Chicago Bears Receivers / Anthony Miller

Anthony Miller

library(tidyverse)
library(dplyr)
library(ggplot2)
library(ggrepel)

pbp_all <- 
  readRDS("pbpdata2009to2019.rds")
pbp_2018 <-
  pbp_all %>% filter(season == 2018)

#Love Copy + Pasting Ben Baldwin's code
pbp_2018_rp <- pbp_2018 %>%
  filter(!is.na(epa), !is.na(posteam), play_type=="no_play" | play_type=="pass" | play_type=="run") %>%
  mutate(
    pass = if_else(str_detect(desc, "(pass)|(sacked)|(scramble)"), 1, 0),
    rush = if_else(str_detect(desc, "(left end)|(left tackle)|(left guard)|(up the middle)|(right guard)|(right tackle)|(right end)") & pass == 0, 1, 0),
    success = ifelse(epa>0, 1 , 0),
    passer_player_name = ifelse(play_type == "no_play" & pass == 1, 
                                str_extract(desc, "(?<=\\s)[A-Z][a-z]*\\.\\s?[A-Z][A-z]+(\\s(I{2,3})|(IV))?(?=\\s((pass)|(sack)|(scramble)))"),
                                passer_player_name),
    receiver_player_name = ifelse(play_type == "no_play" & str_detect(desc, "pass"), 
                                  str_extract(desc, "(?<=to\\s)[A-Z][a-z]*\\.\\s?[A-Z][A-z]+(\\s(I{2,3})|(IV))?"),
                                  receiver_player_name),
    rusher_player_name = ifelse(play_type == "no_play" & rush == 1, 
                                str_extract(desc, "(?<=\\s)[A-Z][a-z]*\\.\\s?[A-Z][A-z]+(\\s(I{2,3})|(IV))?(?=\\s((left end)|(left tackle)|(left guard)|		      (up the middle)|(right guard)|(right tackle)|(right end)))"),
                                rusher_player_name),
    name = ifelse(!is.na(passer_player_name), passer_player_name, rusher_player_name),
    yards_gained=ifelse(play_type=="no_play",NA,yards_gained),
    play=1
  ) %>%
  filter(pass==1 | rush==1)

#First, let's calculate some of Mitch's stats so we can use them later
goat_stats <-
pbp_2018_rp %>% 
  filter(posteam == "CHI",
         play_type == "pass",
         !is.na(receiver_player_name),
         passer_player_name == "M.Trubisky") %>% 
  summarize(success_rate = mean(success), mean_epa = mean(epa), dropbacks = n(), mean_adot = mean(air_yards, na.rm = TRUE))

#Define Bears dataset
pbp_bears_receivers <- 
  pbp_2018_rp %>% 
  filter(posteam == "CHI",
         !is.na(receiver_player_name),
         play_type == "pass",
         down <= 4) %>% 
  group_by(receiver_player_name)

#Success Rate vs EPA

pbp_bears_receivers_epa <-
  pbp_bears_receivers %>% 
  summarize(success_rate = mean(success), mean_epa = mean(epa), targets = n()) %>% 
  filter(targets >= 30)

ggplot(pbp_bears_receivers_epa, aes(x = success_rate, y = mean_epa)) + 
  geom_point(aes(size = targets)) + 
  geom_text_repel(aes(label = receiver_player_name),
                  nudge_y = .0125,
                  nudge_x = .0125) +
  theme_bw() +
  geom_hline(yintercept = goat_stats$mean_epa, lty = 2, color = "blue") +
  geom_vline(xintercept = goat_stats$success_rate, lty = 2, color = "blue") +
  annotate("text", x = .49, y= .225, angle = 0,label = "Mitch Averages", size = 3.5, color = "blue") +
    labs(x = "Success Rate",
       y = "EPA per Target",
       title = "Bears Receivers: Success Rate and EPA", 
       subtitle = "Minimum 30 Targets",
       caption = "Data from nflscrapR \n \n @ChiBearsStats"
  )
ggsave('successratevsepa.png', dpi=1000)


#Receivers aDOT
pbp_bears_receivers_adot <-  
  pbp_bears_receivers %>%
  summarize(adot = mean(air_yards), targets = n()) %>%
  filter(targets >= 30)

ggplot(data = pbp_bears_receivers_adot, aes(x = receiver_player_name, y = adot)) + 
  geom_bar(stat = "identity", color = "blue", fill = "orange") +
  geom_hline(yintercept = goat_stats$mean_adot, lty = 2, color = "blue") +
  annotate("text", x = 4, y= 9.37, angle = 0,label = "Mitch Avg aDOT", size = 3.5, color = "blue") +
    theme_bw() + 
  labs(
    x = "Receiver",
    y = "aDOT",
    title = "Bears Receivers Average Depth of Target",
    subtitle = "Minimum 30 Targets", 
    caption = "Data from nflscrapR \n \n @ChiBearsStats"
    )
ggsave('aDOTeachreceiver.png', dpi=1000)

