At What Age Do AFL Players Peak?
Zac Mawson
1 Introduction
The aim of this project was to find out at what age AFL players generally perform best. Once again the Wheel-o player ratings were used to determine performance. The data was used from Wheel-o ratings and was 2012-2026. Peak age will be separated by and compared between positions.
2 Data and Software Requirements
2.1 Required Data
This analysis uses AFL player statistics from 2012 to 2026.
The files should be saved in the same folder as this R Markdown file.
The files are:
afl-player-stats-2012.csvafl-player-stats-2013.csvafl-player-stats-2014.csvafl-player-stats-2015.csvafl-player-stats-2016.csvafl-player-stats-2017.csvafl-player-stats-2018.csvafl-player-stats-2019.csvafl-player-stats-2020.csvafl-player-stats-2021.csvafl-player-stats-2022.csvafl-player-stats-2023.csvafl-player-stats-2024.csvafl-player-stats-2025.csvafl-player-stats-2026.csv
Data source: https://www.wheeloratings.com/afl_stats.html?comp=afl&season=2026
2.2 Required Package
The tidyverse package is used.
# load library
library(tidyverse)3 Importing the Data
Each season is stored in a separate CSV file.
# import data
afl12 <- read.csv("afl-player-stats-2012.csv")
afl13 <- read.csv("afl-player-stats-2013.csv")
afl14 <- read.csv("afl-player-stats-2014.csv")
afl15 <- read.csv("afl-player-stats-2015.csv")
afl16 <- read.csv("afl-player-stats-2016.csv")
afl17 <- read.csv("afl-player-stats-2017.csv")
afl18 <- read.csv("afl-player-stats-2018.csv")
afl19 <- read.csv("afl-player-stats-2019.csv")
afl20 <- read.csv("afl-player-stats-2020.csv")
afl21 <- read.csv("afl-player-stats-2021.csv")
afl22 <- read.csv("afl-player-stats-2022.csv")
afl23 <- read.csv("afl-player-stats-2023.csv")
afl24 <- read.csv("afl-player-stats-2024.csv")
afl25 <- read.csv("afl-player-stats-2025.csv")
afl26 <- read.csv("afl-player-stats-2026.csv")A season variable is added before combining the data.
# add season
afl12$Season <- 2012
afl13$Season <- 2013
afl14$Season <- 2014
afl15$Season <- 2015
afl16$Season <- 2016
afl17$Season <- 2017
afl18$Season <- 2018
afl19$Season <- 2019
afl20$Season <- 2020
afl21$Season <- 2021
afl22$Season <- 2022
afl23$Season <- 2023
afl24$Season <- 2024
afl25$Season <- 2025
afl26$Season <- 2026The seasons are then combined into one dataset.
# combine data
afl <- bind_rows(
afl12,
afl13,
afl14,
afl15,
afl16,
afl17,
afl18,
afl19,
afl20,
afl21,
afl22,
afl23,
afl24,
afl25,
afl26
)4 Checking the Data
Some basic checks are done before the analysis.
