What to watch from 2026: the horror, comedy and thrillers worth your evening

The year the cheap film won

2026 will be remembered as the year the economics of film inverted in public: a horror picture made for around a million dollars took more than two hundred million, while the worst-reviewed entry in a long-running franchise became its biggest earner. Both facts are true, and together they explain why this year's watchlists disagree with each other more than usual.

Three genres carried the year — horror, comedy and thriller — and in each of them the most profitable release and the best release were different films. That gap is what makes a recommendation list worth writing this year and worthless if written the usual way, by ranking box office and calling it quality. What follows sorts by genre, says plainly which films are good and which are merely successful, and separates the two wherever they diverge.

What the numbers actually did

The clearest figure is Obsession: directed by Curry Barker, distributed by Focus Features, made for roughly a million dollars, and returning more than $240 million worldwide. Against it, Scream 7 posted approximately 30% critical approval — the worst in the franchise's history — while becoming the highest-grossing film the series has produced, at around $214 million. A cheap film that critics liked outperformed a franchise film that they did not, and both made money. The lesson the industry appears to have drawn is that low-budget horror is the most reliable bet available, and the release calendar for the rest of the year reflects it.

How to choose what to watch

Choose by what the film is trying to do to you, not by its score. Rating tells you whether it succeeded; genre and register tell you whether you wanted it.

A high-scoring body horror and a high-scoring comedy are not interchangeable recommendations, and a list that ranks them together is telling you nothing useful about your Friday evening. The practical sort is: what do you want to feel, how much attention are you prepared to give it, and how much unpleasantness will you tolerate. Those three questions eliminate most of a year's output in about ten seconds and leave a shortlist that is genuinely worth reading reviews about.

How to tell a good year from a loud one

A strong year is one with range, not one with hits. Count how many distinct registers produced something good.

By that test 2026 was genuinely strong rather than merely loud. Horror produced a micro-budget hit, a large-scale studio swing, a body-horror art film and a liminal-space experiment. Comedy produced a Pixar film, a legacy sequel and an independent crime comedy. Thriller produced a franchise continuation, a heist picture, a survival film and a period hostage drama. That spread is the actual signal; a year of five successful films in one register looks similar in aggregate and is a much worse year to be a viewer in.

Where recommendation lists go wrong

Ranked lists fail in three predictable ways, and all three are visible in this year's coverage.

The first is ranking across incompatible registers, which produces a number one that nobody should take as a recommendation. The second is confusing novelty with quality — a film doing something unusual gets marked up for the unusualness rather than the execution, which is how trend pieces become buying advice. The third is recency: films released in the last six weeks are systematically over-placed because the writer has not yet stopped being impressed. Discount anything from the last month by a position or two and the lists agree with each other far more.

Picking by mood instead of by rating

In practice the useful approach is to pick a register first and then take the best-reviewed thing inside it, ignoring the overall ranking entirely. Someone who wants to be scared and does not want to be disgusted has a different shortlist from someone who wants the reverse, and both are inside the same genre. Most disappointing evenings come from taking a list's number one without checking which of those two it is.

The case against ranked lists entirely

The honest argument against all lists is that taste in these genres is more personal than in almost any other, and that aggregate scores are close to meaningless for comedy in particular — the one genre where the same film reliably produces both a standing ovation and a walkout. A list is a recommendation from strangers whose sense of humour you have not verified.

That is right about comedy and overstated about the other two. Horror and thriller have craft components — tension construction, geography, sound design, how a scare is set up and paid off — that can be assessed with some objectivity, and a consensus about those is worth something even when the consensus about the film is not. For comedy the answer is to use lists to find candidates and never to settle them, which is what the section below tries to do.

Obsession: the film that reset the economics

The horror picture of the year commercially, and the one every studio will now be trying to reproduce. Directed by Curry Barker and distributed by Focus Features, it was made for roughly a million dollars and has taken more than $240 million worldwide, which is a return ratio that essentially does not occur in modern theatrical release.

The premise is a simple one executed well: Bear, a shy music store employee, acquires a device that makes his crush fall in love with him, with the consequences the setup promises. Ian Machado Garry's co-star Inde Navarrette plays Nikki, the object of the wish, in the performance that made her a name this year.

