← back to standings

Championship Entropy

Measuring competitiveness with information theory. Shannon entropy quantifies how evenly championship points are distributed across the grid — higher entropy means a tighter fight.

What Is Shannon Entropy?

Information theory applied to championship competition

Shannon entropy, introduced by Claude Shannon in 1948, measures the uncertainty or disorder in a probability distribution. When applied to F1 championship points, it answers a simple question: how unpredictable is the championship?

We treat each driver's share of total points as a probability: p_i = points_i / total_points. Drivers with zero points are excluded (log(0) is undefined and they carry no information).

H = − Σ p_i × log₂(p_i)

where p_i = driver i's points / total points, summed over all drivers with points > 0

Maximum Entropy (Perfect Competition)

H_max = log₂(N) when all N drivers have equal points. Every driver equally likely to be champion. Complete uncertainty.

Minimum Entropy (Total Domination)

0.000 bits

H = 0 when one driver has all points. The outcome is certain. No surprise at all.

2026 Championship Entropy

After Round 5

Shannon Entropy (H)

bits

Maximum Possible (H_max)

bits (for 22 drivers)

Competitiveness Index

H / H_max as percentage

Entropy Scale

0 — Total Domination log₂(22) — Perfect Competition

Driver Entropy Contributions

Driver	Team	Points	p_i	−p_i log₂(p_i)	% of Total H

Historical Comparison

Championship entropy across notable F1 seasons — normalized to [0, 1] competitiveness index

Season Entropy Table

Season	Description	H (bits)	H_max	Competitiveness	Character

2026 Round-by-Round Entropy

How championship uncertainty has evolved through the first 5 races

Round-by-Round Values

Round	Race	Leader	Scorers	H (bits)	H / H_max	Trend

Normalized Entropy — The Competitiveness Index

A single number from 0% to 100% that captures how competitive a season is

Raw entropy H depends on the number of drivers — a 22-driver grid naturally has higher maximum entropy than a 20-driver grid. To compare across seasons with different grid sizes, we normalize:

C = H / H_max = H / log₂(N)

C = 1.0 (100%) means perfectly equal points. C = 0 means one driver has everything.

High Competitiveness (>85%)

Multiple drivers with realistic title chances. Points spread evenly. Example: early-season standings, 2010 finale.

Moderate (65%–85%)

Clear frontrunners but midfield still scoring. Some stratification visible. Most completed seasons fall here.

Low Competitiveness (<65%)

Heavy domination by one driver or team. Points heavily concentrated at the top. Example: 2023 Verstappen.