# inspect data
head(afl)## Player Team Age Age_Decimal Matches RatingPoints_Avg
## 1 Aaron Black North Melbourne 22 22.08 3 7.233
## 2 Aaron Cornelius Brisbane 22 22.59 8 7.312
## 3 Aaron Davey Melbourne 29 29.55 8 11.300
## 4 Aaron Edwards North Melbourne 28 28.83 9 6.711
## 5 Aaron Hall Gold Coast 22 22.14 8 8.787
## 6 Aaron Joseph Carlton 23 23.49 13 5.338
## Supercoach_Avg DreamTeamPoints_Avg CoachesVotes_Total CoachesVotes_Avg
## 1 49.667 39.000 0 0.000
## 2 47.875 53.000 5 0.625
## 3 62.625 53.875 0 0.000
## 4 68.667 62.333 3 0.333
## 5 60.250 62.250 0 0.000
## 6 50.077 48.692 1 0.077
## CoachesVotes_MatchesPolled BnFVotes_Total TimeOnGround Kicks Handballs
## 1 0 0 72.667 3.333 4.333
## 2 1 0 79.750 6.250 3.500
## 3 0 0 79.250 6.500 4.125
## 4 1 0 85.444 6.889 3.889
## 5 0 1 72.500 8.125 4.500
## 6 1 0 76.846 7.538 4.769
## Disposals DisposalEfficiency KickingEfficiency HandballEfficiency
## 1 7.667 69.565 80.000 61.538
## 2 9.750 61.538 64.000 57.143
## 3 10.625 75.294 75.000 75.758
## 4 10.778 60.825 59.677 62.857
## 5 12.625 69.307 61.538 83.333
## 6 12.308 73.125 67.347 82.258
## KickPercentage Inside50s Rebound50s Clangers DisposalsPerClanger Turnovers
## 1 43.478 0.667 0.000 0.667 11.500 1.000
## 2 64.103 1.750 0.000 2.250 4.333 2.750
## 3 61.176 1.875 0.750 1.750 6.071 1.875
## 4 63.918 1.222 0.000 1.889 5.706 2.667
## 5 64.356 2.500 0.500 1.250 10.100 2.625
## 6 61.250 1.231 1.462 1.846 6.667 2.000
## DisposalsPerTurnover MetresGained MetresGainedPerDisposal
## 1 7.667 103.000 13.435
## 2 3.545 149.125 15.295
## 3 5.667 155.250 14.612
## 4 4.042 165.222 15.330
## 5 4.810 252.125 19.970
## 6 6.154 150.385 12.219
## ContestedPossessions UncontestedPossessions TotalPossessions
## 1 4.667 3.000 7.667
## 2 4.125 5.125 9.250
## 3 6.375 4.625 11.000
## 4 5.000 5.667 10.667
## 5 3.000 9.625 12.625
## 6 4.615 7.308 11.923
## ContestedPossessionRate Intercepts GroundBallGets GroundBallGetsForward50
## 1 60.870 0.333 4.000 2.333
## 2 44.595 0.500 2.375 1.250
## 3 57.955 1.875 5.125 1.750
## 4 46.875 0.556 3.111 1.667
## 5 23.762 1.500 2.250 0.625
## 6 38.710 4.231 4.000 0.077
## CentreBounceAttendances CentreClearances StoppageClearances TotalClearances
## 1 0.000 0.000 0.000 0.000
## 2 0.000 0.000 0.000 0.000
## 3 0.250 0.250 0.625 0.875
## 4 0.000 0.111 0.333 0.444
## 5 0.125 0.000 0.750 0.750
## 6 1.385 0.077 0.231 0.308
## Marks ContestedMarks MarksInside50 InterceptMarks MarksOnLead FreesFor
## 1 2.667 0.667 1.000 0.000 1.333 0.000
## 2 4.125 1.250 2.125 0.125 1.000 0.500
## 3 1.625 0.000 0.250 0.000 0.250 0.875
## 4 5.000 1.000 2.667 0.000 2.333 0.889
## 5 3.000 0.125 0.750 0.250 0.625 0.250
## 6 3.308 0.154 0.077 0.538 0.000 0.462
## FreesAgainst FreesDiff RuckContests RuckContestPercentage Hitouts
## 1 0.667 -0.667 0.00 0.000 0.00
## 2 0.750 -0.250 0.75 0.755 0.25
## 3 1.250 -0.375 0.00 0.000 0.00
## 4 0.667 0.222 0.00 0.000 0.00
## 5 0.500 -0.250 0.00 0.000 0.00
## 6 1.077 -0.615 0.00 0.000 0.00
## HitoutsWinPercentage HitoutsToAdvantage HitoutsToAdvantagePercentage Tackles
## 1 NA 0 NA 2.000
## 2 33.333 0 0 1.500
## 3 NA 0 NA 4.625
## 4 NA 0 