078: ACC Road records

cbbplotR
gt
geom_tile
Published

February 22, 2024

Data

Code 
library(cbbplotR)
cbbdata::cbd_torvik_game_stats(type = "conf") |> 
  dplyr::filter(conf == "ACC") |> 
  dplyr::group_by(team, year) |> 
  dplyr::summarise(
    away_wins = sum(result == "W" & location == "A"),
    away_loss = sum(result == "L" & location == "A"),
    away_wp = away_wins / (away_wins + away_loss)
  ) |> 
  dplyr::group_by(year) |> 
  dplyr::slice_max(order_by = away_wp, n = 1, with_ties = TRUE) |> 
  dplyr::summarise(
    team = paste(team, collapse = ", "),
    away_wins = max(away_wins),
    away_loss = max(away_loss),
    away_wp = max(away_wp)
  ) |> 
  dplyr::filter(year != 2024) -> acc_road

acc_reg_winners <- tibble::tribble(
  ~year,~winner,~same,
  2008,"North Carolina","Yes",
  2009,"North Carolina","Yes",
  2010,"Duke, Maryland","Yes",
  2011,"North Carolina","Yes",
  2012,"North Carolina","No",
  2013, "Miami FL","Yes",
  2014,"Virginia","Yes",
  2015,"Virginia","Yes",
  2016,"North Carolina","Yes",
  2017,"North Carolina","Yes",
  2018,"Virginia","Yes",
  2019,"Virginia, North Carolina","Yes",
  2020,"Florida State","No",
  2021,"Virginia","Yes",
  2022,"Duke","Yes",
  2023,"Miami FL, Virginia","Yes",
)  |> 
  dplyr::mutate(same = dplyr::if_else(same == "Yes", 
                                      "arrows-left-right", "x"))

# acc reg szn 

acc_reg_szn <- acc_reg_winners |> 
  dplyr::left_join(acc_road, by = "year")

## acc possessions
acc_poss <- readr::read_csv("https://gist.githubusercontent.com/gallochris/38a0da6c27978383e6b9502f14d2b1ca/raw/57ef9b85f48bbc2248bb9d9f7e2b7630f1e32ae6/unc_acc_poss.csv")

GT Table

Code 
title <- cbbplotR::gt_cbb_logo_title(title = 'ACC Regular Season Champions and Best Road Records Since 2008',
                           subtitle = 'The team with the best road record has won at least a share of the regular season title 14 of the last 16 seasons.',
                           type = 'conference',
                           value = 'ACC',
                           logo_height = 65)


acc_reg_szn |> 
  gt::gt() |> 
  gt::fmt_icon(columns = c(same), rows = c(1:4, 6:12, 14:16), 
               stroke_color = "green", fill_color = c( "green")
  ) |> 
  gt::fmt_icon(columns = c(same), rows = c(5, 13), 
               stroke_color = "red", fill_color = c( "red")
  ) |> 
  gtExtras::gt_highlight_rows(rows = c(5, 13), fill = "#fffec8") |> 
  gt::cols_label(
    year = "Season", 
    winner = "Champion(s)", 
    same = "",
    team = "Team(s)",
    away_wins = "W",
    away_loss = "L", 
    away_wp = "Win %"
  ) |> 
  gt::fmt_number(columns = c(away_wp), decimals = 3) |> 
  gt::tab_spanner(columns = c(team, away_wins, away_loss,
                              away_wp), label = "Best Road Record") |> 
  gt::tab_header(title = gt::html(title)) |> 
  gt::tab_source_note(source_note = "Bless your chart | data via cbbdata + cbbplotR | February 22, 2024") |> 
  cbbplotR::gt_theme_athletic() -> acc_gt_tbl
  
gtExtras::gtsave_extra(acc_gt_tbl,
                       filename = "acc_table.png",
                       vheight = 875,
                       vwidth = 850)

