Before head-to-head match results are applied, every team receives a seeded starting rank value between 400 and 2000. This seed is derived from four performance factors that capture a team’s historical achievements — prize winnings, quality of opponents beaten, breadth of opponents beaten, and LAN performance. Seeding ensures that teams enter the Glicko rating phase with a rank that reflects their real-world standing, even before any head-to-head adjustments are made.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/ValveSoftware/counter-strike_regional_standings/llms.txt
Use this file to discover all available pages before exploring further.
Seeding factors
Four factors feed into the seed calculation. Each is a normalized value between 0 and 1, computed from the team’s match history. TheownNetwork factor exists in the codebase but currently carries a coefficient of 0, so it has no effect on the seed.
| Factor | Coefficient | What it measures |
|---|---|---|
bountyCollected | 1 | Quality of opponents beaten (weighted by their prize winnings) |
bountyOffered | 1 | The team’s own prize winnings |
opponentNetwork | 1 | Breadth of quality opponents beaten |
ownNetwork | 0 | Number of distinct opponents defeated (unused) |
lanFactor | 1 | LAN event wins |
Seed value formula
The seed value for a team is the weighted average of its active factors:Mapping seed values to rank range
Once every team’s raw seed value is computed, the values are linearly remapped from the observed minimum–maximum range into the fixed rank window of 400–2000:Seeding in context
Calculate per-team factors
For each team, compute
bountyCollected, bountyOffered, opponentNetwork, ownNetwork, and lanFactor from match history. Each factor is normalized to [0, 1].Compute weighted seed value
Apply
SEED_MODIFIER_FACTORS coefficients to get a single scalar seed value per team.Remap to rank range
Linearly scale all seed values into the [400, 2000] window. This seeded rank becomes the team’s starting point in the Glicko rating phase.
Only the top 10 results are counted for each factor (controlled by
bucketSize = 10 in team.js). This means playing in low-stakes or low-quality matches never hurts a team’s seed — at worst, a weak result simply doesn’t displace one of the top 10.