How ranking works

No one can pay for a better spot

We never take money to list a group or to move it up. Nothing is sponsored. The rules below are the only rules.

Browse

Browse sorts groups from A to Z by name, or by most recent sign of life if you pick that. When you search, the best name matches come first. Filters only hide groups. They never change a group's place in the list.

The match quiz

The quiz uses plain rules, so every result can be explained. Everything runs in your browser, and your answers never leave it.

Step 1: what is ruled out

A group is never shown if any of these is true:

• It looks closed or has gone quiet for years, or it is a partisan group.
• It is not open to you, such as a group only for students at another school, a members only group, or one with a minimum age you do not meet.
• It is a support group, unless you asked for support groups.
• It is a faith community and you chose to leave those out.
• It breaks an answer you locked, such as cost, schedule, distance, access or background checks. If a group's page does not say, we do not rule it out. We show it after the groups we could check, and say what is not listed.
• It does not fit the guide you are using. For court ordered service, only groups with a public source that says they accept those hours are shown.

Step 2: how well it fits

Each group that is left gets a score from these parts. When a group's page does not say something, that part counts as neutral, not as a no. The weights are starting values, and we tune them with practice runs. Every change is written down in the project on GitHub.

• What you are into (25 percent): the topics you picked and the moments you liked.
• Why you want to join (20 percent): whether the group offers the reasons that matter most to you.
• What you would do there (15 percent): hands on, figuring things out, making, helping, leading or organizing, how you like to be with people, and who is in the room.
• Practical fit (15 percent): your schedule, how far it is, how often, group size and cost, for the things you did not lock.
• Your taste test (10 percent): groups like the ones you liked, and not like the ones you passed on.
• Welcome for newcomers (10 percent): rises to 20 percent if you said meeting strangers is hard.
• A regular place (3 percent): groups that meet on a steady schedule in one place build friendships.
• How well we know it (2 percent): how closely we checked it and how recent its last sign of life is.

Step 3: variety

We pick results one at a time. Each time, a group gets a small penalty if we already picked one in the same interest or from the same organization, and no more than two results can share one. Eight results should not be eight running clubs.

Step 4: we explain

Each result says why it fits, in your own words: the answers that mattered most. No mystery scores.

Stretches and the wildcard

Most results are close fits. By default, two are stretches and one is a wildcard. A stretch changes exactly one thing and keeps the rest familiar:

• One step sideways: a neighboring interest, in the same style and with the same kind of crowd.
• Same thing, new way in: the same interest, but a new way to take part, such as teaching it instead of learning it.
• New crowd: the same interest in a group that mixes ages or neighborhoods.
• Next rung: a bigger role in something you already do.

We never stretch cost, schedule, distance, access, age, safety or clearances. We never stretch to support groups. Faith communities are stretches only for people who said yes to them. Every stretch is labeled and says why.

The wildcard is something you would probably never pick: very welcoming, free or cheap, easy to try once, and tied to something you said. A Mummers string band's open rehearsal is the kind of thing we mean.

The adventure dial

You can turn the dial at any time. Gentle shows seven close fits and one stretch. Balanced (the default) shows five close fits, two stretches and one wildcard. Bold shows four close fits, three stretches and one wildcard.