We have no idea how good movies are

Overview

The video argues that film “ratings” (numeric scores, star ratings, and review aggregates from sites like IMDb, Letterboxd, Rotten Tomatoes, and Metacritic) are fundamentally unreliable measures of whether a movie is truly good or bad.

Key Arguments

1) Averages hide emotion and polarization

The speaker claims that very different audience reactions can produce the same average score. For example, a film might receive an identical mean rating whether:

• many viewers are indifferent, or
• half the audience loves it and half hates it.

As a result, score averages often fail to reveal whether a film is merely “meh” or genuinely divisive.

2) Scoring systems flatten important differences

The video critiques how platforms convert reviews into simplified metrics:

• Rotten Tomatoes: criticized for reducing reviews to a binary “fresh/rotten” signal. This makes mild positives and major praise look the same, and mild negatives and strong dislike look the same.
• Metacritic: described as somewhat better because it translates reviews into numeric values and applies weighting, but it still obscures how opinions are distributed across the audience.

3) Distribution patterns don’t match “polarization” expectations

The creator claims they compared “polarizing” films versus “consensus” films using Letterboxd histograms and repeatedly found bell-shaped distributions, not the expected U-shaped pattern for divisive movies.

This undermines the idea that polarization should consistently show a clear statistical signature in those datasets.

4) Ratings may reflect an “insider” subculture, not mass audiences

The video expands into selection bias:

• Most people don’t rate movies or participate online; only a motivated minority does.
• Different platforms have different user cultures and rating behaviors—specifically, how people choose to use the rating scale—making comparisons across sites less reliable.

5) Social and political dynamics distort what ratings represent

The speaker argues that online discourse can be shaped by forces unrelated to the film’s quality, including:

• social media hype/backlash
• identity and clicks
• politics

In this framing, ratings become social signals or identity markers rather than judgments of the movie itself.

The video references a “90-9-1” style idea: most people observe silently, while a vocal minority drives visible discourse.

6) Even “real audience” measures capture bias

Even measures like CinemaScore/PostTrak are presented as closer to ordinary viewers because they poll immediately after screenings, but they mostly capture opening-night excitement, not long-term impact.

7) Box office is not a reliable “good/bad” metric

The video also criticizes simplistic assumptions about commercial success and profitability. It claims:

• Hollywood accounting and conventional profit logic can be “flawed.”
• High box office may reflect momentum and word-of-mouth, but crowds can still be “wrong.”
• Box office success does not guarantee lasting classic status.

Examples referenced include:

• Avatar (as a case where crowds can be fallible)
• F1 (hypothetically, a “box office supernova” that still might not be artistically great)
• Oppenheimer (financially successful, possibly not a lasting classic)
• Moana (confusing signals: low ticket sales but high scores and mixed reviews)

Proposed Alternative: Measure “Durability” Over Time

The core suggestion is to focus on durability rather than immediate aggregated numbers. The video uses a “coastal erosion” metaphor:

• Mass culture is the promontory with mixed content.
• Cultural pressure and scrutiny are the sea/waves.
• “Classics” are the pillars that remain after erosion—films culture can’t fully wear down.

It also notes that IMDb ratings can drop over time as initial enthusiasm erodes.

Retrospective Reassessment Example: The Shining

The video uses The Shining to illustrate how verdicts change with time:

• Initially: met with indifference/hostility
  • commercially outperformed other films at the time
  • nominated for Razzie awards
  • Siskel described it as boring/embarrassing
• Later: major reassessment
  • Roger Ebert later included it among great films with a much higher retrospective score
  • Even the Razzie nomination related to Shelley Duvall was withdrawn years later

Conclusion

The speaker urges viewers to stop treating ratings and letter grades as definitive verdicts. Instead, time and cultural endurance are presented as a better indicator of genuine quality—because a film’s real “goodness” is revealed through lasting resilience rather than immediate aggregated scores.

Presenters / Contributors

• No clear additional contributors are named in the subtitles.
• The video references Gene Siskel and Roger Ebert, and actress Shelley Duvall, but they are referenced as part of the examples, not as presenters of the video.