NHL Visualizations with fastRhockey and sdvplotR

On this page

Ten charts and tables from the 2025-26 NHL season: goal differential in team colors, the final standings, five-on-five expected goals, the champion’s shot map on a rink, finishing and goaltending against expected goals, the scoring leaders with their headshots, one franchise’s marks from Winnipeg to Utah, the points race and the playoffs as a tier list, then the expected-goals chart as an interactive one. The data comes through fastRhockey: its release files (play-by-play with an expected-goals value on every unblocked shot, and box scores), and the NHL’s own APIs for the official standings and the franchise records.

library(sdvplotR)
library(ggplot2)
library(dplyr)
library(gt)

Load the season

The release files are read when this page is built. Each load_nhl_*() call downloads one season from the SportsDataverse data releases; the columns kept here are the ones the examples use.

season <- 2026 # the 2025-26 season, named for the year it ends

pbp <- fastRhockey::load_nhl_pbp_lite(seasons = season) |>
  select(
    game_id, season_type, period, event_type, event_team_abbr, event_team_type,
    home_abbr, away_abbr, strength_state, x_fixed, y_fixed, xg,
    event_goalie_id, event_goalie_name
  )

# game_id 2025020001 is a regular-season game (02); 2025030416 a playoff game
# (03): round 4, series 1, game 6
team_box <- fastRhockey::load_nhl_team_box(seasons = season) |>
  select(game_id, game_date, team_abbrev, goals, goals_against) |>
  mutate(game_type = game_id %/% 10000 %% 100)

skater_box <- fastRhockey::load_nhl_skater_box(seasons = season) |>
  filter(game_id %/% 10000 %% 100 == 2) |>
  select(game_id, team_abbrev, player_id, player_name, goals, assists, points)

c(plays = nrow(pbp), team_games = nrow(team_box), skater_games = nrow(skater_box))
#>        plays   team_games skater_games 
#>       443569         2788        47231

The official standings come from api-web.nhle.com and the franchise records from records.nhl.com:

standings <- fastRhockey::nhl_standings(date = "2026-04-16") |> # the last day of the regular season
  select(
    team_abbr, team_name, conference_name, division_name, division_sequence, games_played,
    wins, losses, ot_losses, points, point_pctg, regulation_wins, goals_for, goals_against,
    goal_differential, streak_code, streak_count
  )

# one franchise's every season: the Winnipeg Jets (1979), the Coyotes, Utah
lineage <- fastRhockey::nhl_records_franchise_season_results() |>
  filter(game_type_id == 2, team_id %in% c(33, 27, 53, 59, 68), season_id <= 20252026) |>
  select(season_id, team_id, tri_code, team_name, games_played, points)

These two calls are read from a snapshot taken on October 05, 2026 (fastRhockey 1.0.0), because the NHL’s APIs are not called when this site is built; data-raw/article_fixtures.R takes it with the same code.

A few of the NHL’s three-letter codes differ from the abbreviations in sdvplotR’s team index. sdv_resolve() converts them once, to the index’s team ids, so every join and color scale below agrees:

standings <- standings |>
  mutate(team = sdv_resolve(team_abbr, league = "nhl"))

standings |>
  mutate(sdv_abbr = sdv_teams("nhl")$abbr[match(team, sdv_teams("nhl")$team_id)]) |>
  filter(team_abbr != sdv_abbr) |>
  select(team_name, team_abbr, sdv_abbr)
#>             team_name team_abbr sdv_abbr
#> 1 Tampa Bay Lightning       TBL       TB
#> 2        Utah Mammoth       UTA     UTAH
#> 3   Los Angeles Kings       LAK       LA
#> 4   New Jersey Devils       NJD       NJ
#> 5     San Jose Sharks       SJS       SJ

api <- "Data: NHL API via fastRhockey | Viz: sdvplotR"
release <- "Data: fastRhockey play-by-play and box score release | Viz: sdvplotR"

1. Goal differential in team colors

Bars in each team’s color from scale_fill_sdv(), with the logo at the end of each bar from geom_sdv_logos().