NA 2.000
## 5 NA 0 NA 3.625
## 6 NA 0 NA 2.231
## TacklesInside50 ContestDefensiveOneOnOnes ContestDefensiveLossPercentage
## 1 1.333 0.000 NA
## 2 0.875 0.125 0
## 3 1.500 0.000 NA
## 4 0.667 0.111 100
## 5 0.750 0.000 NA
## 6 0.077 0.769 20
## ContestOffensiveOneOnOnes ContestOffensiveWinPercentage Goals_Total Goals_Avg
## 1 1.000 33.333 3 1.000
## 2 1.875 26.667 13 1.625
## 3 0.500 25.000 7 0.875
## 4 1.778 31.250 17 1.889
## 5 0.125 0.000 8 1.000
## 6 0.000 NA 1 0.077
## Behinds ShotsAtGoal GoalAssists ScoreboardImpact GoalAccuracy
## 1 0.333 1.667 0.667 10.333 60.000
## 2 0.625 2.375 0.375 12.625 68.421
## 3 0.375 1.375 0.375 7.875 63.636
## 4 0.667 3.111 0.778 16.667 60.714
## 5 0.625 2.000 0.375 8.875 50.000
## 6 0.000 0.077 0.077 0.923 100.000
## ScoreInvolvements ScoreInvolvementPercentage ScoreLaunches KickIns
## 1 3.333 13.699 0.333 0.000
## 2 3.875 19.018 0.250 0.000
## 3 4.125 19.760 0.875 0.125
## 4 5.444 19.291 0.333 0.000
## 5 3.750 19.481 0.625 0.000
## 6 2.308 9.585 0.615 0.077
## KickInsPlayOnPercentage Bounces OnePercenters PressureActs
## 1 NA 0.000 0.333 9.667
## 2 NA 0.000 0.750 10.250
## 3 0 0.125 0.500 17.625
## 4 NA 0.222 1.000 10.667
## 5 NA 0.125 0.625 15.875
## 6 0 0.154 2.154 7.769
## PressureActsDefensiveHalf Spoils Season Position FirstPossessions
## 1 1.667 0.333 2012 <NA> NA
## 2 2.625 0.375 2012 <NA> NA
## 3 4.500 0.125 2012 <NA> NA
## 4 2.111 0.556 2012 <NA> NA
## 5 4.875 0.500 2012 <NA> NA
## 6 5.308 1.615 2012 <NA> NA
## FirstPossessionToClearance HitoutToFirstPossession HitoutToClearance
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NA
## DisposalRetentionPercentage KicksIntoF50 KickIntoF50RetentionPercentage
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NA
## AssistedMetresGained NetMetresGained KickMetresGained RetainedMetresGained
## 1 NA NA NA NA
## 2 NA NA NA NA
## 3 NA NA NA NA
## 4 NA NA NA NA
## 5 NA NA NA NA
## 6 NA NA NA NA
## KickMetresGainedPerKick HandballMetresGained HandballMetresGainedPerHandball
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NA
## Equity_PreClearance Equity_PostClearance Equity_Possession Equity_BallUse
## 1 NA NA NA NA
## 2 NA NA NA NA
## 3 NA NA NA NA
## 4 NA NA NA NA
## 5 NA NA NA NA
## 6 NA NA NA NA
## xThreatPerKick ThreatRating xRetainPerKick RetentionRating
## 1 NA NA NA NA
## 2 NA NA NA NA
## 3 NA NA NA NA
## 4 NA NA NA NA
## 5 NA NA NA NA
## 6 NA NA NA NA
## GroundBallGetsDefensive50 HardBallGets LooseBallGets
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NA
## PostClearanceContestedPossessions PostClearanceGroundBallGets
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 5 NA NA
## 6 NA NA
## GathersFromHitout CrumbingPossessions HandballReceives AverageFieldPosition
## 1 NA NA NA NA
## 2 NA NA NA NA
## 3 NA NA NA NA
## 4 NA NA NA NA
## 5 NA NA NA NA
## 6 NA NA NA NA
## Defensive50Possessions DefensiveMidfieldPossessions
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 5 NA NA
## 6 NA NA
## AttackingMidfieldPossessions Forward50Possessions
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 5 NA NA
