. Even if you get the event right, you also must get the timing right and correctly judge the reaction of the market. Without a framework and probabilistic approach, forecasts are little more than wild guesses.

Of course, if an overconfident pundit is wrong, no one will care to remember because everyone else was likely wrong. When we break down why a prediction missed the mark, we can point to any or all unexpected geopolitical events, or you we just blame the Federal Reserve. 

But everybody keeps doing it because, if you make a lucky guess just one time, it’s a golden ticket to personal prosperity. Wall Street will throw money (clients’ money, not theirs) at you to manage. They will name mutual funds after you. Subscribers will cough up cash for your trading service. You will grace magazine covers. You’ll appear on CNBC’s Squawk Box and drink coffee with Joe Kernan.

And even if you’re wrong next time, you’ll always be known as the person who “called” that one event. 

You will earn a year or two of fame and fortune before the next string of predictions goes wrong and clients give their cash to the next guru in line. Wall Street is littered with one-time famous gurus whose long-term track record doesn’t match bold one-time calls. But – in a world of click bait headlines and two-minute interviews, they’ll always be remembered for a single moment in the sun. 

How people reach their projections is as important as the projections themselves. Attend an investment conference or attempt to dig into some of the forecasts on certain websites and you’ll be quite surprised at the pedestrian nature of some methodologies. It would be funny if Americans’ money wasn’t on the line.

Years ago, several colleagues of mine got together over beer and football, and we decided to create a list of boiler plate predictions that flow from the private pages of self-described market gurus. It’s a road map to become America’s next top CNBC forecaster, even if you have no idea what you’re doing.

First, you’ll only want to forecast on whether the market will go up or down.

The language you use is important. For the stock market, the general rule is to go with the positive drift and predict that stock prices will go higher in the year ahead. However, if we’re in a bull market that has lasted more than the average 51 months, predict a down year. If we are in a bear market that has lasted more than 18 months, you suggest that “prices may continue falling” before “finding a bottom” later this year.

The bond markets also have rules. If stocks are going to be up, say you are positive on bonds. If the model suggests that stocks will fall, then bonds will “test lower levels.” Typically, U.S. dollar moves last between five to seven years, so it’s important to follow an upward trend until year six. Then you’ll “expect weakness.”

Always, always project that U.S. GDP growth will come in “around 3%,” the historical baseline average that everyone uses as a security blanket.

For specific sectors, use both mean reversion and momentum. You’ll want to suggest that last year’s worst sectors will recover and the prior year’s best performers stay strong in the early part of the year and “possibly” fall off in the second half. 

The word “possibly” allows you—without actual numbers or probabilities—to blame a geopolitical event when your forecast falls flat.
For commodities, you can use a two-year mean reversion. Go ahead and say that the best performing commodities from two years ago will decline and the worst will improve.

To make this work, you must be very vague about your forecast. Never give a target price or percentage as people might actually hold you accountable. You also need to make lots of predictions so you have more chances of picking a single needle from the haystack.

If you followed these rules in 2015, here’s how you would have done:

  • You would have predicted that stocks would be down based on the 69-month bull market. (Though you’d be slightly off, you would excuse yourself by suggesting that if you take out the top 10% market cap stocks, then “the rest of the market is down,” and you would have the Fed’s lack of action to blame.)
  • You would have projected a downturn in bonds and nailed the dollar rise. 
  • You would have suggested that utilities and healthcare were strong early. (Utilities did exactly that while healthcare stayed strong). 
  • You would have called for energy and telecom to reverse higher. (Getting one right and one wrong). 
  • You’d have forecasted the selloff in oil and natural gas, but you’d have lost badly on zinc and wheat.

Now we can write this up as you nailed the weakness and volatility in stocks, were spot on in bonds and correctly called for a collapse in metals. You made an epic call in oil and gas while taking small losses when stopped out of wheat and gold. 

You’d have been right to sell any strength in utilities, although you took profits too soon in healthcare stocks. You were stopped out of energy stocks and correctly called the strength in telecom.  You were off a little on GDP, but you can just blame “poor economic policies from the current administration.” 

Remember, you never ever bring up the fact that you were wrong about anything. 

Focus only on the positive forecasts. If asked about your incorrect call on wheat, pivot to oil prices.

You can even go so far as to brand yourself the expert who called the 2015 oil crash. If you are lucky, no one will mention that crude oil began crashing in mid-2014. 

What's ahead in 2016?

And—as you can see—on a model forged on lots of beer and little effort, you’d be on the verge of fame and fortune as a great market predictor if we cherry-pick the results. 

But 2016 looms, and you know how and why “this model” was written. 

Are you willing to follow these rules? Although you might have performed great on some commodities in 2015, you’d have crashed spectacularly on others. In 2016, the hastily written rules and loose indicators suggest that stocks will decline, bonds will decline, and the dollar will rise. Energy and railroad stocks will surge by year’s end. Internet and construction materials will be strong early in the year, but they “probably” will fade later in the year. Energy and agricultural commodities will be higher, while nickel and zinc will weaken. 

Oh, and just in case you forget, GDP growth will be “around 3%.”

About the Author

Tim Melvin is a 27-year veteran of the financial services and investment industry who has served as a broker, advisor and portfolio manager. He specializes in value investing. @timmelvin