It is worth watching for the construction rather than the concept. The film understands that a wish-granting premise only works if the rules are stated early and never bent, and it holds that discipline. Most films with this shape cheat in the third act; this one does not, which is why the ending lands.

Backrooms: the year's most complete piece of craft

Kane Parsons's film, with Chiwetel Ejiofor and Renate Reinsve, is the one critics keep returning to when asked for the best rather than the biggest. It takes the internet's liminal-space mythology and builds an actual film out of it, which had been attempted several times before without success.

The production design and Jeremy Cox's camerawork are the reason. A film set entirely in featureless corridors either finds a visual language or becomes tedious within twenty minutes, and this one finds one — the geography is legible, which is what makes the disorientation frightening rather than merely confusing.

This is the pick for anyone who wants dread rather than shocks. It is quiet, it is patient, and it works considerably better without distraction, which makes it the wrong film to put on while doing something else.

The Bride!: the big swing

Maggie Gyllenhaal's film is the year's largest-scale horror gamble, and Jessie Buckley's performance in it has been described as a force of nature by more than one critic — the sort of consensus that usually survives a year.

It is the entry on this list most likely to divide a room. A stylised, ambitious take on familiar material either carries you or it does not, and the reviews reflect that split more than they reflect any doubt about the craft.

Watch it for the performance if nothing else. Films built around a single central turn are risky and this one is carried, which is not something that can be said of most attempts at the same scale.

Saccharine: the one to skip if you are squeamish

Natalie Erika James's body horror concerns a medical student who discovers that ingesting the ashes of the dead functions as a miracle drug. It is effective and it is genuinely unpleasant, in that order.

Body horror is the sub-genre where the gap between well-made and watchable is widest. This is well made — the escalation is controlled, the premise is followed to its conclusion rather than abandoned — and a substantial share of viewers will still want to stop halfway.

Recommended without reservation for people who know they like this, and not recommended at all as a general pick. It is the clearest example this year of why a single ranked list is a bad instrument.

Send Help: Raimi back in the register that made him

Sam Raimi's return, with Rachel McAdams and Dylan O'Brien as office workers stranded on a deserted island after a plane crash, sits on both the horror and thriller lists this year, which is a fair description of what it is.

It is the most purely entertaining film in either category. Raimi's particular skill — staging violence that is frightening and funny in the same shot, without either undercutting the other — is intact, and the two-hander structure gives it somewhere to go.

The best group-watch on this list. It rewards a room rather than solitude, which is unusual for horror and is the main reason to prefer it over the more technically accomplished films above.

Cold Storage and the liminal turn

The other liminal entry, and a demonstration of how much can be done with one sparse location. It appears on comedy lists as well as horror ones, which tells you something about its register.

The trend it belongs to is the actual story of horror in 2026. Liminal horror — empty spaces, wrong geometry, the sense of a place that was built for people and no longer has any — went from an internet aesthetic to a commercially proven form in a single year.

The reason is partly economic. These films are cheap to make well, because the aesthetic requires emptiness rather than expense, and after Obsession every distributor is looking for the next one that costs nothing.

Scream 7: the money and the reviews part company

Approximately 30% critical approval, the worst in the franchise's history, and around $214 million — the biggest box office any Scream film has produced. Both numbers are worth holding at once.

This is what franchise gravity looks like at full strength. An established audience turns out for the title regardless of the reviews, and the reviews then have no effect on the commercial outcome, which removes the mechanism by which quality usually corrects.

Include it on a watchlist only if you have seen the others. As a standalone recommendation it is the weakest thing in this piece, and as a case study in how sequels get made it is the most instructive.

Hoppers: the funniest thing Pixar has made in years

Described in several reviews as Pixar's funniest in a long time, and unusual for the studio in that the comedy is the point rather than the vehicle. The premise — interacting with the animal kingdom through the body of a robotic beaver — is deliberately silly and the film commits to it.

That commitment is the whole reason it works. Animated comedies that hedge toward sentiment in the third act lose the register they spent an hour building, and this one largely resists.

The safest pick in this entire piece for a mixed-age room, and one of the few films here that genuinely works with children present without being a compromise for the adults.

