Blog, page 14

Field notes — one a day since May 2023. Things we shipped, things we broke, and numbers we went and checked.

2026 2025 2024 2023

1,200 posts. , or browse by year and tag below.

Sep 30, 2025

Four charts I deleted from this site

Each one was accurate and each one made a claim the data did not support.

visualization Visualization

Sep 29, 2025

A essentially none correlation, and what it is not

A measured correlation of -0.02 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Sep 28, 2025

Correction: Median beside the mean

The page claimed more than it should have. What it says now, and where the pattern stops applying.

quality practice Statistics

Sep 27, 2025

Charting sessions by device

Two facts about the columns decide the chart. Cardinality and skew, in that order.

visualization Visualization

Sep 26, 2025

What Regression Lab will not do

Fit an ordinary least squares line to any pair of columns and get the diagnostics that decide whether the fit means anything: residual plot, R², standard errors, and leverage.

visualization Visualization

Sep 25, 2025

Ride-hail trips, in one paragraph

Eight thousand city ride-hail trips with fares, distances, surge multipliers and tips.

quality practice Data quality

Sep 24, 2025

Course note — Designing an experiment that can succeed

Most experiments fail before launch, because they were never large enough to detect the effect anyone hoped for. The fix costs ten minutes and happens before you write code.

practice Experimentation

Sep 23, 2025

Pattern: Median beside the mean

Report both. The gap between them is a free skew diagnostic, and it tells the reader which number to trust.

statistics Statistics

Sep 22, 2025

The time coverage of movie-ratings

901 populated days across a 901-day span, and what that allows you to compute.

engineering quality Data engineering

Sep 21, 2025

The category that eats the chart

7 values with 45.1% concentrated in one of them, and what that does to every chart downstream.

visualization practice Visualization

Sep 20, 2025

Keeping one row per key in ride-hail-trips

deduplicate by rank against a real schema, and the clause in it that is doing the work.

sql Analytics engineering

Sep 19, 2025

Five minutes with origin in flight-delays

8,000 rows, 14 distinct values, and the one fact about origin that changes how you query it.

quality practice Data quality

Sep 18, 2025

The average seats now is not the typical one

A 635.5% gap between mean and median on one column, and which of the two answers the question you were asked.

statistics practice Statistics

Sep 17, 2025

Is device_id + reading_at the grain of sensor-telemetry?

8,000 distinct combinations against 8,000 rows, and what that answers about the table's key.

sql quality warehousing Analytics engineering

Sep 16, 2025

Before you clip the outliers

A max 1.1× the 99th percentile, and the three different things that could be.

statistics quality Statistics

Sep 15, 2025

dep_delay_min by dest: a 38% spread

Real group means for one column across 14 segments, and what the pooled average conceals.

statistics practice Analytics practice

Sep 14, 2025

Charting seats_now by plan

Two facts about the columns decide the chart. Cardinality and skew, in that order.

visualization Visualization

Sep 13, 2025

A strong correlation, and what it is not

A measured correlation of 0.92 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Sep 12, 2025

Correction: Date spine

The page claimed more than it should have. What it says now, and where the pattern stops applying.

quality practice Analytics engineering

Sep 11, 2025

Experiment Calculator works offline

Sample size before you launch, and a confidence interval after you stop. Both directions, with the assumptions written on the screen rather than buried in a footnote.

visualization Visualization

Sep 10, 2025

churned_at runs 2023-03-17 to 2026-01-24

624 populated days across a 1,045-day span, and what that allows you to compute.

engineering quality Data engineering

Sep 9, 2025

What is deliberately wrong with ride-hail-trips

Eight thousand city ride-hail trips with fares, distances, surge multipliers and tips.

quality practice Data quality

Sep 8, 2025

Course note — Writing SQL people can read, and reading the plan

The query you write twice a year is written for the machine. The one in a production model is written for the person debugging it at 3am, who might be you.

practice Analytics engineering

Sep 7, 2025

Gaps and islands, in short

Group consecutive runs — of dates, of statuses, of anything — by subtracting a row number from the value.

sql Analytics engineering