092: ACC Baseball Table

Load data

library(cbbplotR)
acc_base_teams <- readRDS("acc_base_teams.rds")

# find ids
ids <- acc_base_teams$team_id

# function to scrape schedule 
acc_results <- function(id) {
  baseballr::ncaa_schedule_info(team_id = id, year = 2024) |>
    dplyr::filter(home_team_conference == "ACC" & away_team_conference == "ACC") |>
    dplyr::filter(!is.na(home_team_score))
}

# grab acc games
acc_scores <- lapply(ids, acc_results)

acc_results_2024 <- as.data.frame(do.call(rbind, acc_scores)) |>
  dplyr::distinct(contest_id, .keep_all = TRUE ) |> 
  dplyr::mutate(away_team = dplyr::if_else(away_team == "Virginia ", "Virginia", away_team))

# find differential by team 
home_diffs <- acc_results_2024 |>
  dplyr::mutate(h_diff = home_team_score - away_team_score) |>
  dplyr::group_by(home_team) |>
  dplyr::summarize(home_diff = sum(h_diff)) |>
  dplyr::rename(team = home_team)


away_diffs <- acc_results_2024 |>
  dplyr::mutate(a_diff = away_team_score - home_team_score) |>
  dplyr::group_by(away_team) |>
  dplyr::summarize(away_diff = sum(a_diff)) |>
  dplyr::rename(team = away_team)

full_diffs <- merge(home_diffs, away_diffs, by = "team")|>
  dplyr::mutate(full_diff = (home_diff + away_diff))|>
  dplyr::select(team, full_diff, home_diff, away_diff)

# find conference records by team
home_results <- acc_results_2024 |>
  dplyr::mutate(h_result = dplyr::if_else(home_team_score > away_team_score, "W", "L")) |>
  dplyr::group_by(home_team, h_result) |>
  dplyr::count() |>
  tidyr::pivot_wider( 
    names_from = h_result,
    values_from = n)|>
  dplyr::rename(team = home_team, h_w = W, h_l = L)

away_results <- acc_results_2024 |>
  dplyr::mutate(a_result = dplyr::if_else(away_team_score > home_team_score, "W", "L")) |>
  dplyr::group_by(away_team, a_result) |>
  dplyr::count() |>
  tidyr::pivot_wider( 
    names_from = a_result,
    values_from = n) |>
  dplyr::rename(team = away_team, a_w = W, a_l = L) |> 
  dplyr::mutate(across(everything(), ~ replace(.x, is.na(.x), 0))) 

full_recs <- merge(home_results, away_results, by = "team") |>
  dplyr::mutate(W = (h_w + a_w), L = (h_l + a_l))|>
  dplyr::select(team, W, L, h_w, h_l, a_w, a_l)

diffs_recs <- merge(full_diffs, full_recs, by ="team")

RPI Function

# function to get RPI data 
rpi_ranks <- function(url) {
  
  rpi_page <- rvest::read_html(url)
  
  rpi_rk <- rpi_page |> 
    rvest::html_nodes("table") |> 
    rvest::html_table(fill = TRUE)
  
  rpi_table <- as.data.frame(rpi_rk) 
  
  rpi_table <- rpi_table |>
    dplyr::mutate(record = stringr::str_split(Record, "-", simplify = T),
           wins = record[,1],
           losses = record[,2]
    )
}

rpi <- rpi_ranks(url = "https://www.ncaa.com/rankings/baseball/d1/rpi")

# make rpi table
acc_rpi <- rpi |>
  dplyr::filter(Conference == "ACC") |>
  dplyr::select(School, Rank, wins, losses) |>
  dplyr::rename(team = School, rpi = Rank, o_w = wins, o_l = losses)

# atlantic teams 
atl_teams = c("Wake Forest", "Boston College", "Louisville", "NC State", "Notre Dame", "Florida St.", "Clemson")

# make the big table
total_records <- merge(acc_rpi, diffs_recs, by ="team") |> 
  dplyr::mutate(div = dplyr::if_else(team %in% atl_teams, "Atlantic", "Coastal")) |>
  dplyr::select(div, team, rpi, o_w, o_l, W, L, full_diff, h_w, h_l, home_diff, a_w, a_l, away_diff)

Custom GT theme

gt_theme_chris <- function(gt_object, ...) {
  stopifnot("'gt_object' must be a 'gt_tbl', have you accidentally passed raw data?" = "gt_tbl" %in% class(gt_object))
  
