How the Trust Score Works
The public Trust Score is a simple number, but HYFY treats the evidence behind it carefully. The goal is to show current recognition without letting one person, one trait, or one old rating define a profile.
From Rating to Trust Score
1. A rating is attached to a trait
The reviewer chooses a quality and submits the rating allowed for that profile type. People use Chill (3), Fire (4), and Elite (5). Brands, pages, and places use the conventional 1-to-5 scale.
2. HYFY checks eligibility
The platform checks who is rating, which profile is being rated, and whether the chosen trait belongs to that type of profile. People ratings also require a mutual follow. Self-ratings and ratings between profiles controlled by the same account are blocked.
3. One current opinion counts
For each reviewer, person, and trait, only the latest active opinion contributes to the score. Updating a rating replaces its influence; it does not create an extra vote.
4. Evidence is treated with care
Early scores are deliberately cautious. As more independent reviewers contribute, the score can reflect their shared pattern more fully.
HYFY also gives a small, tightly limited amount of extra confidence to evidence from verified or longer-standing accounts. This is not based on follower count, popularity, the reviewer's own Trust Score, or whether they agree with everyone else.
Older ratings gradually matter less, so the score is not permanently controlled by recognition from years ago.
For example, one fresh Elite endorsement on a people trait does not immediately show as a perfect 5.00. The first score is cautious; separate support from more people allows the result to reflect the rating pattern more fully.
5. Traits become an overall score
Trait ratings are combined within their relevant categories, and the represented categories are then brought together into the overall Trust Score. This helps keep a large number of ratings in one category from drowning out every other part of a person's profile.
6. The score is shown with context
HYFY displays the numeric Trust Score together with the number of different raters, rated traits, and current ratings. The counts help you distinguish a newly rated profile from one supported by broader recognition.
| Count | What is counted |
|---|---|
| Hypers | Different people whose current ratings contribute |
| Traits | Different qualities that currently have ratings |
| Ratings | Current reviewer-and-trait ratings in total |
One person can rate several traits. In that case they count once under Hypers, while each trait opinion counts under Ratings.
What the Score Does Not Mean
The Trust Score does not prove someone's identity, guarantee future behavior, or verify that every opinion is correct. It summarizes eligible recognition using the evidence currently available. The score and its supporting counts should be considered together.
Design Principles
| Principle | What it means for users |
|---|---|
| Recognition-first | People can only choose Chill, Fire, or Elite |
| Trait-based | Ratings describe specific qualities, not a blanket judgment |
| Independent support | More different raters provide stronger evidence |
| Current | Older evidence gradually carries less influence |
| Bounded influence | No reviewer receives unlimited voting power |
| Transparent context | The score is shown with raters, traits, and ratings counts |
The scoring details may evolve as HYFY learns from real usage, but these public promises should remain stable: positive-only people ratings, cautious early scores, time awareness, bounded reviewer influence, and visible supporting counts.