The 26.6 million people who tuned in to last year’s Oscars ceremony were the fewest to do so since analytics company Nielsen began estimating the program’s viewership in 1974. That figure continued a four-year downward trend, with viewership for the 2017 broadcast — covering movies released in 2016 — coming in at the third-lowest total. (Only the show that aired in 2008 was lower.)

(The graphs in this article use the film release years, not the ceremony broadcast years.)

Why was last year’s total so low? Any number of reasons, certainly. But one reason may have been a fairly obvious one: People were less likely to have seen the nominated movies and, therefore, to care about who won.

Bruce Nash of data company the Numbers provided The Washington Post with two sets of data for every best picture nominee since 1985: how many tickets were sold for each film and how many tickets had been sold at the time the pictures were nominated. By the date of the announcement of the nominees for 2017 (Jan. 23 of last year), only about 63.8 million tickets for the films had been sold, about the middle of the pack for the 33 years from 1985 to 2017.

But that was for nine films, given that the Academy of Motion Picture Arts and Sciences began allowing more than five Best Picture nominees about a decade ago. On average, each nominee sold about 7.1 million tickets, the 25th best average of the 33 years. The films released in 2016 had the 28th best average at 6.2 million.

If we compare those averages to the broadcast ratings, the pattern doesn’t immediately jump out at us.

Pay attention to that spike on the far right which represents average sales for films released in 2018, the movies nominated in Sunday’s Oscars telecast. We’ll come back to that.

Let’s talk about that big spike in 1997 that correlates to the big spike in ticket sales. That year, film junkies will know, an interesting movie won the best picture award.

Let’s not spoil it yet, though. Nash’s data also allows us to see the average ticket sales after the best picture nominees were announced. The best picture nominees in 1997 on average earned more sales after they were nominated than the average number sold before nomination in 23 of years from 1985 to 2017. Remember: These movies were theoretically released the year before their nominations (though for best picture nominees, that gets a little blurry).

So what happened in 1997?

“Titanic.”

“Titanic” was a popular movie, as it turns out. It sort of breaks the scale, in fact, as we’ll see below.

But notice something else that’s happened in recent years: The pictures that didn’t win best picture have been averaging more post-nomination ticket sales than the ones that did. That happened five times from 1985 to 2009 — and it happened in each of the five years from 2012 to 2016. Last year’s winner, “The Shape of Water,” bested the average for the films it was competing against. In 1997, “Titanic” sold more than twice as many tickets after nominations were announced as its best picture opponents combined.

We got that whole “Titanic” thing out of the way because it’s going to jump out at you on the next graphs. The first compares Nielsen viewership data with average ticket sales for best picture nominees. At upper right, leading the pack on both metrics, is 1997. As the diagonal dashed line shows, there’s a link across years: years in which fewer tickets were sold for best picture nominees generally saw lower Oscar ceremony viewers.

(Circles are scaled to the number of tickets sold after the films were nominated.)

There are a few years that really fall out of that pattern, including 2017 (last year’s broadcast) and 2002, when “Chicago” won. Notice, though, how many more tickets were sold on average in 2018. It’s at the front of the pack on ticket sales for best picture nominees, meaning that the awards ceremony should see higher than normal viewership. (“Roma,” distributed primarily on Netflix, was excluded from our data. More on that in a moment.)

Let’s tweak the metric a bit. If we compare ticket sales for the best picture nominees to all of the tickets sold in the calendar year (using data from BoxOfficeMojo), 2018 stands out even more. The total number of tickets sold for the best picture nominees before they were announced on Jan. 22 was equivalent to 1.5 percent of all tickets sold in 2018, the fifth-highest figure since 1985. That, too, would suggest fairly high Oscar viewership.

Why are 2018′s numbers so good? In part because the top-grossing movie of the year, “Black Panther,” was also a best picture nominee.

But we have to go back to that recent downward slide that left 2017 setting all of the wrong records on viewership. How much of that was a function of services like Netflix eating into movie ticket sales? Well, according to BoxOfficeMojo’s data, sales peaked in 2002 (even though viewership for the ceremony the following year plunged). Since then, it’s been a static downward slide.

2018 saw the third-highest average ticket sales for best picture nominees since 1999 but the fourth-lowest ticket sales since 1995.

This year’s Oscar broadcast, in other words, will test our theory. Viewership should be up, given the popularity of the year’s movies. Maybe enough of the people who saw “A Star is Born” and “Black Panther” will care enough about the award to tune in. If the viewership slide continues — not unthinkable in part given the uncertainty over the broadcast — the problem with the Oscars probably runs a lot deeper than nominating relatively unknown pictures.