About
Most Rewatchable
An unofficial, fan-built index of The Rewatchables—The Ringer’s podcast where Bill Simmons, CR, and a rotating cast rank the movies you can’t stop coming back to.
The catalog is built from the podcast’s RSS feed, TMDB, OMDb, and MDBList—structured data about the show that isn’t published anywhere else.
Spotted a missing episode or a bad data point? Email hello@mostrewatchable.com.
Origin
Why this site exists
The whole site traces back to one episode. I had never seen Flight when its episode came out—I listened anyway, liked the hour so much that I watched the movie the next day, and loved it. That flipped how I use the show: an episode isn’t just a companion piece for movies you already love—sometimes it’s the reason you press play. It’s still the bar I rate episodes against on this site. I call it the Flight test: a Pantheon episode is one good enough to cause a watch. And a Pantheon episode about a movie with a modest score isn’t a contradiction—the tier praises the conversation, the score describes the movie.
The same episode planted the other seed: Van Lathan was hilarious on it, and I loved the giddy energy he elicited out of Chris Ryan and their overall dynamic—enough that I wanted a list of every episode the two of them did together—and, while I was at it, every episode Van Lathan was on at all. That list didn’t exist anywhere. Building the thing that could answer it turned into this site.
The formula
How the score works
Every movie in the catalog gets a Most Rewatchable Score—one number for how rewatchable it is, on a roughly 0–10 scale. The recipe:
- 40%IMDb rating
- 25%Rotten Tomatoes
- 20%Roger Ebert
- 15%Runtime — shorter is more rewatchable
- +1max bonus when the podcast loves it—my episode ratings and the show’s revisits
A bit deeper
The foundation is public data: the IMDb rating, Rotten Tomatoes, Roger Ebert’s star rating, and the runtime. The Rotten Tomatoes signal averages the critics and the audience—the Tomatometer and the Popcornmeter each speak at half the slot, which is how a critically shrugged-at fan favorite gets its due. Runtime is a scoring signal here because it’s the most on-brand thing a rewatchability metric can measure— “it’s 96 minutes, it’s always on, you just watch it” is half the show’s vocabulary. Shorter is more rewatchable; three-hour epics pay for the commitment they ask of you.
On top of that sit two small boosts from the podcast layer. I rate every episode I’ve listened to on a 1–3 scale—3 is Pantheon (outstanding), 2 is Solid (pretty good), 1 is Standard Rewatch (fine, the baseline). The rating is about the episode, not the movie: did the conversation entertain me, and did it make me want to watch the movie? A Pantheon episode gives its movie a small bump, a Solid episode a smaller one, and a Standard Rewatch adds nothing—it just marks a verified listen. The second boost rewards the show itself coming back: a movie the podcast has revisited for a second or third episode earns a little extra credit.
One rule runs under all of it: never publish a rating that doesn’t exist. If a signal is missing— Ebert died in 2013 and never reviewed most recent movies—its weight is redistributed among the signals that do exist. Episodes I haven’t heard yet stay unrated and add nothing. No defaults, no invented opinions.
And the score is deliberately uncapped: a great movie with every bonus maxed out goes past 10, which feels right for a podcast that treats certain movies as beyond criticism.
For the super nerds…(not expecting anyone to read this)
The exact recipe. The movie score is a weighted average of up to four signals, each normalized to a 0–10 subscore:
| Signal | Weight | Normalization |
|---|---|---|
| IMDb rating | 40% | used as-is (already 0–10) |
| Rotten Tomatoes (Tomatometer + audience) | 25% | midpoint of the two, each compressed to a 5–10 band (5 + x⁄20) |
| Ebert stars | 20% | stars × 2.5 |
| Runtime | 15% | tier subscore (next table) |
Runtime maps to its subscore through six tiers—the first hit is a tiny half-step, every later one a full point. Each tier includes its upper bound, so a movie at exactly 2:00 takes only the tiny hit:
| Runtime | Subscore |
|---|---|
| Under 1:40 | 10.0 |
| 1:40 – 2:00 | 9.5 |
| 2:00 – 2:20 | 8.5 |
| 2:20 – 2:40 | 7.5 |
| 2:40 – 3:00 | 6.5 |
| Over 3:00 | 5.5 |
The Rotten Tomatoes compression (applied to the Tomatometer and the audience score alike) and the runtime tiers live in a 5–10 band on purpose: raw tomatometers span roughly 11–100 across this catalog while IMDb ratings compress into about 5.5–9.3, so an uncompressed signal would dominate the board’s spread and bury critically divisive fan favorites.
Missing signals renormalize. An absent signal’s weight is redistributed proportionally among the present ones. No Ebert review, for example, means IMDb speaks at 50%, Rotten Tomatoes at 31.25%, and runtime at 18.75%. Rotten Tomatoes counts as one slot for this: it’s present if either of its numbers exists—both give the midpoint, one alone takes the whole slot. Nothing is ever imputed—a “silent 2 stars” would fabricate an opinion and systematically punish post-2013 movies.
Thin data gets less trust. Signal coverage is the sum of the base weights of a movie’s present signals. At 80% coverage or better—which includes the common missing-Ebert case—the score is fully trusted. Below that, it’s pulled toward 8.0 (the catalog’s typical movie) in proportion to the missing trust: trust = min(1, coverage / 0.8). A score built from almost nothing shouldn’t rank like a conviction, in either direction.
The boosts are quarter-point currency. The episode boost takes the best rating across a movie’s rated episodes: a 3 adds +0.50, a 2 adds +0.25, a 1 adds nothing. Best-of, not a sum—a lower-rated revisit never drags anything down. The revisit boost is +0.25 for each episode beyond the first, capped at +0.50, and it’s deliberately independent of my ratings: episode count is objective data. The symmetry is intentional—the show coming back for a revisit is worth the same quarter point as me bumping an episode to a Solid. Each boost caps at half a point, so the maximum total boost is +1.0. There is no cap on the final score.
A worked example with everything firing:
| IMDb rating | 8.3 | × 40% | 3.32 |
| Rotten Tomatoes (Tomatometer + audience) | 88% / 94% → 9.55 | × 25% | 2.39 |
| Ebert stars | 3.5★ → 8.75 | × 20% | 1.75 |
| Runtime | 170 min → 6.5 | × 15% | 0.98 |
| Movie score | 8.43 | ||
| Episode boost (a Pantheon episode) | +0.50 | ||
| Revisit boost (3 episodes) | +0.50 | ||
| Most Rewatchable Score | 9.43 | ||
Two more properties worth knowing: a movie with no Ebert review is scored by the same arithmetic with the renormalized weights above, and a movie whose only rated episode got a 1 ranks exactly as if it were unrated—the 1 shows up on the episode, never in the math. Every score on the site recomputes from its inputs at query time; hover or tap any score in the catalog to see this exact arithmetic for that movie.
Not affiliated with The Ringer, Spotify, or the hosts of The Rewatchables. This product uses the TMDB API but is not endorsed or certified by TMDB.