## 6 NA NA
## CentreBounceAttendancePercentage CentreClearanceRate RuckHardBallGets
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NA
## ScoreAssists xScoreShots xScorePerShot xScoreRatingPerShot xScoreShots_Set
## 1 NA NA NA NA NA
## 2 NA NA NA NA NA
## 3 NA NA NA NA NA
## 4 NA NA NA NA NA
## 5 NA NA NA NA NA
## 6 NA NA NA NA NA
## xScorePerShot_Set xScoreRatingPerShot_Set xScoreShots_General
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NA
## xScorePerShot_General xScoreRatingPerShot_General KickInPercentage
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 NA NA NA
## 6 NA NA NAdim(afl)## [1] 9976 127names(afl)## [1] "Player" "Team"
## [3] "Age" "Age_Decimal"
## [5] "Matches" "RatingPoints_Avg"
## [7] "Supercoach_Avg" "DreamTeamPoints_Avg"
## [9] "CoachesVotes_Total" "CoachesVotes_Avg"
## [11] "CoachesVotes_MatchesPolled" "BnFVotes_Total"
## [13] "TimeOnGround" "Kicks"
## [15] "Handballs" "Disposals"
## [17] "DisposalEfficiency" "KickingEfficiency"
## [19] "HandballEfficiency" "KickPercentage"
## [21] "Inside50s" "Rebound50s"
## [23] "Clangers" "DisposalsPerClanger"
## [25] "Turnovers" "DisposalsPerTurnover"
## [27] "MetresGained" "MetresGainedPerDisposal"
## [29] "ContestedPossessions" "UncontestedPossessions"
## [31] "TotalPossessions" "ContestedPossessionRate"
## [33] "Intercepts" "GroundBallGets"
## [35] "GroundBallGetsForward50" "CentreBounceAttendances"
## [37] "CentreClearances" "StoppageClearances"
## [39] "TotalClearances" "Marks"
## [41] "ContestedMarks" "MarksInside50"
## [43] "InterceptMarks" "MarksOnLead"
## [45] "FreesFor" "FreesAgainst"
## [47] "FreesDiff" "RuckContests"
## [49] "RuckContestPercentage" "Hitouts"
## [51] "HitoutsWinPercentage" "HitoutsToAdvantage"
## [53] "HitoutsToAdvantagePercentage" "Tackles"
## [55] "TacklesInside50" "ContestDefensiveOneOnOnes"
## [57] "ContestDefensiveLossPercentage" "ContestOffensiveOneOnOnes"
## [59] "ContestOffensiveWinPercentage" "Goals_Total"
## [61] "Goals_Avg" "Behinds"
## [63] "ShotsAtGoal" "GoalAssists"
## [65] "ScoreboardImpact" "GoalAccuracy"
## [67] "ScoreInvolvements" "ScoreInvolvementPercentage"
## [69] "ScoreLaunches" "KickIns"
## [71] "KickInsPlayOnPercentage" "Bounces"
## [73] "OnePercenters" "PressureActs"
## [75] "PressureActsDefensiveHalf" "Spoils"
## [77] "Season" "Position"
## [79] "FirstPossessions" "FirstPossessionToClearance"
## [81] "HitoutToFirstPossession" "HitoutToClearance"
## [83] "DisposalRetentionPercentage" "KicksIntoF50"
## [85] "KickIntoF50RetentionPercentage" "AssistedMetresGained"
## [87] "NetMetresGained" "KickMetresGained"
## [89] "RetainedMetresGained" "KickMetresGainedPerKick"
## [91] "HandballMetresGained" "HandballMetresGainedPerHandball"
## [93] "Equity_PreClearance" "Equity_PostClearance"
## [95] "Equity_Possession" "Equity_BallUse"
## [97] "xThreatPerKick" "ThreatRating"
## [99] "xRetainPerKick" "RetentionRating"
## [101] "GroundBallGetsDefensive50" "HardBallGets"
## [103] "LooseBallGets" "PostClearanceContestedPossessions"
## [105] "PostClearanceGroundBallGets" "GathersFromHitout"
## [107] "CrumbingPossessions" "HandballReceives"
## [109] "AverageFieldPosition" "Defensive50Possessions"
