TMT METHODOLOGY · V1.0 FOUNDATION

How We Rate Movie Trailers

We evaluate a trailer as a creative work and an invitation to see a movie. Each rating belongs to one specific trailer version.

Three different measures

TMT Rating
A structured editorial evaluation of trailer quality, using the six criteria below.
Trailer Score
The site’s quantitative YouTube performance measure. Existing data-ranked collections continue to use this score.
Audience Ranking
A future aggregate of head-to-head audience preferences. It has not launched.

These measures are separate. We never combine performance, editorial judgment and audience preference into one number.

Six criteria, one transparent calculation

15%

Hook & Engagement

How effectively the opening captures attention and the trailer sustains interest.

20%

Editing & Construction

Structure, pacing, shot selection, transitions, escalation and reveals. Assess the assembly here; assess the contribution of sound separately.

20%

Music & Audio

How music, sound design, dialogue, synchronization and silence support storytelling, emotion and impact. A favorite song earns no automatic credit.

20%

Persuasive Effectiveness

How clearly the trailer communicates its intended appeal and creates interest in seeing the film, judged on the trailer's own promise.

15%

Memorability

The lasting impression of this trailer's images, lines, sounds or ideas after viewing, separate from familiarity with the film.

10%

Originality & Historical Significance

How distinctive its approach was at release. Observable originality can be assessed directly; claims of influence or historical significance require credible evidence.

Each component receives a score from 1.0 to 10.0 and a written justification. Multiply each score by its weight and add the results. We retain all six component scores, calculate with full precision and display the overall TMT Rating to one decimal place. Missing scores remain unrated.

What the numbers mean

1–2: seriously ineffective. 3–4: weaknesses outweigh strengths. 5: competent. 6: good. 7: very good. 8: excellent. 9: exceptional. 10: extraordinary execution that serves as a benchmark. These anchors apply to each criterion; a 10 does not require a famous movie or a historically influential campaign.

The weights are fixed for v1.0. We are calibrating the scoring anchors against a candidate set before publishing component ratings.

Judge the trailer in its time

Consider the creative conventions, available techniques and intended audience at the trailer’s original release. Do not penalize older work merely for looking or sounding older, and do not award nostalgia points. A later YouTube upload date is not evidence of the trailer’s original release date.

Observable originality can be evaluated directly. Claims that a trailer pioneered a technique, influenced later campaigns or changed the medium require credible sources that support the specific claim. We record the source, publisher, supporting notes and uncertainty. Missing historical evidence earns neither invented influence credit nor an automatic penalty for originality.

What never goes into a TMT Rating

Movie quality, critic scores, awards, box office, YouTube views, current popularity, franchise fame and personal genre preference are excluded. Persuasion concerns the trailer’s promise, not whether the finished movie fulfilled it. We assess craft within the trailer’s intended genre and audience.

Which videos qualify?

Official theatrical and streaming trailers are eligible. Teasers qualify only where the collection explicitly permits them. TV spots, clips, featurettes, fan trailers, reactions, promos and unofficial edits are excluded.

We record the exact video, version and official source/channel. Another cut, international version or trailer number is a separate creative work. Movie year and genres support collection membership; the original trailer year supplies historical context. Uncertain identity or eligibility stays under review.

From candidate to published rating

Candidate → reviewed → rated → QA approved. Reviewers watch the exact version, record observations and justify each component independently of existing Trailer Scores and list positions. Internal confidence records flag uncertainty. QA checks identity, eligibility, evidence, arithmetic and consistency before a rating can be published.

One evaluation can support decade, genre, all-time and editorial collections without repeating research. Most-viewed collections still order by verified views; an editorial rating cannot stand in for view counts. Collection membership and the measure used to order it are distinct decisions.

Future head-to-head comparisons can test close editorial judgments and collect audience preferences separately. They will not silently change TMT Ratings.

Where we are now

The existing 2000s and 2010s collections are data-ranked by Trailer Score. Their trailers form our initial calibration pool, not a predetermined editorial order. No TMT component ratings have been published. Any future editorial ranking will be explicitly labeled and introduced after calibration.