A single-screen read on any AFL round, built to answer "what happened, who starred, and what did it change?" at a glance. Round Headlines cover completion status, total scoring, average margin vs the previous round, and a margin-spread breakdown (blow-outs / standard / close). Player Headlines surface the round's leaders for rating, goals, disposals, score involvements and tackles, each with season-average context and a form sparkline. Round Climbers and Sliders plots every team on a season attack-vs-defence scatter, with movement trails showing how the round shifted them toward or away from the top-six contention zone. Below: per-game match-up cards (score worms and key-metric splits), team totals by match, top/bottom five rated players, and an interactive Player Performance Analysis scatter for exploring any two player metrics. A global round stepper works across every season. Built on analytics.v_player_match and v_team_matchup, with shots, fixtures and quarter scores from marts.

AFL Analyst | Round Summaries

Build me a single-screen dashboard that gives an at-a-glance overview of an AFL round from my roar_metrics_cloud data — completion status, scoring, competitiveness, player leaders, match-up breakdowns, and an interactive scatter. Default to the latest completed round with a stepper across all seasons.

Reorganise the KPI area into three columns that tell a zoom-out story: round-level team stats, then player headlines, then a season-level view on the far right. Drop the redundant tiles and widen the right column.

Add a premiership-window scatter showing each team on attack vs defence, with the top-6-in-both as the contention zone and trails showing how each round moves them. Include teams on a bye, add tooltips, repel overlapping labels, and handle ties correctly.

Colour the teams by how the round shifted them — climbers (better at both ends) blue, sliders (worse at both) red, mixed neutral — and rename it 'Round Climbers and Sliders'.

Tidy the titles, formatting, labels and captions to a consistent style, fix the sparkline and label bugs, and set up a safe, self-contained database so I can share the dashboard publicly.

Darragh Murray

Hand-drawn WSDOT cartoon map with hover hotspots that surface live WSDOT Traveler Information API data (traffic flow, alerts, mountain passes, border waits, travel times, tolls, cameras, E-WA weather). Backed by the wsdot database, refreshed every 5 min by the wsdot-traveler-sync flight.

WSDOT — Paint Diagram (Live Traveler API)

Using this map created by the WSDOT, I want to now be able to use WSDOT APi's via Motherduck Flights to connect data that refreshes on a 5 min cadence. Allow the user to interact with the map and hover over specific items on the map that then surfaces API information from WSDOT  - Output will be a motherduck dive

WSDOT Live Feeds for Travel Times, Border Waits, and Cameras

Aaron Tham

SpongeBob-themed exploration of 4,731 PADI dive sites across 119 countries. Includes top countries, dive types, depth distribution, marine life sightings, and average depth by type — all with custom SVG illustrations of a scuba duck, pirate ship, treasure chest, and tropical island.

Are Ya Ready, Divers?

Dumky de Wilde

Eval dashboard — modern calm-white theme. Compare jobs across accuracy, cost, tools & tokens; correctness matrix with A/B trace diff; cost-vs-latency outliers.

Eval Explorer (v2)

Jacob Matson

If you are looking for a declarative way to define your charts alongside your dbt pipelines, look no further! Now you can add a dbt package to output your charts as MotherDuck Dives with just 2 config changes. These will even refresh live, while typically dbt Charts are static! 

This Dive embeds the Vega Lite plotting library and pulls the chart specs from a MotherDuck table that is updated as a part of the dbt package during `dbt build`. 

To use this in your own pipelines, grab the package from here: https://github.com/motherduckdb/labs/tree/main/projects/dbt-charts-dives 

dbt: Product Performance

dbt Charts in MotherDuck Dives!

alex+devrel@motherduck.com

Financial statements (P&L, balance sheet, cash movement) read from the DuckLake database finance_books at a named commit (AT (VERSION => n)), pinned to snapshot 2. Commit picker from finance.close_log, compare mode shades cells that differ between two commits and lists the ledger rows posted between them (posted_in), and an Explain panel traces any cell to SQL, rows, and GL entries at that commit. Part of 0903-finance-stable-dives.

Financial statements

A histomap of European armed force, 1720–1871 — adapted from John B. Sparks's flowing-band form (Rand McNally, 1931) to show absolute combined land and naval strength across seven great powers. Unlike Sparks's proportional original, the rising ember crest is itself data: 931,000 men in 1720 to 4.2 million by 1870, a 4.5× growth driven by 26 interactive sparks tracing the institutional, technological, ideological, and competitive causes behind one of history's largest expansions in state military capacity.

Kindling for Mass War

Mehrab Rahman

Compare the cost vs. accuracy and runtime vs. accuracy of frontier cloud LLMs and local LLMs (including Qwen3.8 27B) on the DABStep benchmark.

Agentic SQL Benchmark Including Local Models

Alex Monahan

Artist and credit explorer over genius_music: top artists/producers/writers, search, and recursive collaborator click-through.

Dive Gallery