## [111] "DefensiveMidfieldPossessions" "AttackingMidfieldPossessions"
## [113] "Forward50Possessions" "CentreBounceAttendancePercentage"
## [115] "CentreClearanceRate" "RuckHardBallGets"
## [117] "ScoreAssists" "xScoreShots"
## [119] "xScorePerShot" "xScoreRatingPerShot"
## [121] "xScoreShots_Set" "xScorePerShot_Set"
## [123] "xScoreRatingPerShot_Set" "xScoreShots_General"
## [125] "xScorePerShot_General" "xScoreRatingPerShot_General"
## [127] "KickInPercentage"5 Preparing the Data
Only players who played at least 10 matches are used.
Players with a missing age or Player Rating are also removed.
# create analysis dataset
peak_data <- afl[
afl$Matches >= 10 &
!is.na(afl$Age) &
!is.na(afl$RatingPoints_Avg),
]
# inspect analysis data
dim(peak_data)## [1] 6702 127summary(peak_data$Age)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 19.00 22.00 25.00 25.43 28.00 39.00summary(peak_data$RatingPoints_Avg)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.350 7.139 8.915 9.208 10.938 22.213Using a minimum of 10 matches helps stop a player with only a few games from having too much effect on the results.
6 Average Player Rating by Age
The number of observations at each age is first calculated.
# count observations by age
age_counts <- aggregate(
RatingPoints_Avg ~ Age,
data = peak_data,
FUN = length
)
# rename count column
names(age_counts)[2] <- "N"
# display age counts
age_counts## Age N
## 1 19 231
## 2 20 383
## 3 21 506
## 4 22 608
## 5 23 649
## 6 24 621
## 7 25 610
## 8 26 586
## 9 27 539
## 10 28 477
## 11 29 404
## 12 30 353
## 13 31 259
## 14 32 191
## 15 33 139
## 16 34 70
## 17 35 43
## 18 36 19
## 19 37 8
## 20 38 4
## 21 39 2The average Player Rating is then calculated for each age.
# calculate average rating by age
age_ratings <- aggregate(
RatingPoints_Avg ~ Age,
data = peak_data,
FUN = mean
)
# rename average rating column
names(age_ratings)[2] <- "AverageRating"The average ratings and number of observations are then combined.
# combine age results
age_summary <- merge(
age_ratings,
age_counts,
by = "Age"
)
# only keep ages with at least 20 observations
age_summary <- age_summary[
age_summary$N >= 20,
]
# sort by age
age_summary <- age_summary[
order(age_summary$Age),
]
# display age summary
age_summary## Age AverageRating N
## 1 19 6.748597 231
## 2 20 7.573329 383
## 3 21 8.121115 506
## 4 22 8.722090 608
## 5 23 8.958313 649
## 6 24 9.513810 621
## 7 25 9.651289 610
## 8 26 9.661070 586
## 9 27 9.793699 539
## 10 28 9.790113 477
## 11 29 9.973765 404
## 12 30 9.868402 353
## 13 31 9.877286 259
## 14 32 9.402639 191
## 15 33 9.472705 139
## 16 34 9.622429 70
## 17 35 9.914023 43A minimum of 20 observations is used so very small age groups do not have too much effect.
7 Finding the Highest Rated Age
The age with the highest average Player Rating is found.
# find age with highest average rating
best_age <- age_summary[
which.max(age_summary$AverageRating),
]
# display highest rated age
best_age## Age AverageRating N
## 11 29 9.973765 4048 Plotting Player Rating by Age
The average Player Rating at each age is shown on a graph.