#Receivers YAC
pbp_bears_receivers_yac <-  
  pbp_bears_receivers %>%
  summarize(yac = mean(yards_after_catch, na.rm = TRUE), targets = n()) %>%
  filter(targets >= 30)

ggplot(data = pbp_bears_receivers_yac, aes(x = receiver_player_name, y = yac)) + 
  geom_bar(stat = "identity", color = "blue", fill = "orange") +
  theme_bw() + 
  labs(
    x = "Receiver",
    y = "YAC",
    title = "Bears Receivers Average YAC",
    subtitle = "Minimum 30 Targets", 
    caption = "Data from nflscrapR \n \n @ChiBearsStats"
  )
ggsave('YACeachreceiver.png', dpi=1000)



#Receivers aDOT but a boxplot
pbp_bears_receivers_adot_boxplot <-
  pbp_bears_receivers %>%
  filter(receiver_player_name %in% c("A.Miller", "A.Robinson II", "T.Burton", "T.Cohen", "T.Gabriel")) %>% 
  select(receiver_player_name, air_yards)


ggplot(data = pbp_bears_receivers_adot_boxplot, aes(x = receiver_player_name, y = air_yards)) + 
  geom_boxplot() + 
  geom_hline(yintercept = goat_stats$mean_adot, lty = 2, color = "blue") +
  geom_hline(yintercept = 0, lty = 2, color = "red") +
  theme_bw() + 
  labs(
    x = "Receiver",
    y = "aDOT",
    title = "Bears Receivers Average Depth of Target",
    subtitle = "Minimum 30 Targets (Only Mitch Throws)", 
    caption = "Data from nflscrapR \n \n @ChiBearsStats"
  )

ggsave('adotboxplot.png', dpi=1000)


#Receivers Density Targets
pbp_bears_receivers_adot_density <-
pbp_bears_receivers %>%
  group_by(receiver_player_name,air_yards) %>%
  summarize(targets = n()) %>% 
  filter(receiver_player_name %in% c("A.Miller", "A.Robinson II", "T.Burton", "T.Cohen", "T.Gabriel")) 
  
  
  
ggplot(data = pbp_bears_receivers_adot_density, aes(x=air_yards, y=..density..)) + 
  geom_density(aes(fill= receiver_player_name), position="stack") +
  theme_classic() +
  labs(
    x = "aDOT",
    y = "Density",
    fill = "Receiver Name",
    title = "Mitch Trubisky's Average Depth of Target Density Chart",
    subtitle = "Minimum 30 Targets", 
    caption = "Data from nflscrapR \n \n @ChiBearsStats"
  )

ggsave('adotdensity.png', dpi=1000)

#Now by Violin

pbp_bears_receivers_adot_violin <-
  pbp_bears_receivers %>%
  group_by(receiver_player_name,air_yards) %>%
  summarize(targets = n()) %>% 
  filter(receiver_player_name %in% c("A.Miller", "A.Robinson II", "T.Burton", "T.Cohen", "T.Gabriel")) 

  ggplot(pbp_bears_receivers_adot_violin, aes(factor(receiver_player_name), y= air_yards)) +
  geom_violin(aes(fill = receiver_player_name))  +
  theme_bw() +
  geom_hline(yintercept = goat_stats$mean_adot, lty = 2, color = "blue") +
  geom_hline(yintercept = 0, lty = 2, color = "red") +
  annotate("text", x = 3, y= 11, angle = 0,label = "Average aDOT", size = 3.5, color = "blue") +
  labs(
      x = "Receiver",
      y = "aDOT",
      fill = "Receiver Name",
      title = "Bears Receivers Average Depth of Target",
      subtitle = "Minimum 30 Targets (Only Mitch Throws)", 
      caption = "Data from nflscrapR \n \n @ChiBearsStats"
    )


ggsave('adotviolin.png', dpi=1000)

