Render kinds and options

The trailing RENDER <kind> [WITH (…)] clause decides how a result is displayed. A query with no RENDER clause renders as a table.

Text kinds

Kind	Name	Displays
list	List	Bulleted text list.
table	Table	Plain data table.
leaderboard	Leaderboard	Ranked leaderboard text, optionally @mentioning top rows.

leaderboard is the only kind that @mentions people, and it takes its own option: mentions = <n> pings the top n ranked rows, mentions = all pings every eligible row, and mentions = 0 pings nobody. Omitted, it pings the top three. Rows grouped by anything other than a player never mention anyone, even when a champion name happens to match an alias.

Chart kinds

Kind	Name	Displays
bar_chart	Bar chart	Categorical bar chart.
line_chart	Line chart	Categorical or temporal line chart.
stacked_bar	Stacked bar	Stacked multi-output bar chart.
area_chart	Area chart	Filled categorical or temporal trend.
donut_chart	Donut chart	Part-to-whole chart for bounded rows.
scatter_chart	Scatter chart	Compare two outputs, optionally sized by a third.
heatmap	Heatmap	Two-dimensional matrix colored by an output.
radar_chart	Radar chart	Compare three to eight normalized outputs.
kpi_card	KPI card	One aggregate row displayed as metric cards.
bump_chart	Bump chart	Rank-position movement over time.
calendar_heatmap	Calendar heatmap	Daily activity or performance on a calendar grid.
histogram	Histogram	Distribution over a numeric bucket grouping, such as FLOOR(game_duration_seconds / 300) * 300.
box_plot	Box plot	Five-number summary per group: min, q1, median, q3 and max, named in that order as y.

Charts are rendered to an image and posted as an attachment, so the destination channel needs the Attach Files permission.

WITH options

Option	Syntax	Effect
x	x = <output>	Horizontal channel; defaults to the grouping.
y	y = <output> | (a, b, …)	The numeric series, up to eight.
series	series = <grouping>	Which grouping splits into series.
size	size = <output>	Sizes scatter points.
value	value = <output>	The cell value of a heatmap.
title	title = '…'	Overrides the report's title on the image.
subtitle	subtitle = '…'	A line under the title.
x_axis	x_axis = '…'	Label for the horizontal axis.
y_axis	y_axis = '…'	Label for the vertical axis.
theme	theme = <name>	Chart theme: one of lol_dark, lol_light, minimal_dark, minimal_light.
palette	palette = <name>	Series colours: one of ranked, categorical, team, gold, colorblind.
colors	colors = (#rrggbb, …)	One to eight custom colours, contrast-corrected for the theme.
orientation	orientation = <name>	Bar direction: one of horizontal, vertical.
labels	labels = <name>	Value labels on marks: one of auto, show, hide, value, percent.
legend	legend = <name>	Legend placement: one of auto, none, top, right, bottom.
sort	sort = <name>	Visual row order without changing the query: one of query, asc, desc.
smooth	smooth = true	Curves line and area segments.
rolling	rolling = <n>	Plots a rolling average over n buckets (2–52) instead of raw values.
cumulative	cumulative = true	Accumulates a running total; only valid for additive outputs.
stack	stack = <name>	Stacking for stacked_bar and area_chart: one of none, normal, percent.
trend	trend = true	Overlays a linear trend line.
annotations	annotations = true	Marks notable points on the series.
sparkline	sparkline = true	Attaches a small chart image to a table render.
compare	compare = previous_period	Overlays the equally long period immediately before the window.
format	format = (<output> = <kind>, …)	Overrides an output's inferred display kind: one of count, decimal, percent, duration, ratio, text, timestamp.

Every name inside WITH (…) must be one of the query’s own outputs or groupings. A name the SELECT does not produce is rejected when the query compiles rather than silently plotting nothing.

What each kind needs

Kind	Requires
bar_chart, stacked_bar	y; optionally orientation
line_chart, area_chart	y; a DATE_TRUNC grouping for a time axis
donut_chart	exactly one y, and few rows
scatter_chart	one x and one y, both SELECTed; optionally size
heatmap	value, series, and exactly two groupings
radar_chart	three to eight y outputs over one grouping
kpi_card	no GROUP BY — one grand-total row
bump_chart	one y, a DATE_TRUNC grouping, plus one more grouping to rank
calendar_heatmap	one y and DATE_TRUNC('day', …) as the only grouping
histogram	one numeric bucket grouping and one count y
box_plot	exactly five y outputs — see below
leaderboard	optionally mentions

Comparing against the previous period

compare = previous_period re-runs the same aggregation over the equally long span immediately before the query’s window and overlays it, giving every point a baseline, an absolute delta, and a percentage delta.

It needs two things, and says so if either is missing: a stated time window to take the preceding period of, and a DATE_TRUNC (or patch) grouping to line the two runs up bucket by bucket.

SELECT DATE_TRUNC('week', game_creation_at) AS week,
       COUNT(*) FILTER (WHERE win) AS wins,
       COUNT(*) FILTER (WHERE NOT win) AS losses
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 90 DAY
GROUP BY DATE_TRUNC('week', game_creation_at)
ORDER BY week ASC
RENDER stacked_bar WITH (y = (wins, losses), compare = previous_period)

previous_period is the only comparison in the language.

The box-plot contract

A box plot takes its five numbers explicitly, and their order in y is the encoding: minimum, first quartile, median, third quartile, maximum.

SELECT MIN(kills) AS low,
       QUANTILE_CONT(kills, 0.25) AS q1,
       MEDIAN(kills) AS med,
       QUANTILE_CONT(kills, 0.75) AS q3,
       MAX(kills) AS high
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY champion
ORDER BY med DESC
LIMIT 6
RENDER box_plot WITH (y = (low, q1, med, q3, high))

The names are yours; the positions are not. Nothing is inferred from what a column is called, which is the same bargain the rest of the language makes — q3 could as easily be a 95th percentile, and only the query knows.

