index.Rmd

---
title: "March Mammal Madness"
output:
  html_document:
   toc: true
   toc_float: true
   toc_depth: 5

---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, warning = FALSE, message = FALSE)
```
Are you playing March Mammal Madness this year?? 

Check out [#2023MMM](https://twitter.com/search?q=%232021MMM&src=typed_query) for all the latest tweets, check out the [play by play](http://mammalssuck.blogspot.com/2021/02/march-mammal-madness-2021.html) Or have a look at the [official website](https://libguides.asu.edu/MarchMammalMadness) for all of the latest updates.

# 2023

## 2023 Leaderboard

```{r}
library(tidyverse)
library(here)
library(plotly)
library(readxl)


fils <- list.files (path = "2023", pattern="*.csv") 

# create a list in which to store entries
fl <- list()

#loop through files and add to the list
#using here means that I can easily read files in a subdirectory
for (kk in 1:length(fils)){
  entry <- read.csv(here("2023",fils[kk]))
  
  fl[[kk]] <- as.vector(entry[,1])
}


### counting points 

######################## generating the answers 
mast <- read_excel(here("2023_results.xlsx"))
wild <- mast %>%  select("...2") %>%drop_na() %>% rename(Wildcard = "...2")
a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                      Round_2 = Round_2...6,
                                                      Round_3 = Round_3...7,
                                                      Round_4 = Round_4...8,
                                                      Round_5 = Round_5...9)
b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                        Round_2 = Round_2...14, 
                                                        Round_3 = Round_3...13, 
                                                        Round_4 = Round_4...12, 
                                                        Round_5 = Round_5...11 )
cc <- bind_rows(a,b)
r1 <- cc %>% select(Round_1) %>% drop_na()
r1 <- r1$Round_1

r2 <- cc %>% select(Round_2) %>% drop_na()
r2 <- r2$Round_2

r3 <- cc %>% select(Round_3) %>% drop_na()
r3 <- r3$Round_3

r4 <- cc %>% select(Round_4) %>% drop_na()
r4 <- r4$Round_4

r5 <- cc %>% select(Round_5) %>% drop_na()
r5 <- r5$Round_5

wild <- wild$Wildcard

fin <- mast %>% select(Final) %>% drop_na()
fin <- fin$Final

master <- c(wild,r1,r2,r3,r4,r5,fin)

# 1 = same 0 = different -- ony get points for answers that are the same as the master


# check where in the list the master file is 
#ans


# create a list of players names

players <- sub("_", " ", fils) %>% sub(".csv", "",.)
#players <- sub(".csv", "", players)

### check answers 

scorelist <- list()
for (ii in 1: length(players)){
  
  ## this checks answers for each round 
  playa <- fl[[ii]]
  mstr <- master
  #mstr <- ans[[1]]
  rnd1 <- unique(playa[1:33]) %in% mstr[1:33]
  rnd2 <- unique(playa[34:49]) %in% mstr[34:49]
  sweet <- unique(playa[50:57]) %in% mstr[50:57]
  elite <- unique(playa[58:61]) %in% mstr[58:61]
  roar <- unique(playa[62:63]) %in% mstr[62:63]
  win <- unique(playa[64]) %in% mstr[64]
  
  
  pts <- c(sum(rnd1, na.rm = TRUE),sum(rnd2*2, na.rm = TRUE), sum(sweet*3, na.rm = TRUE), sum(elite*5, na.rm = TRUE), sum(roar*8, na.rm = TRUE),sum(win*13, na.rm = TRUE))
  scorelist[[ii]] <- pts
}

leaderboard <- as.data.frame(do.call(rbind, scorelist))
colnames(leaderboard) <- c("Round_1", "Round_2", "Sweet_16", "Elite_8", "Final_roar", "Winner")
#tardy <- c(1,0,0,0,0,0,1,0,0,1,1,1,0)


leaderboard$Player_name <- players
#leaderboard$Tardigrade_tally <- tardy



leaderboard <- leaderboard %>% 
  select(Player_name,  Round_1, Round_2, Sweet_16, Elite_8, 
         Final_roar, Winner) %>% 
  mutate(Total= rowSums(.[2:7])) %>%
  arrange(desc(Total))

# creating a plot with cumulative points 
#create sum column for after each round 
pl_dat <- leaderboard %>% mutate(r2 = Round_1+Round_2) %>% mutate(r3 = r2+Sweet_16)

pl_dat <- leaderboard %>% 
  mutate(r1 =rowSums(.[2])) %>% 
  mutate(r2 =rowSums(.[2:3])) %>% 
  mutate(r3 =rowSums(.[2:4])) %>%
  mutate(r4 =rowSums(.[2:5])) %>% 
  mutate(r5 =rowSums(.[2:6])) %>% 
  rename(r6 = Total) %>%
  select(Player_name, r1, r2, r3, r4, r5, r6) %>% 
  gather("Round", "points", 2:7)

pl_dat$Round <- gsub("r","",pl_dat$Round)

p <- ggplot(data=pl_dat, aes(x=Round, y=points, group = Player_name, colour = Player_name)) + 
  geom_point() +
  ggtitle("Point accumulation")
#+ scale_colour_manual(values=cols) 
  
# working out the most popular winner
winners <- vector(length = length(players))

for (ii in 1:length(winners)){

  this_one <- fl[[ii]]
  winners[ii] <- this_one[64]
  
}


outcomes <- data.frame(players, winners)
# creating line breaks for plotting 
levels(outcomes$winners) <- gsub(" ", "\n",levels(outcomes$winners))

outcomes <- separate(outcomes, players, into = c("players", "group"), sep = "_(?=[^_]+$)")
```

```{r}
#leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MMM", " MEP")
# create list of MEP, BIO2029 players
leaderboardall <- separate(leaderboard, Player_name, into = c("Player_name", "group"), sep = "_(?=[^_]+$)")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MEP", "")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"_", "")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"BIO2041", "")
leaderboard %>% arrange(-Total) %>% 
knitr::kable()
```

