Why rankings disagree
A ranking answers a question, and different rankings ask different questions. Some ask "who is strongest right now?", others "who has achieved the most this season?", and others "who deserves an invitation to the next event?". The inputs differ too: some count only match results, some weight event prestige, some decay old results quickly and others slowly. When two lists disagree, it is usually because they are measuring different things — not because one of them is wrong.
Six ranking systems you will encounter
1. Elo
Developed for chess by Arpad Elo, the Elo system gives each competitor a rating that rises after a win and falls after a loss. Beating a higher-rated opponent earns more points than beating a weaker one. It is simple and transparent but assumes ratings are equally certain for everyone.
2. Glicko and Glicko-2
Created by Mark Glickman, Glicko adds a "rating deviation" that measures uncertainty. A team that has not played for a while has a wider deviation, so its rating moves more quickly when it returns. Glicko-2 also tracks volatility, capturing how consistent a competitor's performances are.
3. TrueSkill
Developed by Microsoft Research for matchmaking on Xbox Live, TrueSkill handles team games and multiplayer matches with more than two sides — useful for formats like battle royale lobbies.
4. Circuit points
Many publisher-run circuits award points for placements at designated events. Points decide qualification for finals. They reward achievement and attendance rather than pure strength, and they are easy for viewers to follow.
5. Valve Regional Standings
For Counter-Strike 2, Valve publishes Regional Standings that tournament organisers use when issuing invitations. The model considers factors such as results, the strength of opponents and prize earnings, and Valve publishes the methodology openly.
6. Editorial power rankings
Written by journalists or analysts, power rankings mix results with judgement about form, roster changes and playstyle. Long-running examples include the team ranking published by HLTV for Counter-Strike, which combines a points-based model with recent form.
Side-by-side comparison
| System | Main input | Strength | Weakness |
|---|---|---|---|
| Elo | Head-to-head results | Simple and transparent | Ignores uncertainty and inactivity |
| Glicko-2 | Results plus uncertainty | Handles irregular schedules | Harder to explain to viewers |
| TrueSkill | Team and multi-sided results | Suits large lobbies | Designed for matchmaking, not prestige |
| Circuit points | Event placements | Clear qualification paths | Rewards attendance over strength |
| Regional Standings | Results, opponents, earnings | Open methodology tied to invites | Weighting choices invite debate |
| Power rankings | Results plus expert judgement | Captures context and roster changes | Subjective; can reflect bias |
In-game ladders vs professional rankings
The rank badge you see in your own games is driven by matchmaking rating, or MMR, a hidden number often based on Elo- or Glicko-style maths. Its job is to create fair matches, not to crown champions. Professional rankings, by contrast, care about prestige, opposition quality and event tier. A pro player's high ladder rank says something about mechanics, but very little about how their team performs in a structured tournament.
Four pitfalls when reading rankings
- Regional isolation. Teams that rarely face international opponents can inflate ratings within their own region.
- Small samples. A short winning streak can move a rating dramatically early in a season.
- Roster changes. A ranking attached to a team name may describe a lineup that no longer exists.
- Time decay. Check how quickly old results lose weight; slow decay favours past champions, fast decay favours hot streaks.
Algorithmic rankings
- Consistent and reproducible.
- Free from personal bias.
- Slow to react to roster news.
Editorial rankings
- React quickly to context.
- Explain reasoning in words.
- Open to debate and bias.
Which ranking should you trust?
Use more than one. Pair an algorithmic rating for consistency with an editorial ranking for context, and always check the published methodology. When a team climbs, ask which input moved: results, opposition strength or simple activity. For how rankings feed into seeding, read our tournament coverage guide.