The Devil Wears Prada sequel: better than it needed to be

David Frankel returning twenty years after the original is the kind of legacy sequel that usually exists for reasons unrelated to having something to say. This one turned out fresher and funnier than a film about the collapse of media had any right to be.

Meryl Streep gets the biggest laughs, with a new run of Miranda put-downs that land because the character has been given somewhere to go rather than simply reassembled. The film's subject — an industry in structural decline — gives the comedy something to push against.

The pick for anyone who wants to laugh without being asked to work. It is the most conventional film on this list and one of the most reliably enjoyable.

I Love Boosters: the year's sharpest independent comedy

Boots Riley's film has Keke Palmer as Corvette, leading a San Francisco Bay Area shoplifting trio. It is the entry most likely to be missed and the one this piece would push hardest.

Riley's comedy operates the way his previous work does: the jokes carry an argument, and the argument is not softened to make the jokes easier. That produces a film that is genuinely funny and genuinely uncomfortable, sometimes in the same scene.

Not a background film and not a consensus one. Watch it if you want a comedy that leaves something behind, and pick something else if you want ninety minutes of relief.

GOAT, and where else the laughs were

GOAT appears alongside Hoppers and Cold Storage at the top of the aggregated comedy rankings this year, and the spread of that trio is the point: an animated studio film, a sports comedy and a horror-adjacent piece, all landing in the same bracket.

2026's comedy strength came from range rather than from one dominant release. The year's output ran from Sundance independents through mainstream Pixar to darker material from A24, Shudder and IFC, and by several critics' count produced more good comedies in six months than most years manage in twelve.

For a viewer that matters more than any individual recommendation. A year with one great comedy gives you one evening; a year with eight across different registers gives you a season.

28 Years Later: The Bone Temple — the year's best thriller

Ranked first on the thriller lists with reasonable consistency, and the rare franchise continuation that most critics treat as a genuine film rather than as a product.

It is also the most demanding thing recommended here. The film assumes familiarity, moves at its own pace, and does not resolve neatly, which is why it tops critics' lists and does not top box office ones.

Watch it when you have the attention to give it. Watched distracted it will read as slow; watched properly it is the strongest piece of filmmaking in this piece after Backrooms.

Crime 101: the well-made star vehicle

Chris Hemsworth as a jewel thief, Mark Ruffalo as the detective pursuing him, Halle Berry as an insurance broker. The shape is entirely familiar and the execution is the reason to watch.

Films like this live or die on whether the procedural half is as interesting as the criminal half, and the casting suggests the film knows that. A pursuit story where you want the pursuer to succeed is a much better film than one where the police exist to be evaded.

The most straightforwardly satisfying thriller on the list. Nothing about it is surprising and everything about it is competent, which is a genuine recommendation and not a backhanded one.

Apex: the chase in the open

Charlize Theron as an extreme sports athlete, Taron Egerton as the psychopath hunting her. A survival thriller that puts its two leads in open terrain and lets the geography do the work.

The sub-genre is unforgiving. A hunt film has almost no plot to hide behind, so the tension has to come from staging — sightlines, distance, what each character knows about where the other is — and the casting indicates a film that intends to be physical rather than psychological.

The pick for a viewer who wants momentum. It is the least reflective film recommended here and the one least likely to lose you.

Dead Man's Wire: Pacino, and a real story

Gus Van Sant's account of a 1970s Indianapolis hostage standoff, with what several reviewers describe as Al Pacino's best work in years. That phrase gets used loosely and appears to be meant here.

Period hostage dramas work when they resist the temptation to editorialise from the present, and Van Sant is a director with the patience for that. The interest is in how the situation was handled at the time rather than in what it proves.

The most adult film on this list, in the sense of asking most and shouting least. It is also the one most likely to be overlooked in a year dominated by horror.

Is God Is and How to Make a Killing: where the genres met

The year's most interesting films were frequently the ones that refused a single genre. Is God Is is a Southern Gothic revenge thriller with a genuinely comic edge; How to Make a Killing is a dark comedy that functions as a thriller rather than borrowing its furniture.

This blending is why the ranked lists disagree so much in 2026. A film that is 60% thriller and 40% comedy appears in both sets of rankings at different positions, and neither placement describes it accurately.