```{r 2023winners}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
predicted_winners <- outcomes %>% select(winners) %>% group_by(winners) %>% count() %>% rename("num"= "n") %>% arrange(-num) %>% ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```


```{r, 2023bumpall, eval = FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 17-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard)))
```


## BIO2041 Vertebrate Biology Leaderboard
```{r}
leaderboard2041 <- filter(leaderboardall, group == "BIO2041")
Ben <- filter(leaderboardall, Player_name== "Ben Rowland")
Mark <- filter(leaderboardall, Player_name== "Mark Shirley")
Richard <- filter(leaderboardall, Player_name== "Richard Bevan")
# add relevant staff members 
leaderboard2041 <- bind_rows(leaderboard2041,Richard,Ben, Mark ) %>% select(-group)
leaderboard2041$Player_name <- str_replace(leaderboard2041$Player_name,"BIO2041", "")
leaderboard2041 %>% arrange(-Total) %>% 
knitr::kable()
```


```{r 2023winners2041}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
staff <- outcomes %>% filter(players=="Ben Rowland"|players=="Mark Shirley"| players=="Richard Bevan")

predicted_winners <- outcomes %>% filter(group == "BIO2041") %>% 
  bind_rows(., staff) %>% 
  select(winners) %>%
  group_by(winners) %>% 
  count() %>% 
  rename("num"= "n") %>% 
  arrange(-num) %>% 
  ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```

```{r, eval = FALSE}
df <- leaderboard2041
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 15-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall2041 <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall2029, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall2029 %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard2029),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard2029)))
```

## MEP Leaderboard
```{r}
leaderboard <- filter(leaderboardall, is.na(group))
#leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MEP", "")
leaderboard %>% arrange(-Total) %>% 
  select(-group) %>% 
knitr::kable()
```

```{r 2023winnersMEP}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
#outcomes$group <- str_replace(outcomes$group, "MMM", "MEP")
predicted_winners <- outcomes %>% filter(is.na(group)) %>% 
  select(winners) %>%
  group_by(winners) %>% 
  count() %>% 
  rename("num"= "n") %>% 
  arrange(-num) %>% 
  ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```


```{r, eval= FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 16-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankallmep<- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankallmep, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankallmep %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard)))
```

```{r eval= FALSE}
 pl_dat <- leaderboard %>%mutate(r1= Round_1) %>%  mutate(r2 = Round_1+Round_2) %>% mutate(r3 = r2+Sweet_16) %>% 
  mutate(r4=r3+Elite_8) %>% mutate(r5=r4 + Final_roar) %>% mutate(r6=Total) %>% select(Player_name, r1:r6)

pl_dat$r0 <- 0
pl_dat <- pl_dat %>% 
  pivot_longer(cols = r1:r0,names_to = "round", values_to = "points")


ggplot(data=pl_dat, aes(x=round, y=points, group = Player_name, colour = Player_name)) + 
    geom_line() +
  ggtitle("Point accumulation")

# arrange legend 

