Skip to contents

NHL Shot Map

Saiem Gilani
@saiemgilani @saiemgilani

Source: vignettes/recipe-nhl-shot-map.Rmd
recipe-nhl-shot-map.Rmd
On this page

The brief: the morning after the Stanley Cup clincher, post one image of the game: every unblocked shot on a rink, each team in its colors, the goals called out, and the score. It goes out at 1200 x 675 for X and Bluesky and 1600 x 900 for the blog recap. The play-by-play is the SportsDataverse NHL release through fastRhockey, and sdv_surface() draws the rink.

library(sdvplotR)
library(ggplot2)
library(dplyr, warn.conflicts = FALSE)

season <- 2026 # the 2025-26 season: NHL seasons are named by the year they end
cup_final <- 2025030416 # Stanley Cup Final, Game 6
out_dir <- tempfile("sdvplotR-recipe-") # where the exports go; use your own folder
dir.create(out_dir)

1. Get the data

load_nhl_pbp() reads the whole season’s play-by-play, one row per event, and load_nhl_schedule() the games with their final scores. The game’s own row says who played where; the series rows confirm this game ended the Final.

pbp <- fastRhockey::load_nhl_pbp(season)
schedule <- fastRhockey::load_nhl_schedule(season)

game <- filter(schedule, game_id == cup_final)
final <- schedule |>
  filter(playoff_round == 4, game_id <= cup_final) |>
  mutate(winner = if_else(home_score > away_score, home_team_abbr, away_team_abbr))
count(final, winner)
#> # A tibble: 2 × 2
#>   winner     n
#>   <chr>  <int>
#> 1 CAR        4
#> 2 VGK        2
stopifnot(max(count(final, winner)$n) == 4) # the winner's fourth win: the clincher

shots <- pbp |>
  filter(game_id == cup_final, event_type %in% c("SHOT", "MISSED_SHOT", "GOAL")) |>
  transmute(
    event_type, period, period_time,
    team = event_team_abbr, shooter = event_player_1_name,
    x, y, x_fixed, y_fixed, xg
  )
sprintf("%s %d at %s %d", game$away_team_abbr, game$away_score, game$home_team_abbr, game$home_score)
#> [1] "CAR 3 at VGK 0"
count(shots, team, event_type)
#> # A tibble: 5 × 3
#>   team  event_type      n
#>   <chr> <chr>       <int>
#> 1 CAR   GOAL            3
#> 2 CAR   MISSED_SHOT    13
#> 3 CAR   SHOT           20
#> 4 VGK   MISSED_SHOT    15
#> 5 VGK   SHOT           22

Blocked shots are left out: the feed records where a blocked shot was blocked, not where it was taken.

2. The first draft

Scatter the coordinates as they come, one color per team.

ggplot(shots, aes(x, y, color = team)) +
  geom_point()

First draft: the game's unblocked shots as colored dots on a blank grid, with both teams' shots at both ends.

Both teams have shots at both ends: teams switch ends every period, and the feed records where on the ice each shot happened. With no rink the dots float in space, and the goals look like every other shot.

3. One end per team, on a rink

The release already carries x_fixed and y_fixed: every event rotated so the home team shoots right and the visitors left for the whole game. sdv_surface("nhl") draws a regulation rink in the same coordinates (feet, center ice at 0, 0) and returns a ggplot, so the shots are just more layers.

sdv_surface("nhl") +
  geom_point(aes(x_fixed, y_fixed, color = team), data = shots)

The shots on a regulation NHL rink, the visitors' shots all at the left end and the home team's at the right.

4. Team colors, shot types and the goals

Each team’s shots take its primary color through scale_color_sdv(). A shot on goal is a filled dot, a miss an open ring, and each dot is sized by its expected goals (xg, the release’s chance that a shot like it scores), so the dangerous chances stand out from the long-range ones. A goal is a large ringed dot with the scorer’s name and the time in a white box, so it reads over the shots and the rink lines; the labels alternate above and below, so two goals close together keep their labels apart. The drawing goes in a function so the final layout can reuse it.

goals <- shots |>
  filter(event_type == "GOAL") |>
  mutate(
    label = paste0(sub(".* ", "", shooter), " (P", period, " ", period_time, ")"),
    vjust = if_else(row_number() %% 2 == 1, -0.8, 1.8) # above, below, above: close goals stay apart
  )

draw_shots <- function(...) {
  sdv_surface("nhl", ...) +
    geom_point(
      aes(x_fixed, y_fixed, color = team, size = xg),
      data = filter(shots, event_type == "SHOT"), alpha = 0.8
    ) +
    geom_point(
      aes(x_fixed, y_fixed, color = team, size = xg),
      data = filter(shots, event_type == "MISSED_SHOT"), shape = 1, stroke = 0.9
    ) +
    geom_point(
      aes(x_fixed, y_fixed, fill = team),
      data = goals, shape = 21, size = 6, color = "white", stroke = 1.5
    ) +
    geom_label(
      aes(x_fixed, y_fixed, label = label, color = team, vjust = vjust),
      data = goals, size = 2.6, fontface = "bold",
      fill = "white", linewidth = 0, label.padding = unit(0.12, "lines")
    ) +
    scale_color_sdv(league = "nhl", guide = "none") +
    scale_fill_sdv(league = "nhl", guide = "none") +
    scale_size_area(max_size = 4.5, guide = "none")
}

draw_shots()
#> Warning: Removed 1 row containing missing values or values outside the scale range
#> (`geom_point()`).

