［R］5,000打数以上経験した全プレイヤーの三振率とホームラン率（「Rによるセイバーメトリクス入門」（技術評論社）pp.61-63）

> library(Lahman)
> library(tidyverse)
> library(ggplot2)
> Batting %>% group_by(playerID) %>%
+ summarize(tAB = sum(AB, na.rm = TRUE),
+   tHR = sum(HR, na.rm = TRUE),
+   tSO = sum(SO, na.rm = TRUE)) -> long_careers
> Batting_5000 <- filter(long_careers, tAB >= 5000)
> print(head(Batting_5000))
# A tibble: 6 × 4
  playerID    tAB   tHR   tSO
  <chr>     <int> <int> <int>

1 aaronha01 12364   755  1383
2 abreubo01  8480   288  1840
3 adamssp01  5557     9   223
4 adcocjo01  6606   336  1059
5 alfoned01  5385   146   617
6 allendi01  6332   351  1556
> g <- ggplot(Batting_5000, aes(x = tHR / tAB, y = tSO / tAB)) +
+ geom_point() + geom_smooth()
> print(g)
`geom_smooth()` using method = 'loess' and formula = 'y ~ x'