Article shows a graphical tool inspired by FiveThirtyEight’s MESSI* analysis that compares how every athlete played in every men’s World Cup from 1966 to 2018 by generating statistical fingerprints of 5,899 World Cup performance.
User inputs a player name and doppelgangers (similar players) are displayed on the right side of the screen. On the left side of the screen, a radar plot shows various metrics associated with that player.
MESSI here stands for Modeled Event Soccer Similarity Index (MESSI) and is a system that evaluates and compares player performances across 16 metrics.
Each metric is measured on a per-match basis, and for each metric we calculate a z-score — the number of standard deviations above or below average for that World Cup. The similarity between players’ performances is based solely on the average between each of their 16 z-scores
This is the article link: https://projects.fivethirtyeight.com/world-cup-comparisons/
soccer <- read.csv(file = 'https://projects.fivethirtyeight.com/soccer-api/international/2018/world_cup_comparisons.csv')Dataset contains player name, country, year played and 16 metrics used by the article to graphically compare players.
I’ll now create a subset that contains Player, Season, Country and Average of All Z-score.
soccer_subset<-subset(soccer, select=c(player,team,season))
soccer_subset$mean <- rowMeans(soccer[,4:19])head(soccer_subset[order(-soccer_subset$mean),], n=5)## player team season mean
## 66 Neymar Brazil 2018 2.492500
## 4102 Diego Maradona Argentina 1986 2.443750
## 480 Isco Spain 2018 2.342500
## 5802 Eusébio Portugal 1966 2.239375
## 5258 Johan Cruyff Netherlands 1974 2.120000Data set only contains z-score, not sure how to reproduce the article’s radar plots since they show statistics per match.