Agatha All Along (2024) Review
1/10
A witch stripped of her power is living inside a delusion, is snapped out of it by a teenager who wants something from her, and assembles a coven to walk a road that may not exist.
The 1 is about structure instead of craft. There’s real production value here, a committed cast and a good song, and none of that addresses the problem that the series has no reason to exist beyond a character having been popular.
This one stings more than a bad show should. Kathryn Hahn was the best thing in WandaVision, and Marvel took the goodwill she earned and spent it on nine episodes of filler. The worst of it is how much talent is standing around in these episodes waiting for a story to show up.
The Witches’ Road is the clearest symptom. A quest with a new obstacle each week, no cumulative pressure, and a destination the show keeps refusing to define until the mechanism is revealed as something other than what it claimed.
A Scene-Stealer Cannot Be the Whole Scene
Agatha worked in WandaVision because of what she was doing to somebody else. She was a neighbor with a hidden agenda, present in small doses, and her scenes were good because the audience was watching a performance inside a performance. Promoted to protagonist, she loses both conditions at once. There’s nobody for her to be secretly manipulating, because everybody in the coven knows what she is, and there’s no restraint on the screen time, because it’s her show. Kathryn Hahn isn’t the problem and she can’t fix it. She plays contempt, exhaustion and appetite with real precision, and the character has no want the audience can hold beyond getting her power back. That’s a goal instead of a motive.
The series recognizes this late and supplies a backstory about a son. That arrives in the seventh episode and is asked to retroactively supply the interior life the previous six were missing.
That episode is also the best in the season. That’s the problem. A woman who has spent centuries taking power from covens is shown doing it once, to people who trusted her, for a reason, and everything she does in the six episodes before it would have meant something had the audience known.
A supporting character’s appeal usually comes from their relationship to a protagonist and from limited exposure, and promoting them removes both conditions at the same time.
Before you spin somebody off, name what they want that’s nothing to do with the original story. If the answer is that they want to be restored to the position they held before, you have a goal without a motive. The reliable version gives the character a new antagonist and a new stake. What doesn’t work is transplanting the same person into the lead slot and hoping the qualities travel.
The Road Has No Pressure
Each episode presents a trial, the coven attempts it, somebody fails, and the group proceeds. The structure repeats without escalation.
Nothing accumulates because nothing carries forward. A trial that’s survived costs the characters no resource they need later, and the deaths that occur don’t change what remains possible for the group. The show also withholds the nature of the road for too long. Ambiguity works when the reader can see the shape of the question, and here the audience is asked to invest in a journey without knowing what arrival would mean.
When the mechanism is revealed the effect is deflating. The road wasn’t what it appeared, and that turns six episodes of trials into a staged exercise, and staged exercises have no stakes by definition. I felt cheated by that reveal, and I think the people who wrote it should have known that anyone invested in the quest would feel the same.
Compare it to the first season of Loki. That established its rules in the opening hour and obeyed them for six. Rules stated early let an audience calculate risk. Rules withheld until the end mean the audience has spent the whole work unable to assess anything they were watching.
The Coven Is Underwritten
Six women assemble and the series gives two of them enough material to be people. Patti LuPone’s Lilia is the best of them, a diviner who experiences her own life out of order, and the episode built on that idea is the only hour with an interesting structure in it.
Debra Jo Rupp, Ali Ahn and Sasheer Zamata are all capable performers given single traits and a function in a trial. The series keeps introducing what each of them is afraid of and then resolving it in the same episode. Aubrey Plaza’s Rio is the exception and the show is coy about her for too long. A former lover who turns out to be something else entirely is a good idea delivered so late that the relationship it’s meant to complicate has barely been established.
Plaza is doing something specific with the part and it deserves a better vehicle. She plays somebody who has known Agatha for centuries, finds her tiresome and can’t leave. That’s a recognizable kind of relationship and the only one in the series with any history behind it.
Plaza and LuPone are the only two people here I’d happily follow into a better show, and I’m angry this one had them and did so little with them. Plaza especially is playing a sharper series than the one she’s stuck in.
An ensemble assembled for a quest needs each member to want the destination for a different reason, or they’re equipment with dialogue.
The test is whether removing any one of them changes what the group can attempt. If the answer is only that a trial would be harder, you’ve written a skill set instead of a character.
Introduce a fear and resolve it in the same installment and you’ve produced an episode. Fears that persist across a work are what readers attach to.
The Teen Is the Withheld Card
Joe Locke plays a boy who can’t say his own name, follows Agatha onto the road, and is revealed late to be somebody the audience was meant to have guessed. The withholding is the series’ central miscalculation. A character defined by a secret the show refuses to state has no interior life until the statement arrives, so he spends most of the run as a device with an expression. Locke is good in the material he’s finally given, and the last two episodes contain the relationship the series should have been about. A boy who needs a mother figure and a woman who destroyed the last person who needed that from her.
That’s a real idea and it arrives with ninety minutes left in a nine-episode season, after six hours of trials that did nothing to establish it.
The series was also announced as a single season with no continuation planned. That makes the structural choice harder to defend. A show that knows it’s nine episodes and no more has no reason to hold its premise back until the seventh.
That structural choice loses me for good. A nine-episode season with no second one planned has no excuse for hiding its best idea until the seventh hour, and I don’t buy any argument that the mystery was worth it. Joe Locke spends most of the season with nothing to play, and that’s on the writers.
What Works
The Ballad of the Witches’ Road is a real piece of songwriting, repeated across the season in different arrangements, and it’s the most memorable thing the franchise produced in Phase 5.
The practical effects are also a real choice. The road, the houses and several of the trials are physically built, and after a phase defined by digital environments the difference is immediately visible.
The first episode is a good half hour. A television detective procedural pastiche, committed to in the same way WandaVision committed to its sitcoms, and the series abandons the device once it’s served an introduction.
There’s a version of this show that keeps doing that. Each episode a different genre, the way the first one is a Scandinavian-style murder investigation, with the delusion breaking down a little more each time. The series does it once, well, and then becomes a fantasy quest with no format at all.
I’ll give the song and the practical sets real credit, and the 1 still stands. A show with those elements in it and no reason to be nine episodes long is a show I resent for wasting them. Everything that works here is style, and style can’t substitute for a protagonist with a want and a structure with pressure.
It’s also the clearest case on this page of the phase’s defining habit. A character was popular, so Marvel commissioned a series and built it around the popularity instead of around a story somebody needed to tell. Nine hours later the best idea in the series turns out to be a mother and a son, introduced for the first time in the seventh episode.
Style isn’t structure. A distinctive look, a good song and a committed cast will sustain a first episode and none of them address whether the story has anywhere to go.
The diagnostic is to state your protagonist’s want and the obstacle in one sentence each. If the want is to be restored to a previous condition and the obstacle is a sequence of unrelated tests, the shape is a video game instead of a story. When a work’s best material arrives in the last twenty percent, the structure is upside down. Whatever the seventh episode revealed should have been the first episode’s premise.
Should you watch Agatha All Along?
Watch the first episode and the Lilia episode and skip the rest. I gave this a 1 because a good song, some handsome practical sets and two sharp supporting turns can’t cover for a series with no reason to exist, and the people who made it had their best story sitting right there and buried it under trials. There’s a good ninety minutes in here about a woman who destroys everybody who needs her, spread across nine episodes of a quest that turns out not to have been a quest.