goal_diff <- standings |>
  arrange(desc(goal_differential), team) |>
  mutate(
    team = factor(team, levels = team),
    logo_y = goal_differential + if_else(goal_differential >= 0, 9, -9)
  )

ggplot(goal_diff, aes(team, goal_differential)) +
  geom_col(aes(fill = team), width = 0.75) +
  geom_hline(yintercept = 0, colour = "grey20", linewidth = 0.4) +
  geom_sdv_logos(aes(y = logo_y, team = team), league = "nhl", width = 0.027) +
  scale_fill_sdv(league = "nhl") +
  labs(
    title = "NHL goal differential, 2025-26",
    subtitle = "Regular season, goals for minus goals against",
    x = NULL,
    y = "Goal differential",
    caption = api
  ) +
  theme_minimal(base_size = 12) +
  theme(
    legend.position = "none",
    axis.text.x = element_blank(),
    panel.grid.major.x = element_blank(),
    plot.title = element_text(face = "bold")
  )

Bar chart of the 32 NHL teams' 2025-26 regular-season goal differential, sorted from best to worst, each bar in its team's color with the team logo at the end of the bar.

2. Final standings by division

A gt table grouped by division, with gt_sdv_logos() for the logos and gt_theme_athletic() for the look. The goal differential fill goes on after the theme so the theme cannot replace it.

nhl_divisions <- c("Atlantic", "Metropolitan", "Central", "Pacific")
lim <- max(abs(standings$goal_differential))

standings |>
  mutate(
    division = factor(division_name, levels = nhl_divisions),
    logo = team,
    streak = paste0(streak_code, streak_count)
  ) |>
  arrange(division, division_sequence) |>
  select(
    division, logo, team_name, games_played, wins, losses, ot_losses, points,
    point_pctg, regulation_wins, goals_for, goals_against, goal_differential, streak
  ) |>
  gt(groupname_col = "division") |>
  gt_sdv_logos(columns = "logo", league = "nhl", height = 22) |>
  fmt_number(columns = point_pctg, decimals = 3) |>
  cols_align(align = "left", columns = team_name) |>
  cols_label(
    logo = "", team_name = "Team", games_played = "GP", wins = "W", losses = "L",
    ot_losses = "OTL", points = "PTS", point_pctg = "PTS%", regulation_wins = "RW",
    goals_for = "GF", goals_against = "GA", goal_differential = "DIFF", streak = "STRK"
  ) |>
  tab_header(
    title = "NHL standings, 2025-26",
    subtitle = "Final regular season, grouped by division"
  ) |>
  tab_source_note(api) |>
  gt_theme_athletic() |>
  data_color(
    columns = goal_differential,
    palette = c("#c84630", "white", "#2a7ab9"),
    domain = c(-lim, lim)
  )

	Team	GP	W	L	OTL	PTS	PTS%	RW	GF	GA	DIFF	STRK
NHL standings, 2025-26
Final regular season, grouped by division
Atlantic
	Buffalo Sabres	82	50	23	9	109	0.665	42	288	241	47	OT1
	Tampa Bay Lightning	82	50	26	6	106	0.646	40	290	231	59	L1
	Montréal Canadiens	82	48	24	10	106	0.646	34	283	256	27	L1
	Boston Bruins	82	45	27	10	100	0.610	33	272	250	22	W2
	Ottawa Senators	82	44	27	11	99	0.604	38	278	246	32	W1
	Detroit Red Wings	82	41	31	10	92	0.561	30	241	258	-17	L1
	Florida Panthers	82	40	38	4	84	0.512	32	251	276	-25	W3
	Toronto Maple Leafs	82	32	36	14	78	0.476	23	253	299	-46	L5
Metropolitan
	Carolina Hurricanes	82	53	22	7	113	0.689	39	296	240	56	W1
	Pittsburgh Penguins	82	41	25	16	98	0.598	34	293	268	25	L3
	Philadelphia Flyers	82	43	27	12	98	0.598	27	250	243	7	W3
	Washington Capitals	82	43	30	9	95	0.579	37	263	244	19	W4
	Columbus Blue Jackets	82	40	30	12	92	0.561	28	253	253	0	L2
	New York Islanders	82	43	34	5	91	0.555	29	233	241	-8	L3
	New Jersey Devils	82	42	37	3	87	0.530	29	230	254	-24	L1
	New York Rangers	82	34	39	9	77	0.470	25	238	250	-12	W1
Central
	Colorado Avalanche	82	55	16	11	121	0.738	48	302	203	99	W3
	Dallas Stars	82	50	20	12	112	0.683	38	279	226	53	W5
	Minnesota Wild	82	46	24	12	104	0.634	31	272	240	32	W1
	Utah Mammoth	82	43	33	6	92	0.561	33	268	240	28	L1
	St. Louis Blues	82	37	33	12	86	0.524	33	231	258	-27	W4
	Nashville Predators	82	38	34	10	86	0.524	28	247	269	-22	L2
	Winnipeg Jets	82	35	35	12	82	0.500	28	231	260	-29	L4
	Chicago Blackhawks	82	29	39	14	72	0.439	22	213	275	-62	W1
Pacific
	Vegas Golden Knights	82	39	26	17	95	0.579	30	265	250	15	W3
	Edmonton Oilers	82	41	30	11	93	0.567	32	282	269	13	W1
	Anaheim Ducks	82	43	33	6	92	0.561	26	273	288	-15	W1
	Los Angeles Kings	82	35	27	20	90	0.549	22	225	247	-22	L1
	San Jose Sharks	82	39	35	8	86	0.524	27	251	292	-41	W1
	Seattle Kraken	82	34	37	11	79	0.482	26	226	263	-37	L3
	Calgary Flames	82	34	39	9	77	0.470	27	212	259	-47	W1
	Vancouver Canucks	82	25	49	8	58	0.354	15	216	316	-100	L1
Data: NHL API via fastRhockey \| Viz: sdvplotR

