World Junior Championship Analytics

The World Junior Championship

The IIHF World Junior Championship (WJC) is one of hockey's premier showcase events for U20 prospects. Strong WJC performance often correlates with NHL success, but proper statistical context is essential—tournament sample sizes are small, and competition level varies significantly by opponent.

Analyzing WJC Performance

import pandas as pd
import numpy as np
from scipy import stats

# Load World Juniors data
wjc_stats = pd.read_csv('world_juniors_stats.csv')
nhl_careers = pd.read_csv('nhl_career_stats.csv')

# Merge WJC and NHL data
wjc_to_nhl = wjc_stats.merge(
    nhl_careers,
    on='player_id',
    suffixes=('_wjc', '_nhl')
)

# Filter for players who played in NHL
wjc_to_nhl = wjc_to_nhl[wjc_to_nhl['nhl_games'] >= 100]

# Calculate WJC per-game metrics
wjc_to_nhl['wjc_ppg'] = wjc_to_nhl['wjc_points'] / wjc_to_nhl['wjc_games']
wjc_to_nhl['wjc_gpg'] = wjc_to_nhl['wjc_goals'] / wjc_to_nhl['wjc_games']

# Correlation analysis
correlations = {
    'WJC PPG vs NHL PPG': np.corrcoef(
        wjc_to_nhl['wjc_ppg'],
        wjc_to_nhl['nhl_ppg']
    )[0, 1],
    'WJC Goals vs NHL Goals': np.corrcoef(
        wjc_to_nhl['wjc_gpg'],
        wjc_to_nhl['nhl_goals_per_game']
    )[0, 1]
}

print("=== WJC to NHL Correlation Analysis ===")
for metric, corr in correlations.items():
    print(f"{metric}: {corr:.3f}")

# Opponent quality adjustment
opponent_tiers = {
    'CAN': 1, 'USA': 1, 'SWE': 1, 'FIN': 1, 'RUS': 1,
    'CZE': 2, 'SVK': 2, 'SUI': 2,
    'GER': 3, 'LAT': 3, 'NOR': 3, 'AUT': 3
}

def calculate_quality_adjusted_performance(player_stats):
    """Adjust performance based on opponent strength"""
    total_adjusted_points = 0
    total_weight = 0

    for _, game in player_stats.iterrows():
        opponent = game['opponent']
        tier = opponent_tiers.get(opponent, 3)

        # Weight: Tier 1 = 1.0, Tier 2 = 0.8, Tier 3 = 0.6
        weight = 1.2 - (tier - 1) * 0.2

        total_adjusted_points += game['points'] * weight
        total_weight += weight

    if total_weight == 0:
        return 0

    return total_adjusted_points / total_weight

# Role analysis (line deployment)
def analyze_role_impact(df):
    """Analyze performance by role/line"""
    df['role_category'] = pd.cut(
        df['toi_per_game'],
        bins=[0, 12, 15, 18, 30],
        labels=['Limited', 'Middle-6', 'Top-6', 'Star']
    )

    return df.groupby('role_category').agg({
        'wjc_ppg': 'mean',
        'nhl_ppg': 'mean',
        'player_id': 'count'
    }).rename(columns={'player_id': 'player_count'})

role_analysis = analyze_role_impact(wjc_to_nhl)

print("\n=== WJC Role vs NHL Success ===")
print(role_analysis)

# Multi-tournament performance
multiple_wjc = wjc_stats.groupby('player_id').agg({
    'tournament_year': 'count',
    'points': 'sum',
    'games': 'sum'
}).rename(columns={'tournament_year': 'tournaments'})

multiple_wjc['career_wjc_ppg'] = (
    multiple_wjc['points'] / multiple_wjc['games']
)

# Players who played multiple tournaments
multi_tournament_players = multiple_wjc[multiple_wjc['tournaments'] >= 2]

print("\n=== Multi-Tournament Performance ===")
print(f"Players with 2+ tournaments: {len(multi_tournament_players)}")
print(f"Average PPG (multi-tournament): {multi_tournament_players['career_wjc_ppg'].mean():.3f}")

# Medal impact analysis
medal_games = wjc_stats[wjc_stats['game_type'].isin(['QF', 'SF', 'BM', 'GM'])]

medal_performance = medal_games.merge(
    nhl_careers[['player_id', 'nhl_ppg']],
    on='player_id'
)

