At first blush, Logan Lucky does not look like a model of film-industry innovation. If anything, the heist film, starring a glitzy cast (Channing Tatum, Daniel Craig, Riley Keough), seems like a regression for director Steven Soderbergh, who is an old hand at the genre thanks to his twisty and shiny Ocean's movies. But underneath Logan's tale of noble West Virginian thieves and their scheme to rob the Charlotte Motor Speedway lurks a new, brighter future for independent film.

Or that's the hope for Soderbergh, who is using Logan to mark his return from self-imposed feature-film retirement – even if he's spent the past four years since directing Behind the Candelabra helming both seasons of television's The Knick and producing a rash of other projects (from Magic Mike XXL to the series Red Oaks and The Girlfriend Experience). Ahead of Logan Lucky's release Aug. 18, The Globe and Mail had an in-depth conversation with Soderbergh about his quasi-return, the pure cinematic joys of a caper flick and why "it's a full-time job just watching things with my name on it."

Congratulations on the end of your retirement. Did you achieve that level of calibration that you were hoping for, when you first announced the move back in 2013?

[Laughs] Well, this all evolved in a very organic way. I'm a big believer in timing, so I was following what I thought were some planets aligning with coming back to this very specific kind of work, of getting this script by Rebecca Blunt [who may or may not exist; reports that appeared after this conversation took place have postulated that the screenwriter may actually be Soderbergh's wife, Jules Asner, or Soderbergh himself]. I was mostly involved to help find another director for this when I got it in the fall of 2014. It happened to coincide with some conversations I was having with friends of mine in the studio-distribution business and the exhibition business, conversations I'd been having for years. It seemed that there was an opportunity afoot for someone to, with the right kind of project, self-release a movie on several thousand screens. With the technology now, and with a certain kind of film and a certain kind of cast and pedigree, you could do this. You could create your own distribution company and get this out wide. So I said, "I'm going to do this."

What are the details of the distribution model, though? In Canada, it seems that Logan Lucky is coming out in a traditional way, with a theatrical distribution by Entertainment One. What makes your new distribution company, Fingerprint, so innovative?

We're not doing anything that anybody hasn't done before, but what we've been able to do is combine things and recalibrate them in a way that's beneficial to the filmmaking team. There's a long history of movies that were financed by preselling foreign rights in order to cover the budget, and then making a deal in the U.S. that becomes the upside to that film. That's something Dino De Laurentiis was doing in the sixties. So here, E1 would be part of that, with the non-U.S. rights going to various distributors to cover the budget of the film. Then we sell the non-theatrical rights for enough money to cover advertising and marketing, so that when the film actually opens, we're essentially already at zero. The only thing at risk is the time we all took to make the movie, because we all worked for scale. If it works, great. If it doesn't, we learned something.

Are there other projects in the pipeline for this model?

I've got some in the holster waiting to see if Logan works. Our definition of something working, though, is different than the studios'. If our movie opens to something with a "two" in front of it, that's like a studio movie opening to something with a "four" in front of it.

What was it about Rebecca's script that fit so neatly into this model?

It was an extremely commercial piece that we could cast the hell out of. What got me amped was that it's obviously the kind of film I like to make, but a variation. It wasn't exactly like the Ocean's movies; it's a cousin to those films. It inverted the model in an interesting way, so these characters don't have money, technology or opportunities. They're not even criminals at the beginning of the film; you get to watch them become larcenists. The layers of each character also get peeled back gradually over the course of the film. Each character has the opportunity to surprise you.

There's that great line toward the end about this crew being an "Ocean's 7-Eleven." Was that in the original script?

Yeah, and I thought that was an appropriate shout out. Typically, if this were not a comedy, I would've said that's a little too self-conscious. But given that's the universe of this movie, we can get away with that.

It seems like the heist movie is an endlessly enjoyable genre to revisit.

It's a type of story that movies do really well; it just lends itself to the things you can do in movies. It typically features a real eccentric group of characters, and that's fun to cast. They're also like little puzzles. Ever since I was young, I was captivated by caper films. I mean, there's nothing in my upbringing to suggest that I would have that interest, but they always seemed like "movie movies." Movies squared, somehow.

Is this something where you like to reverse engineer the plot to make sure all the puzzle pieces fit?

You have to make sure the math works, but you're balancing the plausibility against granularity. You want to have enough detail so that you're suspending an audience's [disbelief] but also not getting so bogged down that they're ahead of you and waiting for you to finish an idea. On the Ocean's films and on Logan, I went back and did some surgical, strategic reshoots to address various issues by either filling in a piece of detail I thought we needed, or I needed a little bridge piece to make a cut work. There are instances when you screen it, when you test it, and the cards come back and people go "that was easy" or "I didn't understand that."

So you find test screenings useful?

Sometimes. You have to be very clear what you're trying to pull out of those screenings. In this case, it's very clear – you're making a comedy and you want to know if the comedy is working; the pacing and the confusion – is the audience ahead of you? What you don't want to be as a filmmaker is in a situation where the test results are being used as a hammer to make you do something you don't want to do. I've been lucky in that I have enough control to avoid that, or I've had the protection of my producers to keep that from happening.