3. Five-on-five expected goals

Expected goals (xG) weigh every unblocked shot by its chance of scoring, so summed for and against each team at five-on-five they show who drives play, with less luck in them than goals. The y axis is reversed, so the top right is where you want to be; geom_mean_lines() marks the league averages.

five <- pbp |>
  filter(season_type == "R", strength_state == "5v5", !is.na(xg)) |>
  mutate(defending = if_else(event_team_abbr == home_abbr, away_abbr, home_abbr))
xg_for <- five |>
  summarise(xgf = sum(xg), .by = event_team_abbr) |>
  rename(team_abbrev = event_team_abbr)
xg_against <- five |>
  summarise(xga = sum(xg), .by = defending) |>
  rename(team_abbrev = defending)
games_played <- team_box |>
  filter(game_type == 2) |>
  count(team_abbrev, name = "gp")

xg <- xg_for |>
  inner_join(xg_against, by = "team_abbrev") |>
  inner_join(games_played, by = "team_abbrev") |>
  mutate(
    xgf_pg = xgf / gp,
    xga_pg = xga / gp,
    team = sdv_resolve(team_abbrev, league = "nhl")
  )
stopifnot(nrow(xg) == 32)

ggplot(xg, aes(xgf_pg, xga_pg)) +
  geom_mean_lines(aes(x0 = xgf_pg, y0 = xga_pg), colour = "grey60") +
  geom_sdv_logos(aes(team = team), league = "nhl", width = 0.06) +
  scale_y_reverse() +
  labs(
    title = "Who drove play at five-on-five, 2025-26",
    subtitle = "Expected goals per game, regular season",
    x = "5v5 xG for per game",
    y = "5v5 xG against per game (reversed)",
    caption = release
  ) +
  theme_minimal(base_size = 12) +
  theme(plot.title = element_text(face = "bold"))

Scatter plot of the 32 NHL teams in 2025-26, each drawn as its logo, placed by five-on-five expected goals for per game (horizontal) and against per game (vertical, reversed), with dashed lines at the league averages.

4. The champion’s shot map on the rink

sdv_surface("nhl", team) draws a regulation rink with ‘sportyR’, its center line, center faceoff circle and boards in the team’s colors, and the shots go on top as ordinary layers. The champion comes from the data: the team with four wins in round 4. In the release, x_fixed has the home team shooting right and the away team left, so flipping the away shots (both x and y) puts every shot in one attacking end. ggtitle_image() puts the logo in the title.

