NRG SQUASH

The Official, Over-Engineered Squash League!

5 Players. 1 League. Lots of Excuses.

Live results, permanent records and room for plenty of debate.

THE IMPORTANT BIT

League Table

Office Champ is the Elo-based ranking used to order the table. Season Points are calculated as games won plus 0.5 participation points per completed match.
Rank Player Played Won Lost Participation Season Points Office Champ
Loading standings…

OFFICIAL ADMINISTRATION

Enter Match Result

Select two players and enter the result.

PERMANENT EVIDENCE

Recent Results

Loading results…

MEET THE ATHLETES

Player Profiles

Five very different approaches to the same small court, each supported by firm opinions and selective memory.

Loading players…

THE CONSTITUTION

Rules

1. Match result

The player winning the most games wins the match. Results are entered as the number of games won by each player, for example 2–1. Draws are not accepted by the bureaucracy because somebody must eventually take responsibility.

2. Participation

Each player receives 0.5 participation points for every completed match, regardless of the result. It rewards actually arriving and playing, without allowing attendance alone to overpower the results on court.

3. Season Points

Season Points equal games won plus participation points. A 2–1 result therefore awards 2.5 points to the winner and 1.5 points to the loser. A 2–0 result awards 2.5 points to the winner and 0.5 to the loser.

4. Office Champ

Players are ranked by the Office Champ rating, which uses the Elo system. Beating a higher-rated player earns more rating points than beating somebody below you, while losing to a stronger player costs fewer points. This measures playing strength rather than simply rewarding whoever has played the most matches.

5. How the K-factor Works

The K-factor controls how quickly the Office Champ rating changes. A higher K-factor creates larger movements after every match, while a lower value makes the rankings more stable. This league uses a K-factor of 20: upsets are rewarded properly, but one unusual result should not completely rearrange the table.

6. Corrections

Incorrect results may be corrected by Dave's administrator login. When an old result is changed, the system recalculates every later Elo movement in chronological order so the current table remains consistent.

THE MATHEMATICS BEHIND THE ARGUMENTS

How the Office Champ Elo Rating Works

Elo is a rating system, not a league-points system. It estimates playing strength from every recorded match. Everyone starts at 1000. After each match, rating points move from the loser to the winner. The amount depends on how likely each player was to win before the match.

Starting rating 1000
K-factor 20
Current input Match win or loss
Total rating Always conserved

STEP 1

Calculate the expected result

Before a match, the system calculates Player A's expected score:

EA = 11 + 10(RB − RA) / 400

RA and RB are the players' ratings before the match. Player B's expected score is 1 − EA. The two probabilities therefore add up to 100%.

STEP 2

Enter the actual result

The current system treats the entire match as a simple win or loss:

WinnerS = 1
LoserS = 0
The entered game score decides who won and updates Games Won, Games Lost and Season Points. It does not currently change the Elo margin. A 3–0 win and a 3–2 win count as the same Elo match result.

STEP 3

Update both ratings

The new rating is calculated with:

R′ = R + K(S − E)

R is the old rating, R′ is the new rating, K is 20, S is the actual result and E is the expected result. The winner gains exactly what the loser gives up, apart from small display-rounding differences.

EXAMPLE 1 · EVEN MATCH

Dave 1000 vs Jakes 1000

Both players are expected to have a 50% chance. Dave wins.

Dave: 1000 + 20 × (1 − 0.50) = 1010 Jakes: 1000 + 20 × (0 − 0.50) = 990

EXAMPLE 2 · FAVOURITE WINS

Olaf 1100 vs Matthew 900

Olaf is expected to win roughly 76% of the time. Olaf wins, so the result is not surprising.

Olaf: 1100 + 20 × (1 − 0.76) ≈ 1105 Matthew: 900 + 20 × (0 − 0.24) ≈ 895

EXAMPLE 3 · UPSET

Matthew 900 vs Olaf 1100

Matthew is expected to win only roughly 24% of the time. Matthew wins, so the ratings move much more.

Matthew: 900 + 20 × (1 − 0.24) ≈ 915 Olaf: 1100 + 20 × (0 − 0.76) ≈ 1085

THE RICH VS OLAF QUESTION

How can Rich rank above Olaf after losing to him?

Elo considers the complete sequence of results, not only the latest head-to-head match. Olaf gained rating when he beat Rich. Rich then gained rating from several later wins against other players. Those gains can eventually move Rich above Olaf even though Olaf remains unbeaten and won their direct match.

This does not mean the system has forgotten Olaf's win. That result remains part of both ratings. It means Rich has supplied more subsequent evidence through additional matches. Elo measures the accumulated rating produced by all results; it does not enforce a rule that the head-to-head winner must always remain above the loser.

Season Points Who has accumulated the best season?
Office Champ Elo What strength estimate follows from all recorded match results?

SECURITY THEATRE

Player Login

Sign in using your approved NRG office email address.