Coming back to cinema after four years, did you feel like a different kind of director than the one who made Candelabra?

Yeah, and mostly because of The Knick. We were able to schedule Logan much more aggressively than we would have in the past because of the rhythm we fell into on The Knick.

Is that an aggression you'll continue pursuing?

I don't know how to work otherwise now. I like to work fast, and there's no way to go back from that kind of speed. I'll just keep pushing it.

How important is it for you to surround yourself with familiar faces, like Channing and Riley?

You have to cast the best person for the part, but when that's someone who you have a relationship with and a shorthand with, it just makes everything that much more fun. It was pretty obvious reading this script that it was a great Channing part, and Riley is someone who can do pretty much anything. Adam [Driver], I was just a fan of, and Daniel, I've known for a fairly long time, since the mid-aughts when he was in a film I produced called The Jacket. Daniel recognized it's a great opportunity for him to annihilate everything else he's done before. I told him, "I don't care how you look or sound, you get to build this character from the ground up. Just to be clear, I don't expect you to protect anything." And I was thrilled with what he came up with.

I'm curious – you're using your own name here for editing and cinematography, although you usually work under pseudonyms.

Where? What do you mean?

It lists you as the cinematographer and editor on IMDb.

Oh, IMDb is wrong. I would never do that. Yeah, on this film it's officially "Peter Andrews" and "Mary Ann Bernard," as usual.

In the past, you've gotten flack from the guilds for taking on these jobs. Have they let up on you?

I know it's not a popular notion for a director to be their own cinematographer. That being said, over time, people have stopped thinking about it. It's clear that it's something I take seriously and something I'm going to continue to do. I think people realize it's not a joke, and they've relaxed.

Is it just part of your ingrained process now?

Yeah. It's when somebody says, "I don't understand how you divide yourself." To me, it's the opposite. I'm actually not divided now, where I felt like I used to be. This is the way I started, when I began making films. The technology now allows for this kind of process to be really organic, between the ability to shoot with less light to, within an hour and a half after a day's wrap, having the footage on my laptop and being able to cut right away. I wish I had this stuff when I started out.

With the advent of such tech, is it a better time than ever to be an emerging filmmaker?

For sure. I did an [Ask Me Anything] the other day on Reddit and somebody said, "I want to make this film, what do I do?" I said, "Do you have a phone? Then get a group together and shoot it." The technology is insane, what you can do on your phone. It's a 4K capture! I'm shooting stuff all the time to play around with it and see how I can push it. It's jaw dropping. If you have a phone, you're in the movie business.

But what about the other half of the equation, the market?

It's easier than ever to make a movie, but harder than ever to get eyeballs on it. I mean, I'm trying to create a path for a certain kind of filmmaker who wants to make a certain kind of film in wide release. I didn't create Fingerprint to become a boutique distributor.

Is there hope for those smaller companies?

I think you have to be very clear on how you define success. For a young filmmaker starting out, do you look to somebody like Joe Swanberg, who makes a movie every year or so and seems to be doing fine? I don't know Joe, but I believe he's like me, in that he keeps getting to make movies and that makes him happy. When I was growing up, if you told me I'd get paid $250 a week for the rest of my life, but I could only make films for the rest of my life, I would've jumped at that. All I wanted to do was work. You have to determine what your priorities are and what version of the business you want to participate in.

And now Logan's performance is your priority.

My future is sort of an open question, depending on how it performs. It's an interesting time, in terms of what people pay attention to when they're deciding what movie they want to see; it's a continually evolving algorithm. We've been trying, for example, to take advantage of the data mining that's been available, to understand who our potential audience is and how to keep them on the boil until the movie opens. I was always mystified by the traditional approach to marketing, when I saw money being spent on exposing your content to people who are never going to see the film. For instances, 13- to 17-year-olds, for some reason, have really attached themselves to the material for Logan that we've been dropping. They're watching everything! I didn't think that was a target audience making this movie, but the data says they're interested, so we're shifting our approach to the digital spend. I was shocked.

Would you ever let that data influence the content?

No, but you know, as far as trying to find the audience for it, it's very helpful.

What can you tell me about the Mosaic project you've been working on for HBO? It seems rather mysterious.

Well, I'll confirm you can touch it and make it do certain things. You engage with the story, and are presented with opportunities to navigate it. It's a very complex edit, but the technology part of it is almost complete. I've never been through this before, creating an app, and how hard it is to make something feel intuitive to the user and be elegant at the same time. Hopefully this thing will be out for people to touch in November. Finding a different way to lay out a narrative is something I've always been interested in, and this was a unique opportunity to do so. Ed Solomon wrote the whole thing, and if you saw the giant board we had in the writing room, you would get dizzy trying to keep it all sorted out.

So there's that, plus no shortage of other projects. I'm thinking the new seasons of Red Oaks, The Girlfriend Experience …

It will be a full-time job just watching things with my name on it.

This conversation has been condensed and edited.

Logan Lucky opens across Canada Aug. 18.