champion <- team_box |>
  filter(game_type == 3, game_id %/% 100 %% 10 == 4) |>
  summarise(wins = sum(goals > goals_against), .by = team_abbrev) |>
  filter(wins == 4) |>
  pull(team_abbrev)
champ <- sdv_resolve(champion, league = "nhl")
champ_name <- standings$team_name[standings$team == champ]

shots <- pbp |>
  filter(
    event_team_abbr == champion,
    event_type %in% c("SHOT", "MISSED_SHOT", "GOAL"),
    !is.na(x_fixed)
  ) |>
  mutate(
    flip = if_else(event_team_type == "away", -1, 1),
    x = x_fixed * flip,
    y = y_fixed * flip,
    goal = event_type == "GOAL"
  )

sdv_surface("nhl", champ, display_range = "offense") +
  geom_point(aes(x, y), data = filter(shots, !goal), colour = "grey35", alpha = 0.2, size = 0.9) +
  geom_point(
    aes(x, y),
    data = filter(shots, goal), shape = 21, size = 2.2, stroke = 0.3,
    colour = "white", fill = sdv_team_colors("nhl", champ)
  ) +
  ggtitle_image(
    title_image = champ,
    title = paste(champ_name, "shots, 2025-26"),
    subtitle = paste0(
      format(sum(!shots$goal), big.mark = ","), " shots and misses (grey), ",
      sum(shots$goal), " goals (color); regular season and playoffs"
    ),
    image_height = 20,
    league = "nhl"
  ) +
  labs(caption = release) +
  theme_title_image(size = 14, face = "bold") +
  # the rink layers leave the background transparent
  theme(plot.background = element_rect(fill = "white", colour = NA))

Offensive half of a hockey rink in the Stanley Cup champion's colors, with every unblocked shot the champion took in 2025-26 as a faint grey dot and every goal as a dot in the team's color, clustered in front of the net.

5. Finishing: goals above expected

Goals minus expected goals in all situations, shootouts left out. The release’s xG model was fit on earlier seasons and expects more goals than 2025-26 produced, so every team’s xG is first scaled by the league’s goals-to-xG ratio; what is left is finishing relative to the league. scale_x_sdv() and theme_x_sdv() put the logos on the axis.

attempts <- pbp |>
  filter(season_type == "R", !is.na(xg), period <= 4) # period 5 is the shootout
ratio <- sum(attempts$event_type == "GOAL") / sum(attempts$xg)

finishing <- attempts |>
  summarise(goals = sum(event_type == "GOAL"), xg = sum(xg) * ratio, .by = event_team_abbr) |>
  mutate(gax = goals - xg, team = sdv_resolve(event_team_abbr, league = "nhl")) |>
  arrange(desc(gax), team) |>
  mutate(team = factor(team, levels = team))

ggplot(finishing, aes(team, gax, fill = team)) +
  geom_col(width = 0.75) +
  geom_hline(yintercept = 0, colour = "grey20", linewidth = 0.4) +
  scale_fill_sdv(league = "nhl") +
  scale_x_sdv(league = "nhl", size = 20) +
  labs(
    title = "Who finished their chances, 2025-26",
    subtitle = "Goals minus expected goals (scaled to the league), regular season",
    x = NULL,
    y = "Goals above expected",
    caption = release
  ) +
  theme_minimal(base_size = 12) +
  theme_x_sdv() +
  theme(
    legend.position = "none",
    panel.grid.major.x = element_blank(),
    plot.title = element_text(face = "bold")
  )

Bar chart of the 32 NHL teams' 2025-26 goals above expected, sorted from best to worst finishers, each bar in its team's color with the team logo under it on the horizontal axis.

The model expects 12% more goals than the league scored. After scaling, the Boston Bruins finished best (+27.5 goals) and the New Jersey Devils worst (-27.2).

6. Scoring leaders with headshots

The release’s box scores identify players by their NHL API ID (McDavid is 8478402), not their ESPN athlete ID, which is why the headshots pass id_system = "league": it draws NHL API IDs from the NHL’s image CDN. Leave id_system unset for ESPN athlete IDs, such as those in fastRhockey’s espn_nhl_*() data. Both are plain numbers, so the wrong setting draws someone else or nothing. Goals and assists stack into points, and each player’s team is the one he played his last game for.

leaders <- skater_box |>
  arrange(game_id) |>
  summarise(
    name = last(player_name), team_abbrev = last(team_abbrev),
    goals = sum(goals), assists = sum(assists), points = sum(points),
    .by = player_id
  ) |>
  filter(min_rank(desc(points)) <= 10) |>
  mutate(team = sdv_resolve(team_abbrev, league = "nhl")) |>
  arrange(points, desc(name)) |>
  mutate(name = factor(name, levels = name))
bars <- bind_rows(
  transmute(leaders, name, stat = "Goals", n = goals),
  transmute(leaders, name, stat = "Assists", n = assists)
) |>
  mutate(stat = factor(stat, levels = c("Assists", "Goals"))) # goals stack from zero