Every kind, rendered

Each sample below is produced by compiling the query shown and running it through Scout’s actual report renderer — the same function that builds what the bot posts. They are regenerated by bun run generate, so they cannot drift from the renderer.

table

SELECT COUNT(*) AS games, AVG(win::INT) AS win_rate, kda() AS kda
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY player
ORDER BY games DESC
LIMIT 6
RENDER table

Posted to Discord

**Most games played**
```
Player | Games | Win Rate | Kda
--- | --- | --- | ---
Faker | 146 | 57.5% | 4.12
Caps | 111 | 51.4% | 3.48
Chovy | 90 | 54.4% | 3.91
Ruler | 89 | 51.7% | 3.36
Keria | 76 | 46.1% | 2.87
Zeus | 71 | 53.5% | 3.02
```

list

SELECT COUNT(*) AS games, AVG(win::INT) AS win_rate
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY player
ORDER BY games DESC
LIMIT 6
RENDER list

Posted to Discord

**Most games played**
- Faker: Games: 146, Win Rate: 57.5%
- Caps: Games: 111, Win Rate: 51.4%
- Chovy: Games: 90, Win Rate: 54.4%
- Ruler: Games: 89, Win Rate: 51.7%
- Keria: Games: 76, Win Rate: 46.1%
- Zeus: Games: 71, Win Rate: 53.5%

leaderboard

SELECT COUNT(*) AS games, AVG(win::INT) AS win_rate
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY player
ORDER BY games DESC
LIMIT 6
RENDER leaderboard WITH (mentions = 0)

Posted to Discord

**Most games played**
1. Faker — Games: 146, Win Rate: 57.5%
2. Caps — Games: 111, Win Rate: 51.4%
3. Chovy — Games: 90, Win Rate: 54.4%
4. Ruler — Games: 89, Win Rate: 51.7%
5. Keria — Games: 76, Win Rate: 46.1%
6. Zeus — Games: 71, Win Rate: 53.5%

bar_chart

SELECT COUNT(*) AS games
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY player
ORDER BY games DESC
LIMIT 6
RENDER bar_chart WITH (y = games)

line_chart

SELECT DATE_TRUNC('week', game_creation_at) AS week, COUNT(*) AS games
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY DATE_TRUNC('week', game_creation_at)
ORDER BY week ASC
RENDER line_chart WITH (y = games, x_axis = 'Week', smooth = true)

area_chart

SELECT DATE_TRUNC('week', game_creation_at) AS week, COUNT(*) AS games
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY DATE_TRUNC('week', game_creation_at)
ORDER BY week ASC
RENDER area_chart WITH (y = games, x_axis = 'Week', smooth = true)

stacked_bar

SELECT DATE_TRUNC('week', game_creation_at) AS week, COUNT(*) FILTER (WHERE win) AS wins, COUNT(*) FILTER (WHERE NOT win) AS losses
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY DATE_TRUNC('week', game_creation_at)
ORDER BY week ASC
RENDER stacked_bar WITH (y = (wins, losses))

donut_chart

SELECT COUNT(*) AS games
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY queue
ORDER BY games DESC
RENDER donut_chart WITH (y = games)

scatter_chart

SELECT SUM(total_damage_dealt_to_champions) AS damage, kda() AS kda, COUNT(*) AS games
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY player
RENDER scatter_chart WITH (x = damage, y = kda, size = games)

heatmap

SELECT AVG(win::INT) AS win_rate
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY champion, team_position
RENDER heatmap WITH (value = win_rate, series = team_position, palette = gold)

radar_chart

SELECT AVG(win::INT) AS win_rate, kda() AS kda, per_minute(creep_score) AS cs_per_minute, AVG(vision_score) AS vision_score, AVG(total_damage_dealt_to_champions) AS damage
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY player
LIMIT 3
RENDER radar_chart WITH (y = (win_rate, kda, cs_per_minute, vision_score, damage))

kpi_card

SELECT COUNT(*) AS games, AVG(win::INT) AS win_rate, kda() AS kda, AVG(game_duration_seconds) AS avg_length
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
RENDER kpi_card WITH (y = (games, win_rate, kda, avg_length))

bump_chart

SELECT COUNT(*) AS games
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 28 DAY
GROUP BY player, DATE_TRUNC('week', game_creation_at)
ORDER BY games DESC
RENDER bump_chart WITH (y = games)

calendar_heatmap

SELECT COUNT(*) AS games
FROM match_participants
WHERE (game_creation_at AT TIME ZONE 'UTC')::DATE BETWEEN '2026-06-29' AND '2026-07-26'
GROUP BY DATE_TRUNC('day', game_creation_at)
ORDER BY games DESC
RENDER calendar_heatmap WITH (y = games, palette = gold)

histogram

SELECT FLOOR(game_duration_seconds / 300) * 300 AS bucket, COUNT(*) AS games
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY FLOOR(game_duration_seconds / 300) * 300
ORDER BY bucket ASC
RENDER histogram WITH (x = bucket, y = games)

box_plot

SELECT MIN(kills) AS low, QUANTILE_CONT(kills, 0.25) AS q1, MEDIAN(kills) AS med, QUANTILE_CONT(kills, 0.75) AS q3, MAX(kills) AS high
FROM match_participants
WHERE game_creation_at >= CURRENT_TIMESTAMP - INTERVAL 30 DAY
GROUP BY champion
ORDER BY med DESC
LIMIT 6
RENDER box_plot WITH (y = (low, q1, med, q3, high))

Related

• Turn a report into a chart
• Query recipes
• ScoutQL