# Load libraries and settings here
library(tidyverse)
library(here)
library(Lahman)
library(janitor)
library(plotly)
library(ggforce)

knitr::opts_chunk$set(
    warning = FALSE,
    message = FALSE,
    echo = F,
    comment = "#>",
    fig.path = "figs/", # Folder where rendered plots are saved
    fig.width = 8, # Default plot width
    fig.height = 5, # Default plot height
    fig.retina = 3 # For better plot resolution
)

# Put any other "global" settings here, e.g. a ggplot theme:
theme_set(theme_minimal(base_size = 20))

# Write code below here to load any data used in project
payroll_data <- read_csv(here('data_raw', 'payroll_data.csv'))
inflation_factors <- read_csv(here('data_raw', 'inflation_factors.csv'))

# For appendix
payroll_dict <- read_csv(here('data_raw', 'payroll_data_dict.csv'))
infl_dict <- read_csv(here('data_raw', 'inflation_factors_dict.csv'))
teams_dict <- read_csv(here('data_raw', 'team_dict.csv'))
s_p_dict <- read_csv(here('data_raw', 'stats_payroll_dict.csv'))
abbrevs <- read_csv(here('data_raw', 'abbrevs.csv'))
payroll_data <- payroll_data %>% 
    clean_names() %>% 
    mutate(nominal_est_payroll = est_payroll) %>% 
    select(c(year, franch_id, nominal_est_payroll)) %>% 
    filter(is.na(nominal_est_payroll) == F)

inflation_factors <- inflation_factors %>% 
    clean_names() 

teams <- Teams %>% 
    clean_names() %>% 
    select(c(year_id, team_id, franch_id, g, w, l)) %>% 
    filter(year_id >= 1969) %>% 
    mutate(
        team_id = as.character(team_id),
        franch_id = as.character(team_id),
        win_pct = w / g
    )
# CREATE Stats and Payroll Master Data Frame
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="SF","SFG")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="CAL","ANA")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="LAA","ANA")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="CHA","CHW")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="CHN","CHC")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="KCA","KCR")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="LAN","LAD")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="MON","WSN")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="WS2","WSN")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="WAS","WSN")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="NYA","NYY")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="NYN","NYM")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="SDN","SDP")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="SE1","MIL")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="ML4","MIL")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="SFN","SFG")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="SLN","STL")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="FLO","FLA")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="MIA","FLA")
teams$franch_id<-replace(teams$franch_id, teams$franch_id=="TBA","TBD")


# Merging df
stats_and_payroll <- merge(payroll_data, teams, by.x=c("franch_id", "year"), by.y=c("franch_id", "year_id"), all = TRUE) 

# Adjust df for new variables and removing NA
stats_and_payroll <- stats_and_payroll %>% 
    filter(!is.na(nominal_est_payroll) & !is.na(win_pct)) %>% 
    mutate(post_CBT = ifelse(year > 1993, "Post-CBT", "Pre-CBT"))

# Get rid of inflation variables that don't apply 
inflation_factors_adj <- inflation_factors %>% 
    filter(year > 1968)

# Create new df of applicable inflation factors and add to stats_and_payroll
count <- enframe(table(stats_and_payroll$year)) %>% 
    mutate_if(is.table, as.numeric) %>% 
    mutate(name = as.numeric(name))
infl_merge <- merge(count, inflation_factors_adj, by.x="name", by.y = "year")
count_array <- rep(infl_merge$infl_factor, times=infl_merge$value)
stats_and_payroll <- stats_and_payroll[with(stats_and_payroll, order(year)),]
stats_and_payroll <- stats_and_payroll %>% 
    mutate(
        infl_factor = count_array, 
        payroll_adj_for_infl = infl_factor*nominal_est_payroll, 
        ) 