ggplot(leaders, aes(y = name)) +
  geom_col(aes(x = n, fill = stat), data = bars, width = 0.7) +
  geom_sdv_headshots(
    aes(x = -12, player_id = player_id),
    league = "nhl", id_system = "league", height = 0.085
  ) +
  geom_text(aes(x = points + 3, label = points), hjust = 0, fontface = "bold", size = 4) +
  geom_sdv_logos(aes(x = points + 21, team = team), league = "nhl", height = 0.065) +
  scale_fill_manual(values = c(Goals = "#1f3b73", Assists = "#9fb4d8"), breaks = c("Goals", "Assists")) +
  scale_x_continuous(
    breaks = seq(0, 150, 25),
    limits = c(-20, max(leaders$points) + 28),
    expand = expansion(0)
  ) +
  labs(
    title = "NHL points leaders, 2025-26",
    subtitle = "Regular season",
    x = "Points",
    y = NULL,
    fill = NULL,
    caption = release
  ) +
  theme_minimal(base_size = 12) +
  theme(
    legend.position = "top",
    legend.justification = "left",
    panel.grid.major.y = element_blank(),
    panel.grid.minor = element_blank(),
    plot.title = element_text(face = "bold")
  )

Horizontal stacked bar chart of the ten NHL points leaders of the 2025-26 regular season, goals in dark blue and assists in light blue, with each player's headshot at the left, his point total and his team logo at the end of the bar.

7. Goaltending: goals saved above expected

The play-by-play names the goalie in net for every unblocked shot, so the same expected goals grade the goalies: goals saved above expected (GSAx) is the scaled xG a goalie faced minus the goals he allowed. Among the goalies who faced 1,000 unblocked shots, the five best and five worst, in bars of their team’s color (the team he faced his last shot for) with the headshot at the end.

faced <- pbp |>
  filter(
    season_type == "R", event_type %in% c("SHOT", "MISSED_SHOT", "GOAL"),
    !is.na(xg), !is.na(event_goalie_id), period <= 4
  ) |>
  mutate(goalie_team = if_else(event_team_abbr == home_abbr, away_abbr, home_abbr))
ratio_faced <- sum(faced$event_type == "GOAL") / sum(faced$xg)

goalies <- faced |>
  arrange(game_id) |>
  summarise(
    shots = n(), goals = sum(event_type == "GOAL"), xga = sum(xg) * ratio_faced,
    team_abbrev = last(goalie_team),
    .by = c(event_goalie_id, event_goalie_name)
  ) |>
  mutate(gsax = xga - goals) |>
  filter(shots >= 1000)
ends <- goalies |>
  filter(min_rank(desc(gsax)) <= 5 | min_rank(gsax) <= 5) |>
  arrange(gsax) |>
  mutate(
    name = factor(event_goalie_name, levels = event_goalie_name),
    team = sdv_resolve(team_abbrev, league = "nhl"),
    side = sign(gsax)
  )

ggplot(ends, aes(gsax, name)) +
  geom_col(aes(fill = team), width = 0.7) +
  geom_vline(xintercept = 0, colour = "grey20") +
  geom_sdv_headshots(
    aes(x = gsax + 3.5 * side, player_id = event_goalie_id),
    league = "nhl", id_system = "league", height = 0.085
  ) +
  geom_text(
    aes(x = gsax + 7.5 * side, label = sprintf("%+.1f", gsax), hjust = if_else(side > 0, 0, 1)),
    fontface = "bold", size = 3.8
  ) +
  scale_fill_sdv(league = "nhl") +
  scale_x_continuous(expand = expansion(add = 13)) +
  labs(
    title = "Goals saved above expected, 2025-26",
    subtitle = paste(nrow(goalies), "goalies faced 1,000+ unblocked shots; the five best and five worst, regular season"),
    x = "Goals saved above expected",
    y = NULL,
    caption = release
  ) +
  theme_minimal(base_size = 12) +
  theme(
    legend.position = "none",
    panel.grid.major.y = element_blank(),
    panel.grid.minor = element_blank(),
    plot.title = element_text(face = "bold")
  )

Diverging horizontal bar chart of goals saved above expected in the 2025-26 NHL regular season for the five best and five worst goalies who faced at least 1,000 unblocked shots, bars in team colors, with each goalie's headshot and value at the end of his bar.

