Defensive efficiency plot

def_ppp_plot.R

# Attempt to recreate or riff on the plot from the Tufte book
# https://opticalacumen.blogspot.com/2014/07/data-visualization-in-1919.html

# Fetch the player minutes data by game over past three seasons
hubert <-
  cbbdata::cbd_torvik_game_factors(team = "North Carolina") |>
  dplyr::filter(year > 2021) |>
  dplyr::arrange(date) |>
  dplyr::mutate(off_ppp = format(round(off_ppp / 100, 2), nsmall = 2),
                def_ppp = format(round(def_ppp / 100, 2), nsmall = 2)) |>
  dplyr::select(game_id, date, off_ppp, opp, result, def_ppp)

# Add game numbers by month now
def_by_month <-
  hubert |>
  dplyr::mutate(month_year = format(as.Date(date), "%B %Y")) |>
  dplyr::arrange(date) |>
  dplyr::group_by(month_year) |>
  dplyr::mutate(game_num = dplyr::row_number()) |>
  dplyr::select(game_id,
                date,
                month_year,
                game_num,
                off_ppp,
                opp,
                result,
                def_ppp)


# Set custom order
dates_order <- c(
  "November 2021",
  "December 2021",
  "January 2022",
  "February 2022",
  "March 2022",
  "April 2022",
  "November 2022",
  "December 2022",
  "January 2023",
  "February 2023",
  "March 2023",
  "November 2023",
  "December 2023",
  "January 2024"
)

# Set your "month_year" labels as factors with the desired order
def_by_month$date_label <-
  factor(def_by_month$month_year, levels = dates_order)


# Make the plot using hrbrthemes
def_plot <- def_by_month |>
  dplyr::arrange(date) |>
  ggplot2::ggplot(ggplot2::aes(x = date_label, y = game_num)) +
  ggplot2::geom_text(
    data = def_by_month |> dplyr::filter(def_ppp >= 1),
    ggplot2::aes(label = def_ppp),
    size = 3.5,
    color = "#3182bd",
    family = "Roboto Condensed",
  ) +
  ggplot2::geom_text(
    data = def_by_month |> dplyr::filter(def_ppp < 1),
    ggplot2::aes(label = def_ppp),
    size = 3.5,
    color = "#deebf7",
    family = "Roboto Condensed",
    fontface = "bold"
  ) +
  ggplot2::scale_y_continuous(limits = c(1, 9), breaks = seq(1, 9, 1)) +
  ggplot2::scale_x_discrete(labels = scales::label_wrap(5)) +
  hrbrthemes::theme_modern_rc() +
  ggplot2::geom_vline(
    xintercept = c(6.5, 11.5),
    linetype = "dashed",
    color = "#acacac"
  ) +
  ggplot2::annotate(
    "label",
    x = 1.5,
    y = 8.5,
    label = "2021-22 Season \n20 of 39 opponents  \nunder 1 PPP",
    family = "Roboto Condensed",
    size = 3.5,
    color = "#333333",
    fontface = "bold",
    fill = "#deebf7"
  ) +
  ggplot2::annotate(
    "label",
    x = 8.5,
    y = 8.5,
    label = "2022-23 Season \n16 of 33 opponents  \nunder 1 PPP",
    family = "Roboto Condensed",
    size = 3.5,
    color = "#333333",
    fontface = "bold",
    fill = "#deebf7"
  ) +
  ggplot2::annotate(
    "label",
    x = 13,
    y = 8.5,
    label = "2023-24 Season \n12 of 18 opponents  \nunder 1 PPP",
    family = "Roboto Condensed",
    size = 3.5,
    color = "#333333",
    fontface = "bold",
    fill = "#deebf7"
  ) +
  hrbrthemes::theme_modern_rc() +
  ggplot2::theme(
    axis.text.y = ggplot2::element_blank(),
    panel.grid = ggplot2::element_blank(),
  ) +
  ggplot2::labs(
    x = "",
    y = "",
    title = "North Carolina Basketball: Defensive efficiency by month over past three seasons",
    subtitle = "Shows opponent points per possession by game for each month over past three seasons",
    caption = "Bless your chart | January 21, 2024 | data via cbbdata"
  )

# Save the plot
ggplot2::ggsave(
  "def_plot.png",
  def_plot,
  w = 12,
  h = 8,
  dpi = 600,
  type = 'cairo'
)