One hundred and seventy-four days remain until the United States’ midterm elections (902 until the next presidential election, but who’s counting) — which means there’s still time to “evolve” how polling is conducted.

The 2016 presidential election wasn’t polling’s shining moment, with many post-mortems pointing to opinion polls misleading election forecasters and underestimating now-President Trump’s support. It didn’t help that some polls were tied to news organizations that don’t really have the resources anymore to support this work — at least doing this work well. There’s no perfect poll aside from (maybe) the ballot itself, but the polling system — both conducted by the media and reported on in the media — has faced critics since long before November 8, 2016.

These issues contributed to the Associated Press’ and Fox News’ departure from the Election Day polling data shared by the major networks last year. But now the wire service has built a new-and-hopefully-improved election data-gathering system made up of comprehensive surveys and online polling. It tested the new approach in several 2017 state elections; for 2018, both Fox News and The Washington Post have signed up to receive AP’s results from at least some states on Election Day.

(Note that the “old” exit polls aren’t necessarily going away; ABC, CBS, NBC, and CNN have each committed to them through 2020, according to Edison Research, which operates those polls. But Politico described the old consortium, known as the National Election Pool, as being “on life support” a few months back.)

The AP worked with NORC at the University of Chicago to develop this approach, “aimed at evolving the traditional, in-person exit poll” to harness early and absentee voters and dig into the reasoning behind the voters’ picks.

The AP will survey more than 85,000 voters for this year’s midterm elections, compared to the 19,400 interviews with voters in the exit polls four years ago.

The test of this method in the New Jersey, Virginia, and Alabama elections was “promising”:

Instead of stationing in-person interviewers outside of polling places, NORC conducted telephone interviews for AP and Fox News with a random sample of registered voters in New Jersey and Virginia in the Nov. 7 general election and in Alabama for the U.S. Senate special election on Dec. 12. In each state, NORC also interviewed a much larger non-probability sample via the internet, and used sophisticated statistical techniques to combine the two surveys. All interviews — about 4,000 in each state — were conducted beginning 96 hours before Election Day until the polls closed in each state.

At poll close, the new survey approach estimated that Democrat Ralph Northam would win the Virginia governor’s race by a 52-46 margin. His final margin was 54-45. The New Jersey survey estimated at poll close that Democrat Phil Murphy would win that state’s governor’s race 57-38. His actual margin was 56-42. And in Alabama, the survey estimated at poll close that Democrat Doug Jones would win 50-47, while his actual margin was 50-48.

We’ll see how November 6, 2018 goes.