# create average rating by age plot
age_plot <- ggplot(
age_summary,
aes(
x = Age,
y = AverageRating
)
) +
geom_line() +
geom_point() +
labs(
title = "Average AFL Player Rating by Age",
x = "Age",
y = "Average Player Rating"
) +
theme_minimal()
# display age plot
age_plot
9 Quadratic Regression
A quadratic regression is used to model the relationship between age and Player Rating.
The squared age term allows the model to increase and then decrease instead of only fitting a straight line.
# fit quadratic regression
age_model <- lm(
RatingPoints_Avg ~
Age +
I(Age^2),
data = peak_data
)
# display model results
summary(age_model)##
## Call:
## lm(formula = RatingPoints_Avg ~ Age + I(Age^2), data = peak_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.1516 -1.9590 -0.2988 1.6926 12.2189
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -12.858468 1.329039 -9.675 <2e-16 ***
## Age 1.557667 0.102657 15.173 <2e-16 ***
## I(Age^2) -0.026531 0.001953 -13.587 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.739 on 6699 degrees of freedom
## Multiple R-squared: 0.07616, Adjusted R-squared: 0.07588
## F-statistic: 276.1 on 2 and 6699 DF, p-value: < 2.2e-1610 Estimating Peak Age
The coefficients from the quadratic model are used to estimate the age where the curve reaches its highest point.
# save model coefficients
age_coefs <- coef(age_model)
# calculate predicted peak age
predicted_peak_age <- -
age_coefs["Age"] /
(2 * age_coefs["I(Age^2)"])
# display predicted peak age
predicted_peak_age## Age
## 29.35612A range of ages is then created so the regression curve can be plotted.
# create age prediction values
age_range <- data.frame(
Age = seq(
min(age_summary$Age),
max(age_summary$Age),
by = 0.1
)
)
# predict rating across ages
age_range$PredictedRating <- predict(
age_model,
newdata = age_range
)
# find predicted maximum rating
predicted_peak_rating <- max(
age_range$PredictedRating
)
# display predicted peak rating
predicted_peak_rating## [1] 10.0050111 Quadratic Regression Plot
The regression curve is plotted with the observed average ratings.
# create quadratic regression plot
regression_plot <- ggplot(
age_summary,
aes(
x = Age,
y = AverageRating
)
) +
geom_point() +
geom_line(
data = age_range,
aes(
x = Age,
y = PredictedRating
)
) +
labs(
title = "AFL Player Rating and Age",
subtitle = "Quadratic regression curve",
x = "Age",
y = "Average Player Rating"
) +
theme_minimal()
# display regression plot
regression_plot
12 Comparing Positions
Peak age is also compared between AFL positions.
Players without a recorded position are removed.
# create position dataset
position_data <- peak_data[
!is.na(peak_data$Position) &
peak_data$Position != "",
]The number of observations for each age and position is calculated.
# calculate number of observations by age and position
position_counts <- aggregate(
RatingPoints_Avg ~ Age + Position,
data = position_data,
FUN = length
)
# rename count column
names(position_counts)[3] <- "N"The average Player Rating is then calculated for every age and position.
# calculate average rating by age and position
position_ratings <- aggregate(
RatingPoints_Avg ~ Age + Position,
data = position_data,
FUN = mean
)
# rename rating column
names(position_ratings)[3] <- "AverageRating"The results are combined.
Only age and position groups with at least 10 observations are kept.
