There’s no question 2016 got off to a rough start, but it was almost on par with the past two years.

This year, the S&P 500 finished January down 5%. By comparison, the index fell 3.5% in January 2014 and dropped 3.1% in 2015. So the rough start seems to be the norm for recent years.

According to the popular January barometer, the month of January should forecast stock performance for the year. But in 2014 and 2015, it failed to do so. This has caused many analysts to claim the January barometer is broken.

I say it’s not.

In fact, it’s actually signaling a 20% drop in the market this year…

The January barometer is the idea that if the S&P 500 finishes January in the green, the year will likely be positive as well. Likewise, if it finishes in the red, the year should be negative.

According to the January barometer, both 2014 and 2015 should have finished negative because they both started the year in the red. But both finished positive.

In 2014, the S&P 500 was up 12%. In 2015, it finished the year basically flat; down just 0.6%, but after accounting for dividends, it was positive.

As simple as this barometer sounds, it’s still surprisingly accurate. So accurate that it’s telling me there is a 1% chance of the market being positive this year — and, if it’s negative for the year, we are likely in for a market crisis drop of 20% or more.

Low Probability of a Positive Year

Over 80% of the time, the January barometer is correct in foretelling how the year will wrap up. This has stood the test of time, dating back to the 1950s, and has never had three wrong years in a row. In fact, it has had only eight major misses since 1950.

That tells us the odds of going for a third year in a row of not following the January barometer are less than 1%. (To get the odds, you multiply 0.2 — which is the 20% chance of the barometer failing — by 0.2 and then multiply it by 0.2 again, equaling a 0.8% chance of the marketing failing to drop in 2016.)