How The Skry scores

The five dimensions

Every review scores five dimensions on a ten-point scale, to one decimal. They are the whole of the editorial judgment; there is no sixth, unnamed factor. A review that has not scored all five does not produce an Editorial Score at all.

SCRIPTStory

Structure, dialogue, ideas, pacing — what is on the page. SCRIPT asks whether the thing was written well: whether the story holds its shape, whether the scenes earn their length, whether the ideas underneath survive contact with the plot on top of them.

LENSCraft

Direction, cinematography, editing, sound, design — what is on the screen. LENS is the made object: how it was shot and cut and scored, whether the images are doing work or merely covering the dialogue, whether the craft is in service of anything.

CASTPerformance

Performances, chemistry, casting choices — who is on screen. CAST covers the leads and the people around them, whether the ensemble coheres, and whether the casting itself was an act of judgment or a matter of availability.

PULSEImpact

Emotional charge, urgency, force — does it land. PULSE is the dimension a technically accomplished film can fail. It measures whether the work reaches you at all, and how hard, and whether the reach was earned or merely engineered.

REPLAYRewatch

Rewatchability and the case for going back — does it last. REPLAY asks what a second viewing gives you: whether the work has interior worth returning to, or whether knowing the ending empties it out. Some excellent films score low here, and that is a real result, not a penalty.

The Editorial Score

The Editorial Score is the flat mean of those five numbers. Each dimension contributes exactly one fifth — 20% — regardless of genre, format, or anything else. A reader who sees SCRIPT 9.0, LENS 8.0, CAST 9.0, PULSE 8.5, and REPLAY 7.5 can add them up and confirm the 8.4 without taking our word for it.

Genre matters to how a dimension is scored, never to how much it counts. Our internal genre and era guidance teaches what to look for when scoring horror craft or a silent-era performance; it never multiplies a dimension by anything. Weighted-by-genre compositing would reintroduce exactly the opacity we are positioned against.

The no-smoothing rule

Each dimension scores what it scores. The five do not pull toward each other, and none of them pulls toward the average. A film that is extraordinary in one dimension and weak in another should show that spread plainly — CAST 9.4 against SCRIPT 7.0 tells you something specific and true about a film, and flattening it toward a comfortable middle would destroy the only information the profile carries.

The same rule governs the three layers below. They are blended, not reconciled. When our editorial read is a 9.0 and the aggregators average 7.4, the composite reflects that gap honestly and the prose explains it. A number that moves in either direction has to be defended in writing.

The Skry Score

The Skry Score is the cobalt figure at the top of a review: a weighted blend of three layers.

The editorial layer is the Editorial Score above. The platform layer is the mean of up to three recorded aggregator scores — Rotten Tomatoes critics, Metacritic, IMDb and similar — normalized to the same ten-point scale. The reader layer is the aggregate of verified on-site reader ratings, which are submitted from 1.0 to 10.0 in steps of 0.5.

The weights shift as on-site reader signal accumulates. Editorial conviction is most load-bearing when there is little else to go on; readers earn more of the score as more of them show up.

Most titles on The Skry sit in the first row today: editorial 60%, platform 25%, reader 15%.

How reader ratings are handled

On-site ratings begin counting once a title has 5 of them. Below that they are held back, because a handful of ratings is not yet a reader signal. Past the threshold, no single rating can swing the flagship score either, so the reader layer is pulled toward a neutral baseline of 7.0 with a weight equal to 10 ratings. The first ratings barely move it; around 10 real ratings pull it halfway to the true average; past that it converges on what readers actually said.

This adjustment applies to the composite only. The reader average displayed on the page is the raw one — what readers actually submitted, unmodified. Until a title reaches 5 on-site ratings, the layer falls back to the Rotten Tomatoes audience score as a proxy so there is still some reader signal.

When a layer is missing

A layer with no data does not count as a zero. Its weight redistributes proportionally across the layers that do have data. A title with no recorded aggregator scores hands that weight back to the editorial and reader layers in their existing ratio. A title with neither aggregator scores nor any reader signal collapses to the simplest possible case: the Skry Score is the Editorial Score.

The bands

Every Skry Score carries a verdict word. The thresholds are fixed and apply identically to film and television.

Canon-Level is the band for work that defines the form — the films and series a case can be made for in any room, and the ones still worth arguing about decades on. It is not a score we hand out lightly, and it never stands on the number alone: reaching it means the review spends several sentences saying why. Outstanding is reserved for genuine excellence. The discipline runs in both directions: a great comedy scores like a great drama, an animated film scores like a live-action one, and a silent film is measured against any era’s standard rather than graded on a curve for its age.

The editorial floor

The floor for publication is a 7.0 Skry Score. Work below it generally does not get reviewed, because the point of the publication is to tell you what is worth your time rather than to catalogue disappointments. The exception is the title that has to be engaged with — a high-profile failure, a genuine cultural moment — and in that case the low score is the argument, and it gets defended like any other.

No score without a review

The cobalt Skry Score appears only on titles we have actually reviewed. Most of the catalog has no review, and those pages show what they legitimately have — cast, crew, where to watch, the TMDB rating, and aggregator figures where we hold them — with no Skry Score anywhere on the page.

This holds everywhere the score can travel. A share card generated for an unreviewed title carries the brand and no number. The structured data we publish for search engines includes a review rating only when a review exists. There is no synthesized score, no provisional score, and no placeholder that could be mistaken for a verdict we have not made.

Television

Television is scored at three levels, and they are deliberately independent of one another.

A season is the workhorse: a full review of a single season, scored across the same five dimensions as a film, and the unit most TV writing here takes. A series page carries its own score, set as an editorial judgment of the show as a whole — it is not the average of its seasons. The Wire scores what The Wire is worth; the fact that one season scored lower than the others is information for the reader, not an input to the math. Where a series score departs sharply from its seasons, the review says why. An episode review is reserved for landmarks, published sparingly, and assumes spoilers throughout.

Revision

Scores can change, but not casually. A re-evaluation is an editorial decision made when a work has aged in one direction or another, when the context around it has shifted enough to change the reading, or when an anniversary justifies going back. The original score is preserved as part of the record. Reader ratings and aggregator figures, by contrast, move on their own as they accumulate, which means a Skry Score can drift a little over time without anyone rewriting a word.