I compare the top 20 men’s singles and top 20 women’s singles players.
The goal is to answer:
This is a descriptive analysis of the ranking structure, not a causal study.
The data contain 40 elite badminton players:
The rankings used here are from October 13, 2026.
## discipline rank player country points ## 1 Men's Singles 1 Kunlavut Vitidsarn Thailand 91115 ## 2 Men's Singles 2 Anders Antonsen Denmark 86805 ## 3 Men's Singles 3 Christo Popov France 86632 ## 4 Men's Singles 4 Jonatan Christie Indonesia 85031 ## 5 Men's Singles 5 Shi Yu Qi China 84555 ## 6 Men's Singles 6 Chou Tien Chen Chinese Taipei 81220
The order is determined by ranking points, but the size of the point gaps between positions is not constant.
Correlation measures the direction and strength of a linear relationship.
\[ r = \frac{ \sum_{i=1}^{n}(x_i-\bar{x})(y_i-\bar{y}) }{ \sqrt{ \sum_{i=1}^{n}(x_i-\bar{x})^2 \sum_{i=1}^{n}(y_i-\bar{y})^2 } }. \]
Here:
For these data:
\[ r_{\text{men}} = -0.994 \]
\[ r_{\text{women}} = -0.909 \]
Both values are negative because higher rank numbers correspond to lower point totals.
A simple linear regression model summarizes the average trend:
\[ Y_i = \beta_0 + \beta_1X_i + \varepsilon_i. \]
For this project:
\[ \text{Points} = \beta_0+\beta_1(\text{Rank})+\varepsilon. \]
The estimated slopes are:
\[ \hat{\beta}_{1,\text{men}} = -2041.3 \]
\[ \hat{\beta}_{1,\text{women}} = -3120.4 \]
A negative slope means that ranking points generally decrease as we move farther down the top 20.
The women’s No. 1 player has a much larger point total than the rest of the field, while the men’s top positions are more closely grouped.
The women’s fitted line drops more sharply across the top 20, which suggests a larger average decline in ranking points by position.
library(ggplot2) library(plotly) library(dplyr) badminton <- read.csv( "bwf_top20_singles_oct13_2026.csv" ) men <- badminton %>% filter(discipline == "Men's Singles") women <- badminton %>% filter(discipline == "Women's Singles") cor(men$rank, men$points) cor(women$rank, women$points) lm(points ~ rank, data = men) lm(points ~ rank, data = women)
The code separates the two disciplines, calculates correlation, and fits one regression model for each group.
Hover over a point to see the player’s name, country, rank, and ranking points.
The women’s slope is more negative, so ranking points decline faster across these top-20 positions in this snapshot.
Both disciplines show a strong negative relationship between rank number and ranking points.
The point distributions, however, are not identical.
The women’s top 20 show a steeper decline, influenced by a particularly large point total at world No. 1. The men’s top positions are more tightly grouped.
Rank tells us the order of the players, while ranking points show how large the gaps between those positions are.
Because BWF rank is determined by ranking points, this analysis describes the ranking structure. It does not mean that rank causes points to change.
Badmintonic — BWF Men’s Singles World Rankings
Men’s Singles rankings
Badmintonic — BWF Women’s Singles World Rankings
Women’s Singles rankings
Both ranking pages used for this analysis are dated October 13, 2026.