How we score a studio.
Every studio in this index carries a BVS Score from 0 to 10. It is computed from catalog data, not voted on and not negotiated. This page is the whole formula, so any number on this site can be checked.
The four axes
A studio's score is a weighted blend of four measures. Each one is percentile-ranked against every other scored studio, so a value answers "compared to the rest of the industry" rather than "compared to an arbitrary target".
| Axis | Weight | What it measures |
|---|---|---|
| Catalog depth | 30% | How many releases the studio has produced, on a log scale so the difference between 20 and 40 counts for more than the difference between 900 and 920. |
| Production quality | 30% | Average resolution across the catalog, plus the share of releases at 6K and above. One 8K scene in a 4K library does not make an 8K studio. |
| Activity | 25% | How recently the studio released, how much it released in the last year, and how many separate months it was active across the last 24. |
| Roster variety | 15% | Distinct performers per release. A rate, not a count; otherwise it would just restate catalog size. |
Small catalogs are pulled toward the average
A studio with four releases has not proven anything yet, in either direction. Each score is blended with the index average in proportion to how much evidence exists, so a small catalog sits near the middle until it grows. A studio with hundreds of releases is judged almost entirely on its own record.
Dates are measured against the newest release on file
Recency is calculated from the most recent release anywhere in the index, not from today's date. If our catalog ingestion pauses, every studio would otherwise appear to go stale at once. Anchoring to the data keeps the comparison between studios honest.
What we deliberately left out
Popularity. How many people read a studio's profile here tracks its catalog size almost perfectly; including it would count catalog depth twice and turn the score into a measure of our own traffic. Reader interest is published separately, on the rankings page, and never touches the score.
Tag counts. The number of distinct tags across a catalog also tracks its size, so it says nothing about range.
User votes. There is no user rating on this site. A star average assembled from a handful of anonymous clicks would be less trustworthy than the catalog record, not more.
Commercial terms. Pricing and subscription details change faster than any review cycle can track, and are not part of the score.
When a measure is missing
If a studio has no performer credits on file, the roster axis is dropped and the remaining weights are rebalanced. It is scored on what is known, never penalised for a gap in our data. The studio page shows n/a in that case rather than a zero.
Bands
Scores are grouped into five bands. The distribution is genuinely bottom-heavy: roughly half the studios in this industry are dormant back-catalogs, and the score says so.
How often it updates
Scores are recomputed whenever the catalog changes materially, most recently in August 2026. A studio that resumes releasing will see its activity axis recover; one that stops will see it decay.
Editorial judgement, and its limits
Our reviewers may adjust a computed score by at most half a point, and only with a written reason recorded against the review for something the formula cannot see, such as a large catalog that turns out to be re-encodes of much older material. The adjustment is capped precisely so the ranking stays reproducible.
Corrections
Every score can be traced back to the catalog record that produced it. If you represent a studio and believe a figure is wrong, tell us what is wrong and why, and we will re-check the underlying data.