8. One franchise, many marks: Winnipeg, Phoenix, Arizona, Utah

The original Winnipeg Jets moved to Phoenix in 1996, became the Arizona Coyotes in 2014, and their hockey operations moved to Salt Lake City in 2024 as the Utah Hockey Club, the Utah Mammoth since 2025. sdvplotR maps every one of the franchise’s codes, the original Jets’ WIN included, to today’s club, so a dataset from any era gets a logo and a color and groups as one franchise. Today’s Winnipeg Jets, the former Atlanta Thrashers, are WPG:

sdv_resolve(c("WIN", "PHX", "ARI", "UTA", "WPG"), league = "nhl")
#> [1] "28"     "129764" "129764" "129764" "28"

That key draws today’s mark. Given a season, geom_sdv_logos() draws the mark a club wore that season instead. Historical NHL marks are keyed by the NHL’s code for that identity (WIN, PHX, ARI), and seasons are named for the year they end. One mark per logo era, with the club’s name that season from the NHL’s records:

marks <- tibble(
  key = c("WIN", "WIN", "PHX", "PHX", "ARI", "ARI", "UTA", "UTA"),
  season = c(1980, 1991, 1997, 2004, 2015, 2022, 2025, 2026)
) |>
  inner_join(transmute(lineage, season = season_id %% 10000, team_name), by = "season") |>
  mutate(
    x = row_number(),
    # the records write "Winnipeg Jets (1979)" to tell the two Jets apart
    label = paste0(season - 1, "-", substr(season, 3, 4), "\n", sub(" \\(\\d{4}\\)$", "", team_name))
  )

ggplot(marks, aes(x, 1)) +
  geom_sdv_logos(aes(team = key, season = season), league = "nhl", width = 0.08) +
  geom_text(aes(y = 0.55, label = label), size = 3, lineheight = 0.95, vjust = 1) +
  scale_x_continuous(expand = expansion(add = 0.6)) +
  scale_y_continuous(limits = c(0.2, 1.4)) +
  labs(title = "The same franchise, by season") +
  theme_void(base_size = 12) +
  theme(plot.title = element_text(face = "bold"), plot.margin = margin(8, 8, 8, 8))

A row of eight logos worn by one NHL franchise, left to right: two Winnipeg Jets marks, two Phoenix Coyotes marks, two Arizona Coyotes marks, the Utah Hockey Club mark and the Utah Mammoth mark, each labelled with its season and team name.

The NHL’s records list the Utah club as a new franchise from 2024-25, so its two seasons are picked by team ID along with the earlier ones. The franchise’s points percentage across all 46 seasons, one shade per identity:

lineage <- lineage |>
  mutate(season = season_id %% 10000, pts_pct = points / (2 * games_played)) |>
  arrange(season)
bands <- lineage |>
  summarise(from = min(season) - 0.5, to = max(season) + 0.5, .by = tri_code) |>
  mutate(shade = row_number() %% 2 == 1, mid = (from + to) / 2, mark_season = to - 0.5)

ggplot(lineage, aes(season, pts_pct)) +
  geom_rect(
    aes(xmin = from, xmax = to, ymin = -Inf, ymax = Inf, fill = shade),
    data = bands, inherit.aes = FALSE
  ) +
  geom_hline(yintercept = 0.5, linetype = "dashed", colour = "grey50") +
  geom_line(colour = sdv_team_colors("nhl", "UTAH"), linewidth = 0.9) +
  geom_point(colour = sdv_team_colors("nhl", "UTAH"), size = 1.6) +
  geom_sdv_logos(
    aes(x = mid, y = 0.8, team = tri_code, season = mark_season),
    data = bands, league = "nhl", height = 0.13, inherit.aes = FALSE
  ) +
  scale_fill_manual(values = c(`TRUE` = "grey93", `FALSE` = NA), guide = "none") +
  scale_x_continuous(breaks = c(1980, 1990, 2000, 2010, 2020, 2026)) +
  scale_y_continuous(labels = scales::label_number(accuracy = 0.001), limits = c(0.15, 0.87)) +
  labs(
    title = "Winnipeg to Phoenix to Arizona to Utah",
    subtitle = "Regular-season points percentage; each band's logo is the identity's last mark",
    x = "Season (named for the year it ends)",
    y = "Points percentage",
    caption = "Data: NHL records API via fastRhockey | Viz: sdvplotR"
  ) +
  theme_minimal(base_size = 12) +
  theme(panel.grid.minor = element_blank(), plot.title = element_text(face = "bold"))

Line chart of the Winnipeg Jets, Phoenix Coyotes, Arizona Coyotes and Utah franchise's points percentage in every season from 1979-80 to 2025-26, on shaded bands for each identity with the identity's logo at the top of its band and a dashed line at .500.