Both are worth watching and neither is a safe recommendation. They are the entries most likely to be someone's film of the year and someone else's abandoned evening.

What the year's economics mean for next year

A million-dollar film returning $240 million and a badly reviewed sequel returning $214 million teach the industry two different lessons, and it will act on both. Expect substantially more low-budget horror, and expect the franchise sequels to continue regardless of reception.

The first is good news for viewers. Cheap horror is the part of the industry where new directors get made, and a commercially proven route to a first feature produces more interesting films within two or three years.

The second is not. A sequel that earns its highest gross with its worst reviews has demonstrated that quality is optional for that particular audience, and studios do not forget demonstrations like that.

How to actually pick, in ninety seconds

Decide the register before you look at anything: do you want dread, shocks, tension, or relief. Those four words separate almost this entire list, and getting the register wrong is the cause of nearly every disappointing evening.

Then decide the attention level. Backrooms, The Bone Temple and Dead Man's Wire reward full attention and punish distraction. Send Help, Crime 101 and Hoppers do not.

Then check one thing about content rather than quality — how much unpleasantness you will accept. Saccharine and The Bride! sit at one end, Hoppers and the Prada sequel at the other, and the aggregate scores tell you nothing about which is which.

A short list for the reader who wants one film

For dread: Backrooms. For a night with other people: Send Help. For laughing without effort: the Devil Wears Prada sequel. For something that stays with you: Dead Man's Wire. For the one everyone will have seen: Obsession.

Those five cover the year adequately and disagree with every ranked list published, because ranked lists are answering a different question — which film is best — rather than the one a viewer actually has, which is what should I watch tonight.

The rest of this piece exists for the reader who wants more than five. Any of the films above is a defensible evening; the ordering between them is a matter of what you are in the mood for and nothing else.

The performances that carried the year

Three turns account for most of the acting conversation in these genres this year, and all three are in films that would not usually generate one. Jessie Buckley in The Bride!, described repeatedly as a force of nature, is the showiest. Al Pacino in Dead Man's Wire is the one critics reach for the phrase best in years about, and appear to mean.

The third is Inde Navarrette in Obsession, which is the interesting case because the film around her is a genre picture made for a million dollars. A performance that survives that context is doing something the budget did not pay for, and it is the reason she has spent the summer being cast in much larger things.

The pattern is worth noting for anyone who follows casting. Horror and thriller have become the reliable route by which actors demonstrate range at a scale studios notice, which was not true a decade ago and is now the standard path.

Hokum and the rest of the shortlist

Variety's ranked list of the year places Hokum alongside Obsession and Backrooms, and the films below the top three are where the year's depth actually shows. A strong year is not one with three good films; it is one where the eighth-ranked entry is still worth an evening.

That is the test 2026 passes and most years fail. The films outside the top of these lists are not filler — they are smaller, stranger, and frequently more interesting than the ones that broke through commercially.

For a viewer this is the practical argument for reading past the first three entries of any list. The top is decided by consensus, which selects for films that offend nobody; positions four through ten are where the films that someone loves tend to sit.

Why horror keeps producing the year's best cheap films

The economics are unusually favourable and have been for a decade. Horror is the one genre where audiences do not require stars, expensive locations or visual effects to accept a film as legitimate, which means a first-time director can make something that competes with studio output on its own terms.

Obsession is the extreme case rather than an anomaly. A million-dollar budget returning more than two hundred million is unusual in scale and entirely conventional in shape: the genre produces a version of this story every few years, which is why it remains the most reliable entry point into the industry.

The consequence for viewers is that horror is where you find directors before anyone else does. Several of the names on this year's list will be making expensive films within three years, and their first features are usually better than what follows.

What the release calendar still has coming

This piece covers the year to date, and the awards-season window has not opened. Historically the strongest thrillers and the most serious genre work arrive in the final quarter, which means any list published in August is describing two-thirds of a year.

That is a reason to treat current rankings as provisional rather than a reason to ignore them. The films discussed here have been released and can be watched now; the ones arriving later cannot, and a watchlist that waits for completeness is a watchlist nobody uses.