```

# 2022

## 2022 Leaderboard


```{r}
library(tidyverse)
library(here)
library(plotly)
library(readxl)


fils <- list.files (path = "2022", pattern="*.csv") 

# create a list in which to store entries
fl <- list()

#loop through files and add to the list
#using here means that I can easily read files in a subdirectory
for (k in 1:length(fils)){
  entry <- read.csv(here("2022",fils[k]))
  
  fl[[k]] <- as.vector(entry[,1])
}


# format the data creating a single vector with all answers from each player
# ans <- list()
# for(j in 1:length(fl)){
#   
#   mast <- fl[[j]]
#   wild <- mast %>%  select("...2") %>%drop_na() %>% rename(Wildcard = "...2")
#   a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
#                                                         Round_2 = Round_2...6,
#                                                         Round_3 = Round_3...7,
#                                                         Round_4 = Round_4...8,
#                                                         Round_5 = Round_5...9)
#   b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
#                                                           Round_2 = Round_2...14, 
#                                                           Round_3 = Round_3...13, 
#                                                           Round_4 = Round_4...12, 
#                                                           Round_5 = Round_5...11 )
#   
#   cc <- bind_rows(a,b)
#   r1 <- cc %>% select(Round_1) %>% drop_na()
#   r1 <- r1$Round_1
#   
#   r2 <- cc %>% select(Round_2) %>% drop_na()
#   r2 <- r2$Round_2
#   
#   r3 <- cc %>% select(Round_3) %>% drop_na()
#   r3 <- r3$Round_3
#   
#   r4 <- cc %>% select(Round_4) %>% drop_na()
#   r4 <- r4$Round_4
#   
#   r5 <- cc %>% select(Round_5) %>% drop_na()
#   r5 <- r5$Round_5
#   
#   wild <- wild$Wildcard
#   
#   fin <- mast %>% select(Final) %>% drop_na()
#   fin <- fin$Final
#   
#   out <- c(wild,r1,r2,r3,r4,r5,fin)
#   # store the vector in a list
#   ans[[j]] <- out
# }


### counting points 

######################## generating the answers 
mast <- read_excel(here("2022_results.xlsx"))
wild <- mast %>%  select("...2") %>%drop_na() %>% rename(Wildcard = "...2")
a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                      Round_2 = Round_2...6,
                                                      Round_3 = Round_3...7,
                                                      Round_4 = Round_4...8,
                                                      Round_5 = Round_5...9)
b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                        Round_2 = Round_2...14, 
                                                        Round_3 = Round_3...13, 
                                                        Round_4 = Round_4...12, 
                                                        Round_5 = Round_5...11 )
cc <- bind_rows(a,b)
r1 <- cc %>% select(Round_1) %>% drop_na()
r1 <- r1$Round_1

r2 <- cc %>% select(Round_2) %>% drop_na()
r2 <- r2$Round_2

r3 <- cc %>% select(Round_3) %>% drop_na()
r3 <- r3$Round_3

r4 <- cc %>% select(Round_4) %>% drop_na()
r4 <- r4$Round_4

r5 <- cc %>% select(Round_5) %>% drop_na()
r5 <- r5$Round_5

wild <- wild$Wildcard

fin <- mast %>% select(Final) %>% drop_na()
fin <- fin$Final

master <- c(wild,r1,r2,r3,r4,r5,fin)

# 1 = same 0 = different -- ony get points for answers that are the same as the master


# check where in the list the master file is 
#ans


# create a list of players names

players <- sub("_", " ", fils) %>% sub(".csv", "",.)
#players <- sub(".csv", "", players)

### check answers 

scorelist <- list()
for (i in 1: length(players)){
  
  ## this checks answers for each round 
  playa <- fl[[i]]
  mstr <- master
  #mstr <- ans[[1]]
  rnd1 <- playa[1:33] == mstr[1:33]
  rnd2 <- playa[34:49] == mstr[34:49]
  sweet <- playa[50:57]==mstr[50:57]
  elite <- playa[58:61]==mstr[58:61]
  roar <- playa[62:63]==mstr[62:63]
  win <- playa[64]==mstr[64]
  
  
  pts <- c(sum(rnd1),sum(rnd2*2), sum(sweet*3), sum(elite*5), sum(roar*8),sum(win*13))
  scorelist[[i]] <- pts
}

leaderboard <- as.data.frame(do.call(rbind, scorelist))
colnames(leaderboard) <- c("Round_1", "Round_2", "Sweet_16", "Elite_8", "Final_roar", "Winner")
#tardy <- c(1,0,0,0,0,0,1,0,0,1,1,1,0)


leaderboard$Player_name <- players
#leaderboard$Tardigrade_tally <- tardy



leaderboard <- leaderboard %>% 
  select(Player_name,  Round_1, Round_2, Sweet_16, Elite_8, 
         Final_roar, Winner) %>% 
  mutate(Total= rowSums(.[2:7])) %>%
  arrange(desc(Total))

# creating a plot with cumulative points 
#create sum column for after each round 
pl_dat <- leaderboard %>% mutate(r2 = Round_1+Round_2) %>% mutate(r3 = r2+Sweet_16)

pl_dat <- leaderboard %>% 
  mutate(r1 =rowSums(.[2:3])) %>% 
  mutate(r2 =rowSums(.[2:4])) %>% 
  mutate(r3 =rowSums(.[2:5])) %>%
  mutate(r4 =rowSums(.[2:6])) %>% 
  mutate(r5 =rowSums(.[2:7])) %>% 
  rename(r6 = Total) %>%
  select(Player_name, r1, r2, r3, r4, r5, r6) %>% 
  gather("Round", "points", 2:7)

pl_dat$Round <- gsub("r","",pl_dat$Round)

p <- ggplot(data=pl_dat, aes(x=Round, y=points, group = Player_name, colour = Player_name)) + 
  geom_line() +
  ggtitle("Point accumulation")
#+ scale_colour_manual(values=cols) 
  
# working out the most popular winner
winners <- vector(length = length(players))

for (i in 1:length(winners)){

  this_one <- fl[[i]]
  winners[i] <- this_one[64]
  
}




outcomes <- data.frame(players, winners)
# creating line breaks for plotting 
levels(outcomes$winners) <- gsub(" ", "\n",levels(outcomes$winners))

outcomes <- separate(outcomes, players, into = c("players", "group"), sep = "_(?=[^_]+$)")
```

```{r}
#leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MMM", " MEP")
# create list of MEP, BIO2029 and BIO3036 players
leaderboardall <- separate(leaderboard, Player_name, into = c("Player_name", "group"), sep = "_(?=[^_]+$)")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MEP", "")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"_", "")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"BIO2029", "")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"BIO3036", "")
leaderboard %>% arrange(-Total) %>% 
knitr::kable()
```


```{r 2022winners}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
predicted_winners <- outcomes %>% select(winners) %>% group_by(winners) %>% count() %>% rename("num"= "n") %>% arrange(-num) %>% ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```


```{r, 2022bumpall, eval = FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 17-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard)))
```


## BIO2029 Vertebrate Biology Leaderboard
```{r}
leaderboard2029 <- filter(leaderboardall, group == "BIO2029")
 Aileen <- filter(leaderboardall, Player_name== "Aileen Mill")
 Mark <- filter(leaderboardall, Player_name== "Mark Shirley")
Richard <- filter(leaderboardall, Player_name== "Richard Bevan")
# add relevant staff members 
leaderboard2029 <- bind_rows(leaderboard2029,Richard,Aileen, Mark ) %>% select(-group)
leaderboard2029$Player_name <- str_replace(leaderboard2029$Player_name,"BIO2029", "")
leaderboard2029 %>% arrange(-Total) %>% 
knitr::kable()
```


```{r 2022winners2029}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
staff <- outcomes %>% filter(players=="Aileen Mill"|players=="Mark Shirley"| players=="Richard Bevan")

predicted_winners <- outcomes %>% filter(group == "BIO2029") %>% 
  bind_rows(., staff) %>% 
  select(winners) %>%
  group_by(winners) %>% 
  count() %>% 
  rename("num"= "n") %>% 
  arrange(-num) %>% 
  ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```

```{r, eval = FALSE}
df <- leaderboard2029
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 15-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 15-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall2029 <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall2029, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall2029 %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard2029),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard2029)))
```

## MEP Leaderboard
```{r}
leaderboard <- filter(leaderboardall, is.na(group))
#leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MEP", "")
leaderboard %>% arrange(-Total) %>% 
  select(-group) %>% 
knitr::kable()
```

```{r 2022winnersMEP}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
#outcomes$group <- str_replace(outcomes$group, "MMM", "MEP")
predicted_winners <- outcomes %>% filter(is.na(group)) %>% 
  select(winners) %>%
  group_by(winners) %>% 
  count() %>% 
  rename("num"= "n") %>% 
  arrange(-num) %>% 
  ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```