The shot map in team colors: filled dots for shots on goal, open rings for misses, sized by expected goals, and the three goals as large ringed dots labelled with the scorer and time.

5. The frame: score, teams and story

A shot map needs a scoreboard. Each team gets a block on the side of the rink it attacks: its logo, its score and its shots on goal (shots plus goals), placed past the boards. xlims widens the rink’s canvas to make room, and goes to ‘sportyR’ through sdv_surface()’s .... The title says what happened, the subtitle how to read the rink, and the caption carries the source.

club_names <- sdv_teams("nhl") |> select(team_abbr = abbr, team_name = name)
scoreboard <- tibble(
  team = c(game$away_team_abbr, game$home_team_abbr),
  x = c(-122, 122), # outside the boards, on the side each team attacks
  score = c(game$away_score, game$home_score)
) |>
  left_join(
    count(filter(shots, event_type %in% c("SHOT", "GOAL")), team, name = "sog"),
    by = "team"
  ) |>
  left_join(club_names, by = c("team" = "team_abbr"))
winner <- slice_max(scoreboard, score, n = 1)
loser <- slice_min(scoreboard, score, n = 1)

shot_map <- function() {
  draw_shots(xlims = c(-142, 142)) +
    geom_sdv_logos(aes(x = x, y = 14, team = team), data = scoreboard, league = "nhl", height = 0.2) +
    geom_text(aes(x = x, y = -8, label = score), data = scoreboard, size = 11, fontface = "bold") +
    geom_text(
      aes(x = x, y = -21, label = paste(sog, "shots\non goal")),
      data = scoreboard, size = 2.7, color = "grey40", lineheight = 0.9
    ) +
    labs(
      title = sprintf(
        "The %s won the Stanley Cup, %d-%d in Game 6",
        winner$team_name, winner$score, loser$score
      ),
      subtitle = sprintf(
        "Every unblocked shot, %s. %s shoots left, %s right. Dots: shots on goal; rings: misses; size: expected goals.",
        game$venue, game$away_team_abbr, game$home_team_abbr
      ),
      caption = "Data: NHL via fastRhockey  |  Chart: sdvplotR"
    ) +
    theme(
      plot.title = element_text(face = "bold", size = 14, margin = margin(b = 3)),
      plot.subtitle = ggtext::element_textbox_simple(color = "grey40", size = 8.5, margin = margin(b = 4)),
      plot.caption = element_text(color = "grey45", size = 7),
      plot.margin = margin(10, 12, 6, 12),
      plot.background = element_rect(fill = "white", color = NA)
    )
}

shot_map()
#> Warning: Removed 1 row containing missing values or values outside the scale range
#> (`geom_point()`).

The finished shot map: the Stanley Cup clincher's unblocked shots on a rink in team colors, with each team's logo, score and shots on goal beside its end, a headline and a source line.

6. Export for X and the blog

1200 x 675 for X and Bluesky and 1600 x 900 for the blog are the same 16:9 shape, so one 8 x 4.5 in plot serves both; only the dpi changes (150 and 200). Inches times dpi gives the exact pixels, and the logos, sized as a share of the panel, scale with everything else.

chart <- shot_map()
files <- c(
  file.path(out_dir, "nhl_shot_map_1200x675.png"),
  file.path(out_dir, "nhl_shot_map_1600x900.png")
)
ggsave(files[1], chart, width = 8, height = 4.5, dpi = 150)
#> Warning: Removed 1 row containing missing values or values outside the scale range
#> (`geom_point()`).
ggsave(files[2], chart, width = 8, height = 4.5, dpi = 200)
#> Warning: Removed 1 row containing missing values or values outside the scale range
#> (`geom_point()`).
magick::image_info(magick::image_read(files))[c("width", "height")]
#> # A tibble: 2 × 2
#>   width height
#>   <int>  <int>
#> 1  1200    675
#> 2  1600    900
magick::image_read(files[1])

The finished shot map exported at 1200 by 675 pixels for X and Bluesky.

Our Authors

Citation

To cite the sdvplotR R package in publications, use:

BibTeX Citation

@misc{gilani_carl_lee_kay_weatherman_sdvplotR,
  author = {Gilani, Saiem and Carl, Sebastian and Lee, Jared and Kay, Camden and Weatherman, Andrew},
  title = {sdvplotR: The SportsDataverse's R Package for Multi-League Sports Plotting.},
  url = {https://sdvplotR.sportsdataverse.org},
  year = {2026}
}