The practical approach is to work through the year to date now and revisit in January, when the December releases have been seen and the summer films have stopped benefiting from recency. The two lists will differ more than most readers expect.

A note on watching horror well

More than any other genre, horror depends on the conditions it is watched in, and most of the disappointment people report is a viewing problem rather than a film problem. Dread-based films — Backrooms and Cold Storage in particular — do not survive a lit room, a second screen or a paused midpoint.

The mechanism is straightforward. Tension is cumulative and any interruption resets it, which is why a film that unnerved a cinema audience can seem inert at home. Shock-based horror survives this treatment; atmospheric horror does not.

The recommendation is therefore conditional in a way lists rarely admit. If the viewing conditions are compromised, choose Send Help or Scream 7 rather than the better films, because a well-made atmospheric horror watched badly is a worse evening than a competent one watched properly.

Comedy is the genre lists are worst at

Every aggregation problem in film criticism is worse in comedy. A horror film that fails to frighten has failed at something measurable; a comedy that fails to amuse has failed at something that varies by person, by mood and by the number of people in the room.

This is why the comedy entries here are described by register rather than ranked. Hoppers is broad and safe; the Devil Wears Prada sequel is verbal and character-driven; I Love Boosters is political and abrasive. Those three cannot be ordered, and any list that orders them is expressing a preference and calling it a judgement.

The one aggregate signal worth trusting in comedy is unanimity. When critics with different tastes agree a film is funny, as they largely have about Hoppers, that survives the subjectivity problem better than a high average built from wildly split scores.

The thriller revival nobody announced

Thriller had been the quiet genre for several years, squeezed between franchise action above it and prestige drama beside it. 2026 reversed that without any of the coverage that accompanied horror's year.

The evidence is in the range: a franchise continuation taken seriously as filmmaking, a star-led heist picture, a two-hander survival film and a period hostage drama, all released within months and all landing. That is a functioning genre rather than a run of luck.

The likely cause is the same economics driving horror. Thrillers are cheaper than action films, they carry stars well, and they perform on streaming afterwards — which makes them the format studios reach for when budgets tighten, as they have.

What to watch if you only have three evenings

One from each genre, chosen for reliability rather than for critical position. Obsession for horror, because it is short, well constructed and the film everything else this year is measured against. The Devil Wears Prada sequel for comedy, because it is the least likely of the three to misfire in a mixed room.

Crime 101 for thriller, on the same logic — familiar shape, competent execution, no demands on the viewer beyond attention. None of these three is the best film in its genre this year, and all three are the safest.

If you want the best rather than the safest, substitute Backrooms, I Love Boosters and The Bone Temple respectively, and give each of them an undistracted evening. That is a materially better set of films and a materially higher chance of one of them not working for you.

A viewing order that works

Pick the register first, the film second, and ignore the overall ranking entirely.

Start with Obsession if you have not seen it, because it is the film the rest of the year is reacting to and it is short. Follow it with Backrooms when you can give it full attention, which will recalibrate what you expect from the genre. Take Send Help when there are people in the room and the Devil Wears Prada sequel when there are not. Save The Bone Temple and Dead Man's Wire for the evenings you actually want to concentrate, and treat Saccharine as optional in both directions.

Keep your own record, not the aggregate

Note which register each film you enjoyed belonged to rather than the score you would give it. After a dozen entries the pattern is usually clear and frequently surprising — most people discover they consistently prefer tension to shocks, or comedy that argues to comedy that soothes, and neither preference is visible in any published ranking. That record makes next year's lists far more useful, because you will know which reviewers to read and which to discount.

Revisit at the end of the year, when the awards-season releases have arrived and the summer films have stopped being over-rated by recency. A list written in August and a list written in December disagree substantially, and the December one is generally the better guide to what to watch in the years afterwards.

Editorial conclusion

2026 was a strong year to be a viewer for a reason that has nothing to do with its biggest films: it produced good work in a lot of different registers at once. Horror got a micro-budget phenomenon, an art-house body horror and the most technically accomplished film of the year. Comedy got range rather than a single hit. Thriller got a franchise entry that was genuinely good and a period drama nobody will talk about enough. The ranked lists will keep arguing about the order. The more useful question was always which of them you were in the mood for.