```{r, eval= FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 16-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 16-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankallmep<- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankallmep, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankallmep %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard)))
```

```{r eval= FALSE}
 pl_dat <- leaderboard %>%mutate(r1= Round_1) %>%  mutate(r2 = Round_1+Round_2) %>% mutate(r3 = r2+Sweet_16) %>% 
  mutate(r4=r3+Elite_8) %>% mutate(r5=r4 + Final_roar) %>% mutate(r6=Total) %>% select(Player_name, r1:r6)

pl_dat$r0 <- 0
pl_dat <- pl_dat %>% 
  pivot_longer(cols = r1:r0,names_to = "round", values_to = "points")


ggplot(data=pl_dat, aes(x=round, y=points, group = Player_name, colour = Player_name)) + 
    geom_line() +
  ggtitle("Point accumulation")

# arrange legend 

```

# 2021

<!-- <img src = "https://libapps.s3.amazonaws.com/accounts/46633/images/March_Mammal_Madness_2021_Bracket_v1_2_English.png" class = "center"> -->

## 2021 Leaderboard

```{r, eval=FALSE}
library(tidyverse)
library(here)
library(plotly)
library(readxl)


fils <- list.files (path = "2021", pattern="*.xlsx") 

# create a list in which to store entries
fl <- list()

#loop through files and add to the list
#using here means that I can easily read files in a subdirectory
for (k in 1:length(fils)){
  fl[[k]] <- read_excel(here("2021",fils[k]))
}


# format the data creating a single vector with all answers from each player
ans <- list()
for(j in 1:length(fl)){
  
  mast <- fl[[j]]
  wild <- mast %>%  select("...2") %>%drop_na() %>% rename(Wildcard = "...2")
  a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                        Round_2 = Round_2...6,
                                                        Round_3 = Round_3...7,
                                                        Round_4 = Round_4...8,
                                                        Round_5 = Round_5...9)
  b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                          Round_2 = Round_2...14, 
                                                          Round_3 = Round_3...13, 
                                                          Round_4 = Round_4...12, 
                                                          Round_5 = Round_5...11 )
  
  cc <- bind_rows(a,b)
  r1 <- cc %>% select(Round_1) %>% drop_na()
  r1 <- r1$Round_1
  
  r2 <- cc %>% select(Round_2) %>% drop_na()
  r2 <- r2$Round_2
  
  r3 <- cc %>% select(Round_3) %>% drop_na()
  r3 <- r3$Round_3
  
  r4 <- cc %>% select(Round_4) %>% drop_na()
  r4 <- r4$Round_4
  
  r5 <- cc %>% select(Round_5) %>% drop_na()
  r5 <- r5$Round_5
  
  wild <- wild$Wildcard
  
  fin <- mast %>% select(Final) %>% drop_na()
  fin <- fin$Final
  
  out <- c(wild,r1,r2,r3,r4,r5,fin)
  # store the vector in a list
  ans[[j]] <- out
}


### counting points 

######################## generating the answers 
mast <- read_excel(here("2021_results.xlsx"))
wild <- mast %>%  select("...2") %>%drop_na() %>% rename(Wildcard = "...2")
a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                      Round_2 = Round_2...6,
                                                      Round_3 = Round_3...7,
                                                      Round_4 = Round_4...8,
                                                      Round_5 = Round_5...9)
b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                        Round_2 = Round_2...14, 
                                                        Round_3 = Round_3...13, 
                                                        Round_4 = Round_4...12, 
                                                        Round_5 = Round_5...11 )
cc <- bind_rows(a,b)
r1 <- cc %>% select(Round_1) %>% drop_na()
r1 <- r1$Round_1

r2 <- cc %>% select(Round_2) %>% drop_na()
r2 <- r2$Round_2

r3 <- cc %>% select(Round_3) %>% drop_na()
r3 <- r3$Round_3

r4 <- cc %>% select(Round_4) %>% drop_na()
r4 <- r4$Round_4

r5 <- cc %>% select(Round_5) %>% drop_na()
r5 <- r5$Round_5

wild <- wild$Wildcard

fin <- mast %>% select(Final) %>% drop_na()
fin <- fin$Final

master <- c(wild,r1,r2,r3,r4,r5,fin)

# 1 = same 0 = different -- ony get points for answers that are the same as the master


# check where in the list the master file is 
#ans


# create a list of players names

players <- sub("_", " ", fils) %>% sub(".xlsx", "",.)
#players <- sub(".csv", "", players)

### check answers 

scorelist <- list()
for (i in 1: length(players)){
  
  ## this checks answers for each round 
  playa <- ans[[i]]
  mstr <- master
  #mstr <- ans[[1]]
  rnd1 <- playa[1:33] == mstr[1:33]
  rnd2 <- playa[34:49] == mstr[34:49]
  sweet <- playa[50:57]==mstr[50:57]
  elite <- playa[58:61]==mstr[58:61]
  roar <- playa[62:63]==mstr[62:63]
  win <- playa[64]==mstr[64]
  
  
  pts <- c(sum(rnd1),sum(rnd2*2), sum(sweet*3), sum(elite*5), sum(roar*8),sum(win*13))
  scorelist[[i]] <- pts
}

leaderboard <- as.data.frame(do.call(rbind, scorelist))
colnames(leaderboard) <- c("Round_1", "Round_2", "Sweet_16", "Elite_8", "Final_roar", "Winner")
#tardy <- c(1,0,0,0,0,0,1,0,0,1,1,1,0)


leaderboard$Player_name <- players
#leaderboard$Tardigrade_tally <- tardy



leaderboard <- leaderboard %>% 
  select(Player_name,  Round_1, Round_2, Sweet_16, Elite_8, 
         Final_roar, Winner) %>% 
  mutate(Total= rowSums(.[2:7])) %>%
  arrange(desc(Total))