plot <- stats_and_payroll %>% 
    ggplot(aes(x = payroll_adj_for_infl)) +
    geom_histogram(size = 0.5, color = "#002982", fill = "#002982", alpha = 0.7) + 
    labs(
        title = "Payroll Histogram", 
        subtitle = "MLB Franchises Pre and Post-CBT (1969-2021)"
    ) + 
    xlab("Team Payroll (in Millions of $)") +
    ylab("Count") +
    theme_light() + 
    theme(legend.position = "none") +
     scale_x_continuous(
         breaks = c(0, 10^8, 2*10^8, 3*10^8), 
         labels = c("0", "100", "200", "300"),
         limits = c(0, 3.005*10^8)) +
    theme(
        legend.title = element_blank(),
        plot.title=element_text(size=20), 
        axis.title.y = element_text(margin = margin(r=9)),
        axis.title.x = element_text(margin = margin(t=9))
    )
ggplotly(plot)
plot <- stats_and_payroll %>%
    ggplot(aes(x = win_pct)) +
    geom_density(size = 0.5, color = "#002982", alpha = 0.7, fill = "#002982") +
    theme_light() + 
    labs(
        title = "Winning Percent Distribution", 
        subtitle = "MLB Franchises Pre and Post-CBT (1969-2021)"
    ) + 
    xlab("Regular Season Winning Percent") + 
    ylab("Density") +
    theme(
        plot.title=element_text(size=20), 
        axis.title.y = element_text(margin = margin(r=9)),
        axis.title.x = element_text(margin = margin(t=9))
    )

plot
pre <- stats_and_payroll %>% 
    filter(year <= 1993)

post <- stats_and_payroll %>% 
    filter(year >= 1994)

plot <- stats_and_payroll %>% 
    ggplot(aes(
        x = payroll_adj_for_infl, 
        y = win_pct),
        color = post_CBT
    ) +
    geom_point(size = 1, color = '#002982') +
    facet_wrap(vars(post_CBT), nrow = 1) +
    labs(
        title = "Winning Percent based on Payroll", 
        subtitle = "MLB Franchises Pre and Post-CBT (1969-2021)"
    ) + 
    scale_x_continuous(
         breaks = c(0, 10^8, 2*10^8, 3*10^8), 
         labels = c("0", "100", "200", "300"),
         limits = c(0, 3.005*10^8)) +
    xlab("Team Payroll (in Millions of $)") + 
    ylab("Regular Season Winning Percent") + 
    theme_light() + 
    theme(
        legend.title = element_blank(),
        plot.title=element_text(size=20), 
        axis.title.y = element_text(margin = margin(r=9)),
        axis.title.x = element_text(margin = margin(t=9))
    )

plot
preCBT <- stats_and_payroll %>% 
    filter(post_CBT == "Pre-CBT") %>% 
    group_by(franch_id) %>% 
    summarise(avg_pre_win_pct = mean(win_pct)) 

postCBT <- stats_and_payroll %>% 
    filter(post_CBT == "Post-CBT", franch_id != c('ARI', 'TBD')) %>% 
    group_by(franch_id) %>% 
    summarise(avg_post_win_pct = mean(win_pct))

lineGraphData <- merge(preCBT, postCBT)

scatterData <- lineGraphData %>% 
    rename(Team=franch_id,
           "Pre-CBT"=avg_pre_win_pct,
           "Post-CBT"=avg_post_win_pct) %>%
    gather("Pre-CBT", "Post-CBT", key = "CBT",value = number)

combinedAvg <- inner_join(scatterData, lineGraphData, by = c("Team" = "franch_id"))

plot <- combinedAvg %>% 
    ggplot() +
    geom_segment(size = 0.8, color ="grey",
        aes(
            x = avg_pre_win_pct,
            xend = avg_post_win_pct,
            y = reorder(Team, avg_post_win_pct),
            yend = Team
        )
    ) +
    geom_point(aes(y = reorder(Team, avg_post_win_pct), x=number, color=CBT, shape=CBT),) +
    scale_color_manual(values = c("Pre-CBT" = "#002982","Post-CBT"= "#D50032")) +
    scale_shape_manual(values = c("Pre-CBT" = 16, "Post-CBT" =  17)) +
    labs(
        title = "Average Winning Percent Pre- and Post-CBT", 
        subtitle = "By MLB Franchise"
    ) + 
    ylab("Team Franchise ID") + 
    xlab("Regular Season Winning Percent") + 
    theme_light() + 
    theme(
        plot.title=element_text(size=18), 
        axis.title.y = element_text(margin = margin(r=9)),
        axis.title.x = element_text(margin = margin(t=9))
    )

plot
knitr::kable(payroll_dict)
knitr::kable(infl_dict)
knitr::kable(teams_dict)
knitr::kable(s_p_dict)
knitr::kable(abbrevs)