Prediction of 2019 MLB home run total at midseason (through games of June 27)

home_run_prediction.R

# 2019 statcast data is in data frame sc2019

# collect number of home runs in each game

library(tidyverse)

sc2019 %>%
  group_by(game_pk) %>%
  summarize(HR = sum(events == "home_run",
                     na.rm = TRUE)) -> S

# construct bar graph of home run distribution

library(TeachBayes)
bar_plot(S$HR) +
  increasefont() +
  ggtitle("Number of Home Runs in a Game - 2019") +
  centertitle() +
  xlab("HRs in Game")

# played 1211 games so far
# 2430 - 1211 = 1219 remaining games to play

current_hr <- 3311

HR_Predicted <- current_hr + 
                replicate(10000,
                sum(sample(S$HR,
                           replace = TRUE,
                           size = 1219)))

# graph predictions and show 95% interval estimate

(pred_limits <- quantile(HR_Predicted, c(0.025, 0.975)))

bar_plot(HR_Predicted) +
  increasefont() +
  ggtitle("Predicted 2019 Home Runs") +
  centertitle() +
  geom_vline(xintercept = pred_limits, size = 1.5)