# Compare medal game performance to NHL success
high_pressure_correlation = np.corrcoef(
    medal_performance.groupby('player_id')['points'].sum(),
    medal_performance.groupby('player_id')['nhl_ppg'].first()
)[0, 1]

print(f"\n=== Medal Round Performance Correlation ===")
print(f"Medal games PPG vs NHL PPG: {high_pressure_correlation:.3f}")

# Age factor in WJC performance
def analyze_age_impact(df):
    """Younger players face tougher competition"""
    df['age_category'] = pd.cut(
        df['age_at_tournament'],
        bins=[16, 18, 19, 20],
        labels=['U18', '18-19', '19-20']
    )

    return df.groupby('age_category').agg({
        'wjc_ppg': 'mean',
        'nhl_ppg': 'mean',
        'player_id': 'count'
    })

age_impact = analyze_age_impact(wjc_to_nhl)

print("\n=== Age Impact on WJC Performance ===")
print(age_impact)

# Current year prospects
current_wjc = pd.read_csv('current_wjc_players.csv')

def calculate_wjc_prospect_score(row):
    """Score prospects based on WJC performance"""
    score = 0

    # Production (0-40 points)
    ppg = row['points'] / row['games'] if row['games'] > 0 else 0
    if ppg >= 1.50:
        score += 40
    elif ppg >= 1.20:
        score += 35
    elif ppg >= 1.00:
        score += 30
    elif ppg >= 0.80:
        score += 25
    else:
        score += 15

    # Role/ice time (0-20 points)
    if row['toi_per_game'] >= 18:
        score += 20
    elif row['toi_per_game'] >= 15:
        score += 15
    else:
        score += 10

    # Medal round performance (0-20 points)
    if row['medal_round_points'] >= 3:
        score += 20
    elif row['medal_round_points'] >= 2:
        score += 15
    elif row['medal_round_points'] >= 1:
        score += 10
    else:
        score += 5

    # Team success (0-10 points)
    medal_points = {'Gold': 10, 'Silver': 8, 'Bronze': 6}
    score += medal_points.get(row['team_medal'], 0)

    # Age adjustment (0-10 points)
    if row['age'] <= 18:
        score += 10
    elif row['age'] == 19:
        score += 7
    else:
        score += 5

    return score

current_wjc['prospect_score'] = current_wjc.apply(
    calculate_wjc_prospect_score, axis=1
)

top_wjc_prospects = current_wjc.sort_values(
    'prospect_score', ascending=False
).head(20)

print("\n=== Top WJC Performers (Current Year) ===")
print(top_wjc_prospects[[
    'name', 'country', 'age', 'points', 'games',
    'toi_per_game', 'medal_round_points', 'prospect_score'
]])

library(tidyverse)
library(scales)

# Load WJC and NHL data
wjc_stats <- read_csv("world_juniors_stats.csv")
nhl_careers <- read_csv("nhl_career_stats.csv")

# Merge datasets
wjc_to_nhl <- wjc_stats %>%
  inner_join(nhl_careers, by = "player_id", suffix = c("_wjc", "_nhl")) %>%
  filter(nhl_games >= 100) %>%
  mutate(
    wjc_ppg = wjc_points / wjc_games,
    wjc_gpg = wjc_goals / wjc_games
  )

# Correlation analysis
cat("=== WJC to NHL Correlation Analysis ===\n")
cat(sprintf("WJC PPG vs NHL PPG: %.3f\n",
            cor(wjc_to_nhl$wjc_ppg, wjc_to_nhl$nhl_ppg)))
cat(sprintf("WJC Goals vs NHL Goals: %.3f\n",
            cor(wjc_to_nhl$wjc_gpg, wjc_to_nhl$nhl_goals_per_game)))

# Role analysis
role_analysis <- wjc_to_nhl %>%
  mutate(
    role_category = cut(
      toi_per_game,
      breaks = c(0, 12, 15, 18, 30),
      labels = c("Limited", "Middle-6", "Top-6", "Star")
    )
  ) %>%
  group_by(role_category) %>%
  summarise(
    avg_wjc_ppg = mean(wjc_ppg),
    avg_nhl_ppg = mean(nhl_ppg),
    player_count = n()
  )

cat("\n=== WJC Role vs NHL Success ===\n")
print(role_analysis)