9. The points race, by division

Standings points game by game (two for a win, one for an overtime or shootout loss), measured against a .500 pace of one point a game so the lines spread apart instead of all climbing together. The box scores have the score, the play-by-play says which games went past regulation, and the totals match the official standings for every team.

Division rivals often share a color, so each line takes its team’s primary color from sdv_team_colors() unless that is too close to one already used in its division, and then its secondary.

last_period <- summarise(pbp, last_period = max(period), .by = game_id)
stopifnot(identical(class(team_box$game_id), class(last_period$game_id)))

race <- team_box |>
  filter(game_type == 2) |>
  inner_join(last_period, by = "game_id") |>
  mutate(pts = case_when(goals > goals_against ~ 2, last_period > 3 ~ 1, .default = 0)) |>
  arrange(game_date, game_id) |>
  mutate(game = row_number(), points = cumsum(pts), .by = team_abbrev) |>
  inner_join(
    select(standings, team_abbr, team, division_name, division_sequence, official = points),
    by = c(team_abbrev = "team_abbr")
  ) |>
  mutate(above = points - game, division = factor(division_name, levels = nhl_divisions))
final <- filter(race, game == 82)
stopifnot(nrow(final) == 32, all(final$points == final$official))

# the primary, unless it is within 20 (CIE 2000 color difference) of the white
# background or of a color already taken, and the secondary is farther away
distinct_colours <- function(primary, secondary, min_gap = 20) {
  taken <- "#FFFFFF"
  gap <- function(x) {
    min(farver::compare_colour(
      farver::decode_colour(x), farver::decode_colour(taken), "rgb",
      method = "cie2000"
    ))
  }
  for (i in seq_along(primary)) {
    swap <- gap(primary[i]) < min_gap && gap(secondary[i]) > gap(primary[i])
    taken <- c(taken, if (swap) secondary[i] else primary[i])
  }
  taken[-1]
}
line_colours <- final |>
  arrange(division, division_sequence) |>
  inner_join(sdv_team_colors("nhl", which = "all"), by = c(team = "team_id")) |>
  mutate(colour = distinct_colours(primary, secondary), .by = division)

# teams that finish close together would stack their logos, so spread the logos
# at least `gap` points apart (keeping their order and average) and draw a
# connector from each line's end to its logo
spread <- function(y, gap) {
  o <- order(y)
  s <- y[o]
  for (i in seq_along(s)[-1]) s[i] <- max(s[i], s[i - 1] + gap)
  y[o] <- s - mean(s) + mean(y)
  y
}
final <- mutate(final, logo_y = spread(above, gap = 6), .by = division)

ggplot(race, aes(game, above, colour = team)) +
  geom_hline(yintercept = 0, colour = "grey60") +
  geom_line(linewidth = 0.8) +
  geom_segment(aes(x = 82, xend = 86, yend = logo_y), data = final, linewidth = 0.3) +
  # inherit.aes = FALSE: a colour aesthetic would tint the logos
  geom_sdv_logos(
    aes(x = 89, y = logo_y, team = team),
    data = final, league = "nhl", height = 0.075, inherit.aes = FALSE
  ) +
  scale_colour_manual(values = setNames(line_colours$colour, line_colours$team)) +
  scale_x_continuous(breaks = c(1, 20, 40, 60, 82), limits = c(1, 92)) +
  facet_wrap(~division, ncol = 2) +
  labs(
    title = "The 2025-26 points race, by division",
    subtitle = "Standings points above a .500 pace (one point a game)",
    x = "Game",
    y = "Points above .500 pace",
    caption = release
  ) +
  theme_minimal(base_size = 12) +
  theme(
    legend.position = "none",
    panel.grid.minor = element_blank(),
    strip.text = element_text(face = "bold"),
    plot.title = element_text(face = "bold")
  )

Four small line charts, one per NHL division, showing each team's standings points above a .500 pace game by game through the 2025-26 regular season, lines in team colors, with the team logo at the end of each line.