# creating a plot with cumulative points 
#create sum column for after each round 
pl_dat <- leaderboard %>% mutate(r2 = Round_1+Round_2) %>% mutate(r3 = r2+Sweet_16)

pl_dat <- leaderboard %>% 
  mutate(r1 =rowSums(.[2:3])) %>% 
  mutate(r2 =rowSums(.[2:4])) %>% 
  mutate(r3 =rowSums(.[2:5])) %>%
  mutate(r4 =rowSums(.[2:6])) %>% 
  mutate(r5 =rowSums(.[2:7])) %>% 
  rename(r6 = Total) %>%
  select(Player_name, r1, r2, r3, r4, r5, r6) %>% 
  gather("Round", "points", 2:7)

pl_dat$Round <- gsub("r","",pl_dat$Round)

p <- ggplot(data=pl_dat, aes(x=Round, y=points, group = Player_name, colour = Player_name)) + 
  geom_line() +
  ggtitle("Point accumulation")
#+ scale_colour_manual(values=cols) 
  
# working out the most popular winner
winners <- vector(length = length(players))

for (i in 1:length(winners)){

  this_one <- ans[[i]]
  winners[i] <- this_one[64]
  
}




outcomes <- data.frame(players, winners)
# creating line breaks for plotting 
levels(outcomes$winners) <- gsub(" ", "\n",levels(outcomes$winners))

outcomes <- separate(outcomes, players, into = c("players", "group"), sep = " ")
```

```{r, eval=FALSE}
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MMM", " MEP")
# create list of MEP, BIO2029 and BIO3036 players
leaderboardall <- separate(leaderboard, Player_name, into = c("Player_name", "group"), sep = " ")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MEP", "")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"BIO2029", "")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"BIO3036", "")
leaderboard %>% arrange(-Total) %>% 
knitr::kable()
```

```{r, 2021bumpall, eval = FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 17-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard)))
```


```{r 2021winners, eval=FALSE}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
predicted_winners <- outcomes %>% select(winners) %>% group_by(winners) %>% count() %>% rename("num"= "n") %>% arrange(-num) %>% ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```



## BIO2029 Vertebrate Biology Leaderboard
```{r, eval=FALSE}
leaderboard2029 <- filter(leaderboardall, group == "BIO2029")
Aileen <- filter(leaderboardall, Player_name== "AileenMill")
leaderboard2029 <- bind_rows(leaderboard2029,Aileen) %>% select(-group)
leaderboard2029$Player_name <- str_replace(leaderboard2029$Player_name,"BIO2029", "")
leaderboard2029 %>% arrange(-Total) %>% 
knitr::kable()
```
 

```{r 2021winners2029, eval=FALSE}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
predicted_winners <- outcomes %>% filter(group == "BIO2029") %>% 
  select(winners) %>%
  group_by(winners) %>% 
  count() %>% 
  rename("num"= "n") %>% 
  arrange(-num) %>% 
  ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```

```{r, eval=FALSE}
df <- leaderboard2029
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 17-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall2029 <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall2029, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall2029 %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard2029),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard2029)))
```


<!-- - bump diagram  -->
## BIO3036 Mammal Surveying Leaderboard

```{r, eval=FALSE}
leaderboard3036 <- filter(leaderboardall, group == "BIO3036") 

#select(-group)
leaderboard3036$Player_name <- str_replace(leaderboard3036$Player_name,"BIO3036", "")
leaderboard3036 <- bind_rows(leaderboard3036,Aileen) %>% select(-group)

leaderboard3036 %>% arrange(-Total) %>% 
knitr::kable()
```

```{r 2021winners3036, eval=FALSE}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
predicted_winners <- outcomes %>% filter(group == "BIO3036") %>% 
  select(winners) %>%
  group_by(winners) %>% 
  count() %>% 
  rename("num"= "n") %>% 
  arrange(-num) %>% 
  ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```

```{r, eval=FALSE}
df <- leaderboard3036
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 17-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall3036 <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall3036, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall3036 %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard3036),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard3036)))
```


## MEP Leaderboard
```{r, eval=FALSE}
leaderboard <- filter(leaderboardall, group == "MEP"| group =="")
leaderboard$Player_name <- str_replace(leaderboard$Player_name,"MEP", "")
leaderboard %>% arrange(-Total) %>% 
knitr::kable()
```

```{r 2021winnersMEP, eval=FALSE}
library(viridis)

outcomes$winners <- as.character(outcomes$winners)
outcomes$group <- str_replace(outcomes$group, "MMM", "MEP")
predicted_winners <- outcomes %>% filter(group == "MEP"| group == "MMM") %>% 
  select(winners) %>%
  group_by(winners) %>% 
  count() %>% 
  rename("num"= "n") %>% 
  arrange(-num) %>% 
  ungroup()


ggplot2::ggplot(data = predicted_winners, aes(x = reorder(winners,num), y = num)) + 
  geom_bar( stat = "identity", aes(fill = winners))+
   scale_fill_manual(values = viridis(n = nrow(predicted_winners)))+
  ggtitle("Predicted winners")+
  theme_classic()+
  coord_flip()+
 theme(legend.position = "none")+
  labs(x = "")


```


```{r, eval=FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 17-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankallmep<- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankallmep, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankallmep %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, nrow(leaderboard),1))+
  scale_color_manual(values = viridis(n = nrow(leaderboard)))
```