# Multi-tournament analysis
multiple_wjc <- wjc_stats %>%
  group_by(player_id) %>%
  summarise(
    tournaments = n_distinct(tournament_year),
    total_points = sum(points),
    total_games = sum(games),
    career_wjc_ppg = total_points / total_games
  ) %>%
  filter(tournaments >= 2)

cat("\n=== Multi-Tournament Performance ===\n")
cat(sprintf("Players with 2+ tournaments: %d\n", nrow(multiple_wjc)))
cat(sprintf("Average PPG (multi-tournament): %.3f\n",
            mean(multiple_wjc$career_wjc_ppg)))

# Age impact analysis
age_impact <- wjc_to_nhl %>%
  mutate(
    age_category = cut(
      age_at_tournament,
      breaks = c(16, 18, 19, 20),
      labels = c("U18", "18-19", "19-20")
    )
  ) %>%
  group_by(age_category) %>%
  summarise(
    avg_wjc_ppg = mean(wjc_ppg),
    avg_nhl_ppg = mean(nhl_ppg),
    player_count = n()
  )

cat("\n=== Age Impact on WJC Performance ===\n")
print(age_impact)

# Current year prospects
current_wjc <- read_csv("current_wjc_players.csv")

# Prospect scoring function
calculate_wjc_prospect_score <- function(points, games, toi_per_game,
                                        medal_round_points, team_medal, age) {
  score <- 0
  ppg <- ifelse(games > 0, points / games, 0)

  # Production (0-40)
  score <- score + case_when(
    ppg >= 1.50 ~ 40,
    ppg >= 1.20 ~ 35,
    ppg >= 1.00 ~ 30,
    ppg >= 0.80 ~ 25,
    TRUE ~ 15
  )

  # Role (0-20)
  score <- score + case_when(
    toi_per_game >= 18 ~ 20,
    toi_per_game >= 15 ~ 15,
    TRUE ~ 10
  )

  # Medal round (0-20)
  score <- score + case_when(
    medal_round_points >= 3 ~ 20,
    medal_round_points >= 2 ~ 15,
    medal_round_points >= 1 ~ 10,
    TRUE ~ 5
  )

  # Team success (0-10)
  medal_scores <- c("Gold" = 10, "Silver" = 8, "Bronze" = 6)
  score <- score + ifelse(team_medal %in% names(medal_scores),
                         medal_scores[team_medal], 0)

  # Age (0-10)
  score <- score + case_when(
    age <= 18 ~ 10,
    age == 19 ~ 7,
    TRUE ~ 5
  )

  return(score)
}

current_wjc <- current_wjc %>%
  rowwise() %>%
  mutate(
    prospect_score = calculate_wjc_prospect_score(
      points, games, toi_per_game, medal_round_points, team_medal, age
    )
  ) %>%
  ungroup()

top_wjc_prospects <- current_wjc %>%
  arrange(desc(prospect_score)) %>%
  head(20)

cat("\n=== Top WJC Performers (Current Year) ===\n")
print(top_wjc_prospects %>%
  select(name, country, age, points, games, toi_per_game,
         medal_round_points, prospect_score))

# Visualization: WJC vs NHL performance
ggplot(wjc_to_nhl, aes(x = wjc_ppg, y = nhl_ppg)) +
  geom_point(aes(color = age_at_tournament), size = 3, alpha = 0.6) +
  geom_smooth(method = "lm", se = TRUE, color = "blue") +
  scale_color_gradient(low = "green", high = "red") +
  labs(title = "World Junior Championship Performance vs NHL Career",
       x = "WJC Points Per Game",
       y = "NHL Points Per Game",
       color = "Age at WJC") +
  theme_minimal()

# Team success visualization
team_medals <- current_wjc %>%
  count(country, team_medal) %>%
  filter(!is.na(team_medal))

ggplot(team_medals, aes(x = reorder(country, n), y = n, fill = team_medal)) +
  geom_col() +
  coord_flip() +
  scale_fill_manual(values = c("Gold" = "gold", "Silver" = "grey70",
                               "Bronze" = "#CD7F32")) +
  labs(title = "World Junior Championship Medal Distribution",
       x = "Country", y = "Medal Count", fill = "Medal") +
  theme_minimal()

Context Matters in WJC Analysis

World Junior performance should be contextualized carefully. A player scoring 10 points in 7 games against weak opponents differs from 7 points against medal contenders. Ice time, linemates, and role all significantly impact production. The best prospects dominate in both round-robin and medal games.