10. The playoffs as a tier list

How far every team went, from the playoff game IDs (the seventh digit is the round), with regular-season points deciding the order within a tier. sdv_team_tiers() draws each team’s dark-background logo on its dark background, so the Capitals’, Lightning’s and Maple Leafs’ marks stay visible.

reached <- team_box |>
  filter(game_type == 3) |>
  summarise(round = max(game_id %/% 100 %% 10), .by = team_abbrev)

tiers <- standings |>
  left_join(reached, by = c(team_abbr = "team_abbrev")) |>
  mutate(tier_no = case_when(
    team_abbr == champion ~ 1,
    is.na(round) ~ 6,
    .default = 6 - round
  )) |>
  arrange(tier_no, desc(points), team) |>
  mutate(tier_rank = row_number(), .by = tier_no)
stopifnot(sum(tiers$tier_no < 6) == 16)

sdv_team_tiers(
  select(tiers, tier_no, tier_rank, team),
  league = "nhl",
  title = "The 2026 Stanley Cup playoffs: how far everyone got",
  subtitle = paste(champ_name, "won the Cup; teams ordered by regular-season points within each tier"),
  caption = release,
  tier_desc = c(
    "1" = "Champion",
    "2" = "Lost the Final",
    "3" = "Conference final",
    "4" = "Second round",
    "5" = "First round",
    "6" = "Missed the playoffs"
  ),
  width = 0.05
)

Tier list of the 32 NHL teams in 2025-26, drawn as logos in six rows: the Stanley Cup champion, the other finalist, the conference finalists, the second-round teams, the first-round teams and the teams that missed the playoffs.

11. Make it interactive

Section 3’s expected-goals scatter, built with native plotly. A transparent marker under each team carries the hover text, and add_sdv_logos() places the logos as layout images at the same coordinates. Hover a logo for the team’s five-on-five expected goals for and against per game and its share of the two, the numbers the static chart can only place.

xg_hover <- xg |>
  left_join(select(standings, team, team_name), by = "team") |>
  mutate(tip = sprintf(
    "<b>%s</b><br>5v5 xG for: %.2f per game<br>5v5 xG against: %.2f per game<br>xG share: %.1f%% (%d games)",
    team_name, xgf_pg, xga_pg, 100 * xgf / (xgf + xga), gp
  ))

pad <- 0.1
avg_line <- list(color = "grey", dash = "dash", width = 1)

plotly::plot_ly(
  xg_hover, x = ~xgf_pg, y = ~xga_pg, type = "scatter", mode = "markers",
  marker = list(size = 30, opacity = 0), text = ~tip, hoverinfo = "text"
) |>
  plotly::layout(
    title = list(text = "Who drove play at five-on-five, 2025-26", x = 0.02),
    xaxis = list(title = "5v5 xG for per game", range = range(xg_hover$xgf_pg) + c(-pad, pad), zeroline = FALSE),
    yaxis = list(
      title = "5v5 xG against per game (reversed)", zeroline = FALSE,
      range = rev(range(xg_hover$xga_pg) + c(-pad, pad))
    ),
    shapes = list(
      list(type = "line", x0 = mean(xg_hover$xgf_pg), x1 = mean(xg_hover$xgf_pg), yref = "paper", y0 = 0, y1 = 1, line = avg_line),
      list(type = "line", y0 = mean(xg_hover$xga_pg), y1 = mean(xg_hover$xga_pg), xref = "paper", x0 = 0, x1 = 1, line = avg_line)
    )
  ) |>
  add_sdv_logos(xg_hover$xgf_pg, xg_hover$xga_pg, xg_hover$team, "nhl", size = 0.08)

A range given high to low reverses the axis, as scale_y_reverse() did in section 3. plotly loads each logo from its URL, so the page stays light. Interactive charts covers plotly, ggiraph and leaflet, and the Interactive NHL Shot Map recipe makes a shot map like section 4’s interactive.

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}
}

sdvplot - the Python plotting package
sdvplot-js - the JavaScript / TypeScript plotting packages
cfbfastR - college football
hoopR - men’s basketball
wehoop - women’s basketball
baseballr - baseball
fastRhockey - hockey
oddsapiR - betting odds
sportyR - playing surfaces
sportsdataverse-py - the Python package
sportsdataverse-R - the R meta-package

Saiem Gilani