# 2020 

<!-- <img src= "https://1.bp.blogspot.com/-I1rQLndyvd8/Xlji_FnnCNI/AAAAAAAAMus/0DfvLvTwEU0zRtG5KrAYdcWubh77nx6kQCNcBGAsYHQ/s640/Mammal%2BMarch%2BMadness%2B2020%2BBracket%2Bv1_0_English.png" class= "center"> -->
## MEP Leaderboard

```{r 2020mep, eval=FALSE}
library(tidyverse)
library(here)
library(plotly)
library(readxl)


fils <- list.files (path = "2020MEP", pattern="*.xlsx") 

# create a list in which to store entries
fl <- list()

#loop through files and add to the list
#using here means that I can easily read files in a subdirectory
for (k in 1:length(fils)){
  fl[[k]] <- read_excel(here("2020MEP",fils[k]))
}


# format the data creating a single vector with all answers from each player
ans <- list()
for(j in 1:length(fl)){
  
  mast <- fl[[j]]
  wild <- mast %>%  select(`Wild Card`) %>%filter(!is.na(`Wild Card`)) %>% rename(Wildcard = `Wild Card`)
  a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                        Round_2 = Round_2...6,
                                                        Round_3 = Round_3...7,
                                                        Round_4 = Round_4...8,
                                                        Round_5 = Round_5...9)
  b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                          Round_2 = Round_2...14, 
                                                          Round_3 = Round_3...13, 
                                                          Round_4 = Round_4...12, 
                                                          Round_5 = Round_5...11 )
  
  cc <- bind_rows(a,b)
  r1 <- cc %>% select(Round_1) %>% filter(!is.na(Round_1))
  r1 <- r1$Round_1
  
  r2 <- cc %>% select(Round_2) %>% filter(!is.na(Round_2))
  r2 <- r2$Round_2
  
  r3 <- cc %>% select(Round_3) %>% filter(!is.na(Round_3))
  r3 <- r3$Round_3
  
  r4 <- cc %>% select(Round_4) %>% filter(!is.na(Round_4))
  r4 <- r4$Round_4
  
  r5 <- cc %>% select(Round_5) %>% filter(!is.na(Round_5))
  r5 <- r5$Round_5
  
  wild <- wild$Wildcard
  
  fin <- mast %>% select(Final) %>% filter(!is.na(Final))
  fin <- fin$Final
  
  out <- c(wild,r1,r2,r3,r4,r5,fin)
  # store the vector in a list
  ans[[j]] <- out
}


### counting points 

######################## generating the answers 
mast <- read_excel(here("2020AAnswers.xlsx"))
wild <- mast %>%  select(`Wild Card`) %>%filter(!is.na(`Wild Card`)) %>% rename(Wildcard = `Wild Card`)
a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                      Round_2 = Round_2...6,
                                                      Round_3 = Round_3...7,
                                                      Round_4 = Round_4...8,
                                                      Round_5 = Round_5...9)
b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                        Round_2 = Round_2...14, 
                                                        Round_3 = Round_3...13, 
                                                        Round_4 = Round_4...12, 
                                                        Round_5 = Round_5...11 )
cc <- bind_rows(a,b)
r1 <- cc %>% select(Round_1) %>% filter(!is.na(Round_1))
r1 <- r1$Round_1

r2 <- cc %>% select(Round_2) %>% filter(!is.na(Round_2))
r2 <- r2$Round_2

r3 <- cc %>% select(Round_3) %>% filter(!is.na(Round_3))
r3 <- r3$Round_3

r4 <- cc %>% select(Round_4) %>% filter(!is.na(Round_4))
r4 <- r4$Round_4

r5 <- cc %>% select(Round_5) %>% filter(!is.na(Round_5))
r5 <- r5$Round_5

wild <- wild$Wildcard

fin <- mast %>% select(Final) %>% filter(!is.na(Final))
fin <- fin$Final

master <- c(wild,r1,r2,r3,r4,r5,fin)

# 1 = same 0 = different -- ony get points for answers that are the same as the master


# check where in the list the master file is 
#ans


# create a list of players names

players <- sub("_", " ", fils) %>% sub(".xlsx", "",.)
#players <- sub(".csv", "", players)

### check answers 

scorelist <- list()
for (i in 1: length(players)){
  
  ## this checks answers for each round 
  playa <- ans[[i]]
  mstr <- master
  #mstr <- ans[[1]]
  rnd1 <- playa[1:33] == mstr[1:33]
  rnd2 <- playa[34:49] == mstr[34:49]
  sweet <- playa[50:57]==mstr[50:57]
  elite <- playa[58:61]==mstr[58:61]
  roar <- playa[62:63]==mstr[62:63]
  win <- playa[64]==mstr[64]
  
  
  pts <- c(sum(rnd1),sum(rnd2*2), sum(sweet*3), sum(elite*5), sum(roar*8),sum(win*13))
  scorelist[[i]] <- pts
}

leaderboard <- as.data.frame(do.call(rbind, scorelist))
colnames(leaderboard) <- c("Round_1", "Round_2", "Sweet_16", "Elite_8", "Final_roar", "Winner")
#tardy <- c(1,0,0,0,0,0,1,0,0,1,1,1,0)


leaderboard$Player_name <- players
#leaderboard$Tardigrade_tally <- tardy



leaderboard <- leaderboard %>% 
  select(Player_name,  Round_1, Round_2, Sweet_16, Elite_8, 
         Final_roar, Winner) %>% 
  mutate(Total= rowSums(.[2:7])) %>%
  arrange(desc(Total))

# creating a plot with cumulative points 
#create sum column for after each round 
pl_dat <- leaderboard %>% mutate(r2 = Round_1+Round_2) %>% mutate(r3 = r2+Sweet_16)

pl_dat <- leaderboard %>% 
  mutate(r1 =rowSums(.[2:3])) %>% 
  mutate(r2 =rowSums(.[2:4])) %>% 
  mutate(r3 =rowSums(.[2:5])) %>%
  mutate(r4 =rowSums(.[2:6])) %>% 
  mutate(r5 =rowSums(.[2:7])) %>% 
  rename(r6 = Total) %>%
  select(Player_name, r1, r2, r3, r4, r5, r6) %>% 
  gather("Round", "points", 2:7)

pl_dat$Round <- gsub("r","",pl_dat$Round)

p <- ggplot(data=pl_dat, aes(x=Round, y=points, group = Player_name, colour = Player_name)) + 
  geom_line() +
  ggtitle("Point accumulation")
#+ scale_colour_manual(values=cols) 
  
# working out the most popular winner
winners <- vector(length = length(players))

for (i in 1:length(winners)){

  this_one <- ans[[i]]
  winners[i] <- this_one[64]
  
}

outcomes <- data.frame(players, winners)
# creating line breaks for plotting 
levels(outcomes$winners) <- gsub(" ", "\n",levels(outcomes$winners))


