Stocks quoted in this article:
Ah, momentum. It's a topic of debate in both sports and finance. Sportscasters are fond of the concept, crediting momentum over and over again any time a team strings two good plays together. The stats community is rather dubious of the concept. It confuses coincidence and the random nature of tiny sample sizes for cause and effect.
Advanced NFL Stats has run a series of tests that call into question the whole concept of momentum in football games. In his most recent edition, he ran something called the "Runs Test" on several years worth of NFL games.
This article will examine how 'streaky' NFL games tend to be. If momentum is real and it affects game outcomes, it would result in streaks of success and failure that are longer than we would expect by chance. But if consecutive plays are independent of previous success, the streaks of success and failure will tend to be no longer than expected by chance. This method of analysis does not rely on any particular definition of a precipitating momentum-swing, as it looks at entire games to measure whether success begets further success and whether failure leads to more failure.
For momentum to have a tangible effect on games, it does not require completely unbroken strings of successful or unsuccessful plays. But if success does enhance the chance of subsequent success, then the streaks of outcomes will be longer than if by chance alone.
For this analysis, I applied the Runs Test to the sequence of plays in a game. This produces a statistic indicating how streaky a string of results is compared to what would be expected by chance. For example, consider the following three strings of results of flipping a coin eight times:
HTHTHTHT, HHHHTTTT, HTTTHTHH
The Runs Test works like this:
The average expected number of runs of the same result (the 'mu' symbol) is calculated based on the number of 'successful' trials (N+) and the number of 'unsuccessful' trials (N-). In the three examples above, the test says we should expect an average of five unbroken 'runs' of one result or the other:
(2 * 4 * 4 / 8) + 1 = 5
Long story shot … by this metric, there's a lot less "momentum" in NFL games than we all tend to believe.
Anyways, it got me thinking. What if I applied this Runs Test to our markets. We're just off a record-tying run of VIX lifts. Does the CBOE Volatility Index (INDEXCBOE:VIX) demonstrate any sort of momentum or streakiness?
Well, by this methodology, we should have seen 115.44 VIX streaks up to the close of Dec. 3. In actuality, we saw 129 runs, which would indicate there's literally less than random correlation in VIX direction from one day to the next.
That begs the question of whether this year's VIX action is typical or not in that respect. And the answer is that it's pretty normal. Here's the VIX Runs Test for each year going back to 2004, and the test over the period as a whole.
And as you can see, there's not much in the way of VIX-mentum. Only three years saw fewer streaks than the Runs Test predicted. I would have expected for 2008 to show the most momentum, but in actuality, it showed the least.
But perhaps it makes perfect sense. Remember that VIX is a statistic, prone to all sorts of day of week and holiday quirks, So it would figure that we'd see random action on a day-over-day basis.
So what if we apply this methodology to the SPDR S&P 500 ETF Trust (NYSEARCA:SPY). Surely there's more evidence of momentum in the market itself.
Well, not really. It's modestly less random, but all in all, not much different from VIX.
So does momentum in the markets not really exist?
I wouldn't go that far. This is just one test, and one specific way of looking at it. We're just analyzing the direction of one day versus the next day. It speaks nothing of the magnitudes or the momentum within a given day. It does suggest, though, that knowing which way the market went today doesn't give you as much help in predicting what it does tomorrow as we would think.
Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.