ELO System Analysis

What is Rating Pool Isolation?

Rating pool isolation occurs when separate groups of players rarely interact, creating rating "bubbles" that are internally consistent but not comparable across groups. A 1600-rated player in one bubble could be genuinely better or worse than a 1600 in another — making that number meaningless for cross-group comparisons like tournament seedings.

Strengths — Why Our System Is Resilient

• 1

  Division-anchored initial ratings

  Every player's rating is seeded from their division placement — not a generic default. This means the rating scale has meaningful real-world anchoring from day one.

  Division	Starting Rating
  Premier League	2000
  1st Division	1800
  2nd Division	1600
  3rd Division	1500
  4th Division	1400
  5th Division	1300
  6th Division	1200
  7th Division	1100

  A Premier player's 2000 and a 5th Division player's 1300 are calibrated to reflect actual competitive tier, not just a product of who they happened to play first. Players with no known division default to 1200.

• 2

  Single unified rating pool across all formats

  Everything feeds one rating_history table — CTTT League (all divisions), Summer League, CTTT Junior League, Super Vets League, Open Tournaments, and Knockout rounds. This is the most important structural advantage. Most club systems keep league and tournament ratings in separate silos. Ours doesn't. A Premier player beating a 5th Division player in a tournament — that exchange is real, recorded, and affects both ratings.

• 3

  Tournaments are the great equaliser

  Open tournaments mix players across all divisions and clubs in a single draw. Round 1 can pit a Premier player against a 5th Division player. These cross-boundary games are the structural anchors that prevent pools from drifting apart. As long as regular open tournaments run with mixed fields, pool isolation stays largely theoretical.

• 4

  The division rating bump system

  When a player gets promoted or relegated, they don't just carry their old rating into a new division — they receive a rating adjustment to better reflect their new competitive context. This actively combats isolation in two ways:

  • It prevents rating deflation in lower divisions — a relegated player doesn't drag others' ratings down by being systematically too strong
  • It prevents rating inflation in higher divisions — a promoted player doesn't inflate others' ratings by being systematically too weak
  • Every division crossing creates a calibrated bridge point between the two pools

  This is genuinely sophisticated — many systems at national level don't do this properly.

• 5

  Inactivity decay

  The decay system (3-month grace → 2 pts/week for months 3–9 → 3 pts/week for 9+, floored at 1100, capped at 25% of division floor) serves a dual purpose most people don't think about:

  • It keeps ratings current and meaningful
  • It actively fights pool inflation — stale high ratings from long-absent players stop anchoring the pool at an inflated level
  • When an inactive player returns, their decayed rating means wins against them don't over-reward opponents and losses don't over-penalise them

  The 25% cap is smart here: it prevents decay from being so aggressive that a returning player becomes an easy rating farm for everyone they face.

• 6

  Promotion/relegation creates vertical mixing over time

  Players moving between divisions bring their adjusted ratings with them. A player who rises from 5th → 4th → 3rd over three seasons has played competitively at each level and been calibrated accordingly. Over multiple seasons this creates a web of indirect connections that ties all pools together.

• 7

  K-factor is universal

  The same K-factor formula (32 below 1500, 24 from 1500–2000, 16 above 2000) applies regardless of which league or format the game was played in. No "soft" local K-factor that creates a walled-off micro-economy.

Where We Are Genuinely Vulnerable

• 1

  Division depth disparity

  Cross-divisional games happen mainly at boundaries and in tournaments. Premier players almost never directly face 7th Division players except in open events. The pools are linked through a chain of divisions, not direct contact. This is inherent to any tiered league and can't be eliminated — only managed.

• 2

  Satellite league isolation risk

  Super Vets and Junior League players who never participate in CTTT League matches or tournaments could develop a slightly drifted pool. It's not a structural flaw — it's a participation pattern issue. Any player who cross-participates immediately bridges the gap.

• 3

  New player calibration period

  New players not yet in a league default to a 1200 seed. Even with division-anchored starting ratings, a genuinely exceptional new player placed in the wrong division will take 10–15 games to converge to their true level. During that window there's minor distortion to the players around them.

  This is mitigated by admin discretion: if a new player is well known within the local community, an admin can assign them a division placement or a more appropriate initial seed. The system uses that division's standard starting rating automatically. It doesn't need to be exact — the ELO self-corrects regardless — but a better seed means faster calibration and less early-period distortion. The K=32 floor accelerates convergence either way.

Compared to the Classic Problem

Verdict for Tournament Seedings