How many people are switching?
About 5 million people signed up for Netflix, Hulu, and other low-cost -streaming-video services in 2013, so they could watch shows and movies via the internet. Many of these people simultaneously "cut the cord" on their pay-TV subscriptions, resulting in a net decline of 250,000 cable subscribers that year — the first time that total has gone down. While cable companies still have 56 million subscribers among them, Netflix alone has rapidly amassed 36 million. That number is expected to keep rising, as people switch to the cheaper alternative of watching TV and movies on computers and TVs hooked up to the internet. Some networks already offer cable subscribers the option to stream their content, but next year both HBO and CBS will launch streaming services that don't require a cable subscription — the start, some say, of an "à la carte" subscription model. "The television industry is in transformation," says media analyst Jeffrey Kagan. "This entire space is going to be a completely different space in five years."

Why are cable subscribers moving to streaming?
The biggest factor is cost. Cable subscribers pay an average of $85.80 a month (with many paying well over $100), yet watch only about 17 of the 189 channels in their cable bundle. Netflix, in contrast, costs just $9.99 a month. While that subscription doesn't provide access to the latest episodes of CBS's Big Bang Theory, HBO's Game of Thrones, or AMC's The Walking Dead — hit shows generally arrive on Netflix a year after their network premiere — it does open up a vast library of Hollywood films, documentaries, and TV shows. To entice potential subscribers, Netflix, Amazon Prime, and other streaming services also produce their own exclusive series, such as Netflix's acclaimed political drama, House of Cards. Young people, in particular, are adopting this new model of obtaining entertainment: Only 24 percent of 18- to 24-year-olds have cable, whereas 61 percent pay for a stand-alone streaming service.

What are they watching?
The rise of streaming has created a bigger appetite for big-budget, high-quality dramatic series like House of Cards and Showtime's Homeland. Unlike regular, "appointment" TV sitcoms and dramas created by CBS, NBC, and ABC, which limit viewers to one or two episodes a week, streaming allows people to devour several hours of a narrative show in one sitting. Streaming firms, in fact, actively encourage this "binge watching": When Netflix releases shows, including its own original series, it makes all the episodes available at once; when viewers finish one episode, the next one automatically loads up. For one unnamed serialized drama on Netflix, 25 percent of viewers finished the entire 13-episode season in just two days, while 48 percent took only a week. "The internet," says Netflix executive Ted Sarandos, "is attuning people to get what they want when they want it."

Does that mean cable is doomed?
No. Many viewers aren't willing to wait months or a year for their favorite shows to be available online. And an "à la carte" system may be more expensive than people think. The most popular series are scattered across a number of channels, and the cost of subscribing to all these individually — HBO's service alone will be at least $15 a month — could well be as much as a cable bundle. But the biggest weapon in cable's armory is sports. As long as football, basketball, baseball, and hockey fans require a cable subscription to follow their teams, then Verizon, Time Warner, and the others can continue wrapping sports channels into big, expensive bundles. Given how lucrative the status quo is for everyone involved, traditional networks will mightily resist changing their business model. So HBO, CBS, and other traditional networks will try to walk a fine line by pricing their "over-the-top" streaming services low enough to appeal to the growing army of cord cutters but not so low as to encourage most cable subscribers to join the exodus.

So what is going to change?
For many TV viewers — sports fans in particular — not much, at least not in the immediate future. Change in media-consumption habits doesn't happen overnight; after all, AOL still has 2.3 million customers paying for dial-up internet. But as more and more people switch to streaming over the next few years, and tech-savvy millennials replace Baby Boomers as the largest segment of TV consumers, market pressures will grow on networks and cable companies to offer cheaper, more customized menus of content. Inevitably, streaming will disrupt TV the same way the internet disrupted the music and print-media industries — by "unbundling'' content and making it cheaper. "The convenience of on-demand viewing just makes sense for a lot of people," says Roy Price, vice president of Amazon Studios. "This is the new world of TV."

Data-driven drama
When Netflix committed $100 million for the first two seasons of House of Cards, it wasn't as big a gamble as it looked. Netflix executives knew from data analysis of its subscribers' viewing habits that many of them enjoyed the work of creator David Fincher and star Kevin Spacey, as well as the British miniseries on which the show was based. Different Netflix users were even shown targeted trailers: Subscribers who'd previously watched Spacey movies saw trailers focused on him; viewers who'd liked Thelma & Louise were shown the drama's female characters. Streaming is a two-way street of data, enabling Netflix to study not just what subscribers watch but also how they watch it. It can even tell whether viewers fast-forward through the credits or when they pause a show. This feedback helps show creators to tailor their decisions to viewers' preferences about which characters to develop and which color palettes and scenery to feature. Rather than "shooting a shotgun into the night sky and hoping they hit something," says Rick Smolan, author of The Human Face of Big Data, "these guys know what they're aiming at." The danger, of course, is that high-quality TV series will go down the same path as Hollywood movies, with creators and networks looking to replicate past success instead of striving for quality and originality.