```



```{r eval=FALSE}
knitr::kable(leaderboard)
```

```{r eval=FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "max")) %>% 
  mutate(Ranking = 17-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "max")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)
# 
round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 17-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)

# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall <- bind_rows(round1, round2, round3,round4,round5,round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, 16,1))+
  scale_color_manual(values = viridis(n = 16))
```

```{r, eval=FALSE}
#what is the most popular winner 

ggplot2::ggplot(data = outcomes) + 
  geom_bar(mapping = aes(x = winners,  fill = players))+
  ggtitle("Predicted winners")+
  theme_classic()



```



## 2020 BIO2029 Leaderboard
```{r 2020bio2029, eval=FALSE}
library(tidyverse)
library(here)
library(plotly)
library(readxl)



fils <- list.files (path = "2020BIO2029", pattern="*.xlsx") 

# create a list in which to store entries
fl <- list()

#loop through files and add to the list
#using here means that I can easily read files in a subdirectory
for (k in 1:length(fils)){
  fl[[k]] <- read_excel(here("2020BIO2029",fils[k]))
}


# format the data creating a single vector with all answers from each player
ans <- list()
for(j in 1:length(fl)){
  
  mast <- fl[[j]]
  wild <- mast %>%  select(`Wild Card`) %>%filter(!is.na(`Wild Card`)) %>% rename(Wildcard = `Wild Card`)
  a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                        Round_2 = Round_2...6,
                                                        Round_3 = Round_3...7,
                                                        Round_4 = Round_4...8,
                                                        Round_5 = Round_5...9)
  b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                          Round_2 = Round_2...14, 
                                                          Round_3 = Round_3...13, 
                                                          Round_4 = Round_4...12, 
                                                          Round_5 = Round_5...11 )
  
  cc <- bind_rows(a,b)
  r1 <- cc %>% select(Round_1) %>% filter(!is.na(Round_1))
  r1 <- r1$Round_1
  
  r2 <- cc %>% select(Round_2) %>% filter(!is.na(Round_2))
  r2 <- r2$Round_2
  
  r3 <- cc %>% select(Round_3) %>% filter(!is.na(Round_3))
  r3 <- r3$Round_3
  
  r4 <- cc %>% select(Round_4) %>% filter(!is.na(Round_4))
  r4 <- r4$Round_4
  
  r5 <- cc %>% select(Round_5) %>% filter(!is.na(Round_5))
  r5 <- r5$Round_5
  
  wild <- wild$Wildcard
  
  fin <- mast %>% select(Final) %>% filter(!is.na(Final))
  fin <- fin$Final
  
  out <- c(wild,r1,r2,r3,r4,r5,fin)
  # store the vector in a list
  ans[[j]] <- out
}


### counting points 

######################## generating the answers 
mast <- read_excel(here("2020AAnswers.xlsx"))
wild <- mast %>%  select(`Wild Card`) %>%filter(!is.na(`Wild Card`)) %>% rename(Wildcard = `Wild Card`)
a <- select(mast, Round_1...5:Round_5...9) %>% rename(Round_1 = Round_1...5,
                                                      Round_2 = Round_2...6,
                                                      Round_3 = Round_3...7,
                                                      Round_4 = Round_4...8,
                                                      Round_5 = Round_5...9)
b <- select(mast, Round_5...11:Round_1...15) %>% rename(Round_1 = Round_1...15,
                                                        Round_2 = Round_2...14, 
                                                        Round_3 = Round_3...13, 
                                                        Round_4 = Round_4...12, 
                                                        Round_5 = Round_5...11 )
cc <- bind_rows(a,b)
r1 <- cc %>% select(Round_1) %>% filter(!is.na(Round_1))
r1 <- r1$Round_1

r2 <- cc %>% select(Round_2) %>% filter(!is.na(Round_2))
r2 <- r2$Round_2

r3 <- cc %>% select(Round_3) %>% filter(!is.na(Round_3))
r3 <- r3$Round_3

r4 <- cc %>% select(Round_4) %>% filter(!is.na(Round_4))
r4 <- r4$Round_4

r5 <- cc %>% select(Round_5) %>% filter(!is.na(Round_5))
r5 <- r5$Round_5

wild <- wild$Wildcard

fin <- mast %>% select(Final) %>% filter(!is.na(Final))
fin <- fin$Final

master <- c(wild,r1,r2,r3,r4,r5,fin)

# 1 = same 0 = different -- ony get points for answers that are the same as the master


# check where in the list the master file is 
#ans


# create a list of players names

players <- sub("_", " ", fils) %>% sub(".xlsx", "",.)
#players <- sub(".csv", "", players)

### check answers 

scorelist <- list()
for (i in 1: length(players)){
  
  ## this checks answers for each round 
  playa <- ans[[i]]
  mstr <- master
  #mstr <- ans[[1]]
  rnd1 <- playa[1:33] == mstr[1:33]
  rnd2 <- playa[34:49] == mstr[34:49]
  sweet <- playa[50:57]==mstr[50:57]
  elite <- playa[58:61]==mstr[58:61]
  roar <- playa[62:63]==mstr[62:63]
  win <- playa[64]==mstr[64]
  
  
  pts <- c(sum(rnd1),sum(rnd2*2), sum(sweet*3), sum(elite*5), sum(roar*8),sum(win*13))
  scorelist[[i]] <- pts
}

leaderboard <- as.data.frame(do.call(rbind, scorelist))
colnames(leaderboard) <- c("Round_1", "Round_2", "Sweet_16", "Elite_8", "Final_roar", "Winner")
#tardy <- c(1,0,0,0,0,0,1,0,0,1,1,1,0)


leaderboard$Player_name <- players
#leaderboard$Tardigrade_tally <- tardy



leaderboard <- leaderboard %>% 
  select(Player_name,  Round_1, Round_2, Sweet_16, Elite_8, 
         Final_roar, Winner) %>% 
  mutate(Total= rowSums(.[2:7])) %>%
  arrange(desc(Total))

# creating a plot with cumulative points 
#create sum column for after each round 
pl_dat <- leaderboard %>% mutate(r2 = Round_1+Round_2) %>% mutate(r3 = r2+Sweet_16)

pl_dat <- leaderboard %>% 
  mutate(r1 =rowSums(.[2:3])) %>% 
  mutate(r2 =rowSums(.[2:4])) %>% 
  mutate(r3 =rowSums(.[2:5])) %>%
  mutate(r4 =rowSums(.[2:6])) %>% 
  mutate(r5 =rowSums(.[2:7])) %>% 
  rename(r6 = Total) %>%
  select(Player_name, r1, r2, r3, r4, r5, r6) %>% 
  gather("Round", "points", 2:7)

pl_dat$Round <- gsub("r","",pl_dat$Round)

p <- ggplot(data=pl_dat, aes(x=Round, y=points, group = Player_name, colour = Player_name)) + 
  geom_line() +
  ggtitle("Point accumulation")
#+ scale_colour_manual(values=cols) 
  
# working out the most popular winner
winners <- vector(length = length(players))

for (i in 1:length(winners)){

  this_one <- ans[[i]]
  winners[i] <- this_one[64]
  
}

outcomes <- data.frame(players, winners)
# creating line breaks for plotting 
levels(outcomes$winners) <- gsub(" ", "\n",levels(outcomes$winners))