# combine position results
position_summary <- merge(
position_ratings,
position_counts,
by = c("Age", "Position")
)
# only keep groups with at least 10 observations
position_summary <- position_summary[
position_summary$N >= 10,
]
# sort position results
position_summary <- position_summary[
order(
position_summary$Position,
position_summary$Age
),
]
# display position results
position_summary## Age Position AverageRating N
## 1 19 Gen. Defender 6.934435 23
## 7 20 Gen. Defender 6.728825 40
## 14 21 Gen. Defender 7.586371 70
## 21 22 Gen. Defender 7.753122 74
## 28 23 Gen. Defender 8.064024 82
## 35 24 Gen. Defender 8.166943 70
## 42 25 Gen. Defender 9.081400 70
## 49 26 Gen. Defender 8.628753 81
## 56 27 Gen. Defender 8.478716 74
## 63 28 Gen. Defender 8.861134 67
## 70 29 Gen. Defender 8.856492 61
## 77 30 Gen. Defender 8.540000 50
## 84 31 Gen. Defender 9.027816 38
## 91 32 Gen. Defender 7.882750 28
## 98 33 Gen. Defender 8.218789 19
## 2 19 Gen. Forward 6.497703 37
## 8 20 Gen. Forward 6.684408 49
## 15 21 Gen. Forward 7.190161 62
## 22 22 Gen. Forward 7.175136 66
## 29 23 Gen. Forward 7.622439 82
## 36 24 Gen. Forward 7.938653 75
## 43 25 Gen. Forward 8.490438 64
## 50 26 Gen. Forward 8.182283 60
## 57 27 Gen. Forward 8.337880 50
## 64 28 Gen. Forward 8.410821 39
## 71 29 Gen. Forward 8.862447 38
## 78 30 Gen. Forward 8.184806 36
## 85 31 Gen. Forward 8.241350 20
## 92 32 Gen. Forward 8.577235 17
## 99 33 Gen. Forward 8.982167 12
## 9 20 Key Defender 7.066700 10
## 16 21 Key Defender 7.092923 13
## 23 22 Key Defender 8.032654 26
## 30 23 Key Defender 7.843511 47
## 37 24 Key Defender 8.708386 44
## 44 25 Key Defender 8.747896 48
## 51 26 Key Defender 8.228463 41
## 58 27 Key Defender 8.930347 49
## 65 28 Key Defender 8.276085 47
## 72 29 Key Defender 8.871667 45
## 79 30 Key Defender 8.524515 33
## 86 31 Key Defender 7.952520 25
## 93 32 Key Defender 8.046722 18
## 4 19 Key Forward 5.032000 13
## 10 20 Key Forward 6.107536 28
## 17 21 Key Forward 6.908722 36
## 24 22 Key Forward 7.150600 35
## 31 23 Key Forward 8.010250 36
## 38 24 Key Forward 7.729607 28
## 45 25 Key Forward 7.834135 37
## 52 26 Key Forward 8.383946 37
## 59 27 Key Forward 8.747226 31
## 66 28 Key Forward 7.952516 31
## 73 29 Key Forward 9.006458 24
## 80 30 Key Forward 8.351765 17
## 87 31 Key Forward 9.766643 14
## 94 32 Key Forward 9.048824 17
## 101 33 Key Forward 8.097000 16
## 108 34 Key Forward 8.939667 12
## 5 19 Mid-Forward 6.468562 16
## 11 20 Mid-Forward 8.090826 23
## 18 21 Mid-Forward 7.988423 26
## 25 22 Mid-Forward 7.907333 27
## 32 23 Mid-Forward 7.819750 28
## 39 24 Mid-Forward 9.930792 24
## 46 25 Mid-Forward 9.574850 20
## 53 26 Mid-Forward 9.664571 14
## 60 27 Mid-Forward 10.433833 18
## 67 28 Mid-Forward 10.400462 13
## 74 29 Mid-Forward 9.971600 10
## 81 30 Mid-Forward 10.593800 10
## 6 19 Midfielder 7.092250 36
## 12 20 Midfielder 8.192887 62
## 19 21 Midfielder 8.712352 91
## 26 22 Midfielder 9.564765 102
## 33 23 Midfielder 10.662200 95
## 40 24 Midfielder 10.570161 93
## 47 25 Midfielder 10.740144 97
## 54 26 Midfielder 10.852263 95
## 61 27 Midfielder 10.719330 88
## 68 28 Midfielder 11.069112 80
## 75 29 Midfielder 11.709629 70
## 82 30 Midfielder 11.123159 63
## 89 31 Midfielder 11.223157 51
## 96 32 Midfielder 10.203313 32
## 103 33 Midfielder 10.358417 24
## 110 34 Midfielder 10.918000 11
## 27 22 Ruck 8.034154 13
## 34 23 Ruck 9.558412 17
## 41 24 Ruck 10.926500 22
## 48 25 Ruck 11.386375 24
## 55 26 Ruck 11.432286 21
## 62 27 Ruck 11.040955 22
## 69 28 Ruck 11.575263 19
## 76 29 Ruck 11.643214 14
## 83 30 Ruck 12.529167 18
## 90 31 Ruck 11.368211 19
## 97 32 Ruck 11.702214 14
## 104 33 Ruck 10.937000 1113 Finding Peak Age by Position
The age with the highest average Player Rating is found for each position.
