• by David Goetzl  @dgoetzl, May 8, 2013

TV viewing may be increasing on a national basis, but that’s hardly the rule in major markets. By one metric, Nielsen says about half of them saw viewing declines in February, including one by an average of nearly 40 minutes a day.

Among the 25 markets measured by LPMs (local people meters), 12 saw a drop in average daily viewing -- live plus time-shifted -- for the 25-to-54 demo. The February data came during the first sweeps period of the year. 

Reviewing the research, there doesn’t appear to be any rhyme or reason to the year-over-year fluctuations. Why did Charlotte have the largest drop -- a 39-minute daily average -- and Orlando the highest increase at 25 minutes? Why did Dallas, Denver and Philadelphia -- from three different regions -- all experience an average 11-minute fall-off?

Certainly, the Feb. 3 Super Bowl would send viewership soaring in the two markets with teams competing?

Nope. With the winning team, Baltimore had an 8-minute average increase. But losing San Francisco had an 8-minute drop.

Perhaps ecstatic Marylanders kept watching afterglow coverage in heavy numbers all month, while many San Franciscans just couldn’t bear to turn on the set for a while.

(The collective data doesn't necessarily reflect success or failure at any particular station.)

The country’s three largest markets didn’t move in lockstep. New York had an average daily increase of 21 minutes, while Los Angeles was down 15. Chicago was up 3.

A contributor to the variations may be the interest in local news at any one point. Washington, for example, saw an average 15-minute increase not long after President Obama began his second term.

Cleveland had only a 1-minute bump. But the current story about three young women saved after years in captivity should bring a major bump when May sweeps data comparisons are computed in 2014.

Does weather play a role? Without knowing specifics about rain levels or unseasonably hot or cold temps, it doesn’t appear that way.

Consider Florida. Orlando had the 25-minute increase in viewing, but Miami had a 23-minute drop. (Tampa was down by 6 minutes.)

Chew on this: Chilly Pittsburgh and Minneapolis saw viewership up notably, each over 20 minutes. Phoenix? Up 17 minutes.

Could levels of cord-cutting be a factor in certain markets? Markets with lots of over-the-top (OTT) streaming could experience a viewing decrease.

Yes, Seattle had more streaming than any other LPM market with a daily average of 8 minutes among 25-to-54 year-olds. And TV viewing was down by a 24-minute per day average.

San Francisco and Detroit tied for second in OTT streaming levels at 7-minute averages. And TV viewing was down in the Bay Area. But the Motor City breaks the pattern with an increase. 

More on the impact of streaming: Nielsen data shows the San Francisco market had the highest tablet penetration at 36% during the February sweeps. Boston was second at 35%. Both tech-oriented areas saw TV viewing fall.

Surely, people were spending time with iPads and turning away from the set? Well, Washington and New York both had similarly high tablet penetration and significant viewership bumps.

Want more confusion? One could argue that with their cutting-edge technology, smart TVs might increase traditional or DVR viewing. Of course, if people are using the Internet-connected sets to watch Hulu or Netflix instead of CBS or NBC that might offer contrary evidence.

Make of this what you will: San Francisco had the highest smart TV penetration among the LPM markets at 8%. Seattle was next at 7.6%. Both saw TV viewing down.

Viewership was also down in Los Angeles and Denver, two other top-5 markets in smart TV penetration. But Washington was also in the group, but it had a 15-minute average viewing bump.

A cynic – which might be every station chief with a viewership drop – might blame Nielsen for a slide, suggesting its methodology leads to wide swings. But does that hold any more credence than any other theory? American TV viewers are one of the most inscrutable species around. Maybe that's just it.