acc_gt_tbl
ACC Regular Season Champions and Best Road Records Since 2008
The team with the best road record has won at least a share of the regular season title 14 of the last 16 seasons.
Season Champion(s) Best Road Record
Team(s) W L Win %
2008 North Carolina Arrows Left Right North Carolina 8 0 1.000
2009 North Carolina Arrows Left Right North Carolina 6 2 0.750
2010 Duke, Maryland Arrows Left Right Duke, Florida St., Maryland 5 3 0.625
2011 North Carolina Arrows Left Right North Carolina 6 2 0.750
2012 North Carolina X Duke 8 0 1.000
2013 Miami FL Arrows Left Right Miami FL 7 2 0.778
2014 Virginia Arrows Left Right Pittsburgh, Syracuse, Virginia 7 2 0.778
2015 Virginia Arrows Left Right Virginia 8 1 0.889
2016 North Carolina Arrows Left Right North Carolina 6 3 0.667
2017 North Carolina Arrows Left Right North Carolina, Notre Dame, Virginia 5 4 0.556
2018 Virginia Arrows Left Right Virginia 9 0 1.000
2019 Virginia, North Carolina Arrows Left Right North Carolina 9 0 1.000
2020 Florida State X Virginia 7 3 0.700
2021 Virginia Arrows Left Right Virginia 6 3 0.667
2022 Duke Arrows Left Right Duke 9 1 0.900
2023 Miami FL, Virginia Arrows Left Right Miami FL, Pittsburgh 6 4 0.600
Bless your chart | data via cbbdata + cbbplotR | February 22, 2024

Geom tile possession chart

Code 
cuse_poss <- acc_poss |> 
  dplyr::filter(ID == 5618376) |> 
  dplyr::mutate(scoring = dplyr::if_else(PTS > 0, TRUE, FALSE))

# idea is to use geom_tile to show different empty or scoring possessions 
# tricky part is Poss_Num is sort of double-counting? 
# adjusting the x axis manually with labels to try and clear it up 
# basically think you can divide by 2? 

# find half time 
row_index <- which(cuse_poss$Half_Status == 2)[1]

first_poss_num <- cuse_poss$Poss_Num[row_index] /2

cuse_plot <- cuse_poss |> 
    ggplot2::ggplot(ggplot2::aes(x = Poss_Num/2, y = Poss_Team, fill = scoring)) +
    ggplot2::geom_tile(width = 0.75, height = 0.25) +
    ggplot2::scale_fill_manual(values = c("lightpink", "lightgreen")) +
    ggplot2::scale_x_continuous(breaks = seq(0, 70, 5), limits = c(0, 66)) +
    ggplot2::geom_vline(xintercept = first_poss_num, linetype = "dashed", color = "#333333") +
    ggplot2::theme_classic() +
    ggplot2::theme(legend.position = "none",
                   axis.text.y = cbbplotR::element_cbb_teams(size = 0.9)) +
    ggplot2::labs(x = "Possessions", y = "", 
                  title = "Syracuse 86, North Carolina 79",
                  subtitle ="Green tile represents scoring possession  \nPink tile represents empty possession", caption = "Bless your chart & Fifth Factor | data via bigballR")


ggplot2::ggsave(
  "cuse-plot.png",
  cuse_plot,
  w = 6,
  h = 4,
  dpi = 300,
  type = 'cairo'
)

cuse_plot

Road records since 2012 - UVA

Code 
since_uva <- cbbdata::cbd_torvik_game_stats(type = "conf", team = "North Carolina") |> dplyr::filter(location == "A" & year > 2012) |> 
  dplyr::group_by(opp) |> 
  dplyr::summarise(
    W = sum(result == "W"), 
    L = sum(result == "L"),
    win_pct = W / (W+L)
  ) |> 
  dplyr::arrange(win_pct)

since_uva
# A tibble: 15 × 4
   opp                    W     L win_pct
   <chr>              <int> <int>   <dbl>
 1 Virginia               0     8   0    
 2 Duke                   4     7   0.364
 3 Wake Forest            3     4   0.429
 4 Notre Dame             3     3   0.5  
 5 Virginia Tech          3     3   0.5  
 6 Louisville             4     3   0.571
 7 Miami FL               4     3   0.571
 8 Pittsburgh             4     3   0.571
 9 Syracuse               4     3   0.571
10 Florida St.            5     3   0.625
11 Georgia Tech           5     3   0.625
12 Clemson                6     2   0.75 
13 North Carolina St.     9     3   0.75 
14 Boston College         7     0   1    
15 Maryland               1     0   1