```


```{r eval=FALSE}
df <- leaderboard
round1 <- select(df, Player_name, Round_1) %>% arrange(desc(Round_1)) %>% 
  mutate(Ranking = rank(Round_1, ties.method = "min")) %>% 
  mutate(Ranking = 13-Ranking) %>% 
  mutate(Round = 1) %>% 
  select(-Round_1)

round2 <- select(df, Player_name,Round_1:Round_2) %>%
  mutate(Round_2 = Round_1 + Round_2) %>%
  arrange(desc(Round_2)) %>%
  mutate(Ranking = rank(Round_2, ties.method = "min")) %>%
  mutate(Ranking = 13-Ranking) %>%
  mutate(Round = 2) %>%
  select(- Round_1, -Round_2)

round3 <- select(df, Player_name, Round_1, Round_2,Sweet_16) %>%
  mutate(Sweet_16 = Round_1 + Round_2 + Sweet_16) %>%
  arrange(desc(Sweet_16)) %>%
  mutate(Ranking = rank(Sweet_16, ties.method = "min")) %>%
  mutate(Ranking = 13-Ranking) %>%
  mutate(Round = 3) %>%
  select(Player_name, Ranking, Round)

round4 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8) %>%
  mutate(Elite_8 = Round_1 + Round_2 + Sweet_16 + Elite_8) %>%
  arrange(desc(Elite_8)) %>%
  mutate(Ranking = rank(Elite_8, ties.method = "min")) %>%
  mutate(Ranking = 13-Ranking) %>%
  mutate(Round = 4) %>%
  select(Player_name, Ranking, Round)

round5 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar) %>%
  mutate(Final_roar = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar) %>%
  arrange(desc(Final_roar)) %>%
  mutate(Ranking = rank(Final_roar, ties.method = "min")) %>%
  mutate(Ranking = 13-Ranking) %>%
  mutate(Round = 5) %>%
  select(Player_name, Ranking, Round)

round6 <- select(df, Player_name, Round_1, Round_2, Sweet_16,Elite_8, Final_roar, Winner) %>%
  mutate(Winner = Round_1 + Round_2 + Sweet_16 + Elite_8 + Final_roar+ Winner) %>%
  arrange(desc(Winner)) %>%
  mutate(Ranking = rank(Winner, ties.method = "min")) %>%
  mutate(Ranking = 13-Ranking) %>%
  mutate(Round = 6) %>%
  select(Player_name, Ranking, Round)
# 
# rankall <- bind_rows(round1,round2,round3,round4,round5, round6)

rankall <- bind_rows(round1, round2,round3,round4,round5, round6)
source(here::here("geom_bump.R"))
library(cowplot)
library(viridis)
library(ggrepel)
ggplot(rankall, aes(Round, Ranking, color = Player_name)) +
  geom_point(size = 7) +
  #geom_text(data = rankall %>% filter(Round == min(Round)),
           # aes(x = 0.5, label = Player_name), position=position_jitter(height=01),size = 4) +
  geom_text_repel(data = rankall %>% filter(Round == max(Round)),
            aes(x = Round, label = Player_name),size = 4, hjust = -0.5) +
  geom_bump(aes(smooth = 8), size = 2) +
  scale_x_continuous(limits = c(1, 7),
                     breaks = seq(1, 6, 1)) +
  theme_minimal_grid(font_size = 14, line_size = 0) +
  theme(legend.position = "none",
        panel.grid.major = element_blank()) +
  labs(y = "RANK",
       x = "Round") +
  scale_y_reverse(breaks = seq(1, 16,1))+
  scale_color_manual(values = viridis(n = 16))
```





```{r eval=FALSE}
knitr::kable(leaderboard)
```


```{r, eval=FALSE}
#what is the most popular winner 

ggplot2::ggplot(data = outcomes) + 
  geom_bar(mapping = aes(x = winners,  fill = players))+
  ggtitle("Predicted winners")+
  theme_classic()



```