# create empty peak position table
position_peaks <- data.frame()
# find peak age for each position
for (position in sort(
unique(position_summary$Position)
)) {
# select position
current_position <- position_summary[
position_summary$Position == position,
]
# find highest rated age
peak_row <- current_position[
which.max(
current_position$AverageRating
),
]
# add result
position_peaks <- rbind(
position_peaks,
data.frame(
Position = position,
PeakAge = peak_row$Age,
PeakRating = peak_row$AverageRating,
N = peak_row$N
)
)
}
# round ratings
position_peaks$PeakRating <- round(
position_peaks$PeakRating,
2
)
# display peak age by position
position_peaks## Position PeakAge PeakRating N
## 1 Gen. Defender 25 9.08 70
## 2 Gen. Forward 33 8.98 12
## 3 Key Defender 27 8.93 49
## 4 Key Forward 31 9.77 14
## 5 Mid-Forward 30 10.59 10
## 6 Midfielder 29 11.71 70
## 7 Ruck 30 12.53 1814 Plotting Age by Position
The relationship between age and Player Rating is shown separately for each position.
# create position plot
position_plot <- ggplot(
position_summary,
aes(
x = Age,
y = AverageRating
)
) +
geom_line() +
geom_point() +
facet_wrap(
~ Position
) +
labs(
title = "Average AFL Player Rating by Age and Position",
x = "Age",
y = "Average Player Rating"
) +
theme_minimal()
# display position plot
position_plot
15 Final Tables
A small table is created with the main overall values.
# create final results
final_results <- data.frame(
Result = c(
"Highest Observed Average Age",
"Quadratic Model Peak Age",
"Quadratic Model Peak Rating"
),
Value = c(
best_age$Age,
round(predicted_peak_age, 2),
round(predicted_peak_rating, 2)
)
)
# display final results
knitr::kable(
final_results,
caption = "Overall Peak Age Results"
)| Result | Value |
|---|---|
| Highest Observed Average Age | 29.00 |
| Quadratic Model Peak Age | 29.36 |
| Quadratic Model Peak Rating | 10.01 |
# display position peak table
knitr::kable(
position_peaks,
caption = "Observed Peak Age by AFL Position"
)| Position | PeakAge | PeakRating | N |
|---|---|---|---|
| Gen. Defender | 25 | 9.08 | 70 |
| Gen. Forward | 33 | 8.98 | 12 |
| Key Defender | 27 | 8.93 | 49 |
| Key Forward | 31 | 9.77 | 14 |
| Mid-Forward | 30 | 10.59 | 10 |
| Midfielder | 29 | 11.71 | 70 |
| Ruck | 30 | 12.53 | 18 |
16 Exporting the Results
The main tables can also be saved as CSV files.
# save age results
write.csv(
age_summary,
"afl_rating_by_age.csv",
row.names = FALSE
)
# save position results
write.csv(
position_peaks,
"afl_peak_age_by_position.csv",
row.names = FALSE
)