  gt_object |> 
    gt::tab_options(
      heading.align = "left",
      table.border.top.style = "none",
      column_labels.border.bottom.style = "none",
      column_labels.border.bottom.color = "black",
      table_body.border.top.style = "none",
      table_body.border.bottom.color = "white",
      heading.border.bottom.style = "none",
      data_row.padding = gt::px(7),
      column_labels.font.size = gt::px(14),
      column_labels.border.bottom.width = gt::px(2),
      ...
    ) |> 
    gt::tab_style(
      style = gt::cell_text(
        color = "#333333",
        font = gt::google_font("Roboto"),
        transform = "uppercase"
      ),
      locations = list(
        gt::cells_column_labels(),
        gt::cells_stubhead()
      )
    ) |> 
    gt::tab_style(
      style = gt::cell_text(
        font = gt::google_font("Roboto"),
        weight = 800
      ),
      locations = gt::cells_title(groups = "title")
    ) |> 
    gt::tab_style(
      style = gt::cell_text(
        font = gt::google_font("Roboto"),
        weight = 400
      ),
      locations = gt::cells_body()
    ) |> 
    gt::tab_style(
      locations = gt::cells_title(groups = "subtitle"),
      style = gt::cell_text(
        size = "small"
      )
    ) |> 
    gt::tab_style(
      locations = gt::cells_title(groups = "title"),
      style = gt::cell_text(
        size = "x-large"
      )
    ) |> 
    gt::tab_style(
      locations = gt::cells_source_notes(),
      style = gt::cell_text(
        size = "x-small"
      )
    ) |> 
    gt::tab_style(
      style = gt::cell_borders(sides = "top", color = 'black', weight = gt::px(1.5), style = 'solid'),
      locations = gt::cells_body(
        rows = gt::everything()
      )
    )  |> 
    gt::tab_style(
      locations = gt::cells_row_groups(groups = gt::everything()),
      style = list(
        gt::cell_text(
          size = "medium",
          weight = "bold"
        ), 
        gt::cell_fill(color = "#FFFEE0")
      )) |> 
    gt::tab_style(
      style = list(
        gt::cell_fill(color = "#FFFEE0"),
        gt::cell_text(color = "#333333", size = gt::px(16), weight = "bold"),
        gt::cell_borders(sides = "right", color = "black", weight = gt::px(2)),
        gt::cell_borders(sides = "bottom", color = "black", weight = gt::px(2))
      ),
      locations = list(
        gt::cells_column_spanners(
          spanners = gt::everything()
        )
      )
    )
} 

GT Table for standings

total_records |> 
  cbbplotR::gt_cbb_teams(team, team) |> 
  dplyr::arrange(-W, -full_diff)|> 
  gt::gt(groupname_col = "div")|>
  gt::fmt_markdown(team) |> 
  gt::cols_label(
    # rename columns
    team = "",
    rpi = "RPI",
    o_w = "W",
    o_l = "L",
    full_diff = "+/-",
    h_w = "W",
    h_l = "L",
    home_diff = "+/-",
    a_w = "W",
    a_l = "L",
    away_diff = "+/-",
  )|>
  gt::tab_spanner(label = "Overall",
                  columns = c(rpi, o_w, o_l))|>
  gt::tab_spanner(label = "ACC Only",
              columns = c(W, L, full_diff))|>
  gt::tab_spanner(label = "Home (ACC)",
              columns = c(h_w, h_l, home_diff))|>
  gt::tab_spanner(label = "Away (ACC)",
              columns = c(a_w, a_l, away_diff))|>
  gt::fmt(
    columns = c(full_diff, home_diff, away_diff),
    fns = function(x) {
      ifelse(x > 0, paste0("+", x), x)
    }
  )|>
  gtExtras::gt_hulk_col_numeric(columns = c(full_diff, away_diff, home_diff)) |> 
  gtExtras::gt_hulk_col_numeric(columns = c(rpi), reverse = TRUE) |> 
  gt::tab_header(title = "ACC Baseball Standings: League Records and Run Differentials by Location", 
                 subtitle = "Shows RPI, overall record, and conference records by location with run differentials through May 18 games.") |>
  gt::tab_source_note(source_note = gt::html("<hr><b>Table by Chris (@dadgumboxscores) + Bless your chart | data via baseballr and ncaa.org</b>")) |>
  gt_theme_chris() |> 
  gt::tab_footnote(footnote = 
                     gt::md(
                       "UNC clinched first place in Coastal Division."
                     ),
                   locations = gt::cells_body(
                     columns = W,
                     rows = 1
                   )) |> 
  gt::cols_align(align = "left", columns = c(team)) |>
  gtExtras::gt_add_divider(columns = c(o_l, full_diff, home_diff, away_diff), 
                           sides = "right", color = "black") -> acc_tbl 

gtExtras::gtsave_extra(acc_tbl,
                       filename = "acc_base_table.png",
                       vheight = 875,
                       vwidth = 875)

acc_tbl

ACC Baseball Standings: League Records and Run Differentials by Location
Shows RPI, overall record, and conference records by location with run differentials through May 18 games.
	Overall			ACC Only			Home (ACC)			Away (ACC)
	RPI	W	L	W	L	+/-	W	L	+/-	W	L	+/-
Coastal
North Carolina	4	41	12	1 22	8	+95	14	1	+74	8	7	+21
Virginia	12	40	14	18	12	+42	9	6	-2	9	6	+44
Duke	20	35	18	16	14	+31	8	7	+8	8	7	+23
Georgia Tech	46	31	21	15	15	+12	9	6	+29	6	9	-17
Virginia Tech	61	32	20	14	16	-9	7	8	+8	7	8	-17
Miami FL	90	25	29	11	19	-38	7	8	+1	4	11	-39
Pittsburgh	83	26	27	10	20	-70	5	10	-54	5	10	-16
Atlantic
Clemson	7	40	13	20	10	+54	11	4	+46	9	6	+8
North Carolina St.	14	32	19	18	11	-6	13	2	+20	5	9	-26
Florida St.	8	39	14	17	12	+40	11	3	+39	6	9	+1
Louisville	52	32	22	16	14	-51	9	6	-17	7	8	-34
Wake Forest	11	36	19	15	15	+17	8	7	+12	7	8	+5
Notre Dame	77	27	25	9	21	-31	9	6	+43	0	15	-74
Boston College	81	22	31	8	22	-86	3	12	-50	5	10	-36
Table by Chris (@dadgumboxscores) + Bless your chart | data via baseballr and ncaa.org
1 UNC clinched first place in Coastal Division.