#aDOT vs. Catch Rate
pbp_receivers_adot_vs_catch <-
pbp_2018_rp %>%
  filter(!is.na(receiver_player_name) & play_type == "pass" & down <= 4) %>%
  group_by(posteam, receiver_player_name) %>%
  summarize(adot = mean(air_yards), targets = n(), catch_rate = sum(complete_pass)/targets) %>%
  filter(adot >= 5 & targets >= 50)

ggplot(data = pbp_receivers_adot_vs_catch, aes(x = adot, y = catch_rate)) + 
    geom_point() + 
    geom_text_repel(
      data = subset(pbp_receivers_adot_vs_catch, (posteam == "CHI")),
      aes(label = receiver_player_name),
      color = "blue",
      fontface = "bold",
      nudge_y  = .045,
      nudge_x =  .55) +
    geom_text_repel(
      data = subset(pbp_receivers_adot_vs_catch, 
                    ((posteam != "CHI") & catch_rate > .8
                    |(posteam != "CHI") & catch_rate < .5
                    |(posteam != "CHI") & adot > 15)),
      aes(label = paste(receiver_player_name, posteam, sep=" "))) +
  geom_smooth(method = lm, se = FALSE, linetype = "dashed", color = "black") + 
  scale_y_continuous(breaks = c(.4,.45,.5,.55,.6,.65,.7,.75,.8,.85)) + 
  scale_x_continuous(breaks = c(6,8,10,12,14,16,18,20)) + 
  geom_hline(yintercept = mean(pbp_receivers_adot_vs_catch$catch_rate), lty = 2, color = "red") +
  geom_vline(xintercept = mean(pbp_receivers_adot_vs_catch$adot), lty = 2, color = "red") +
    theme_bw() + 
  labs(x = "aDOT", 
       y = "Catch Rate",
       title = "Catch Rate vs aDOT",
       subtitle = "Minimum 50 Targets", 
       caption = "Data from nflscrapR \n \n @ChiBearsStats"
  )

ggsave('catchratevsadot.png', dpi=1000)


#Not a single outlier for Anthony Miller? Curiousity's sake:
doesanthonymillergodeep <-
  pbp_bears_receivers %>% 
  filter(receiver_player_name == "A.Miller") %>% 
  arrange(desc(air_yards)) 

doesanthonymillergodeep[(1:5), "air_yards"]


#Maybe Better in Red Zone?

pbp_bears_receivers_epa_redzone <-
  pbp_bears_receivers %>% 
  mutate(redzone = (ifelse(yardline_100 <= 20, "redzone", "not redzone"))) %>% 
  group_by(receiver_player_name, redzone) %>% 
  summarize(success_rate = mean(success), mean_epa = mean(epa), targets = n()) %>% 
  filter(receiver_player_name %in% c("A.Miller", "A.Robinson II", "T.Burton", "T.Cohen", "T.Gabriel")) 

#Let's get target distribution

ggplot(data = pbp_bears_receivers_epa_redzone , aes(x=receiver_player_name, y= targets)) + 
  geom_bar(aes(fill= redzone), stat = "identity") +
  geom_text(x = 1, y = 7.5, label = "20.3%") +
  geom_text(x = 2, y = 7.5, label = "12.8%") +
  geom_text(x = 3, y = 7.5, label = "18.4%") +
  geom_text(x = 4, y = 7.5, label = "15.6%") +
  geom_text(x = 5, y = 5, label = "8.7%") +
  theme_bw() +
  labs(
    x = "Receiver",
    y = "Total Targets",
    fill = "Redzone",
    title = "Bears Receivers Average Depth of Target",
    subtitle = "Minimum 30 Targets (Only Mitch Throws)", 
    caption = "Data from nflscrapR \n \n @ChiBearsStats"
  )

ggsave('redzonetargets.png', dpi=1000)

#Then by EPA

ggplot(pbp_bears_receivers_epa_redzone, aes(x = receiver_player_name, y = mean_epa)) + 
  geom_point(aes(color = receiver_player_name, shape = redzone), size = 5) +
  labs(x = "Receiver",
       y = "EPA per Target",
       shape = "Field Position",
       color = "Receiver Name",
       title = "Bears Receivers: Difference in Red Zone EPA?", 
       subtitle = "Minimum 30 Targets",
       caption = "Data from nflscrapR \n \n @ChiBearsStats")
  
ggsave('eparedzone.png', dpi=1000)

#Now by Success Rate

ggplot(pbp_bears_receivers_epa_redzone, aes(x = receiver_player_name, y = success_rate)) + 
  geom_point(aes(color = receiver_player_name, shape = redzone), size = 5) +
  labs(x = "Receiver",
       y = "Success Rate",
       shape = "Field Position",
       color = "Receiver Name",
       title = "Bears Receivers: Difference in Success Rate?", 
       subtitle = "Minimum 30 Targets",
       caption = "Data from nflscrapR \n \n @ChiBearsStats")

ggsave('successrateredzone.png', dpi=1000)