Chronicles of Denver's Shadows

"In the data, patterns emerge like constellations—except these constellations steal cars."

467,928 incidents analyzed • 19 dimensions explored • 0 crimes committed in the making of this notebook

Having lived in Denver until recently, I did what any reasonable person does after moving away: downloaded the city's entire public crime dataset and refused to sleep until it made sense. Armed with Python and the confidence of a wizard who has not yet seen the data quality, I waded into nearly half a million incidents spanning years of urban life.

The mission was genuinely serious: organize this chaotic sea of records so Denver's law enforcement could allocate resources more efficiently. To predict where the shadows fall next, you first have to map where they have been. The shadows, for the record, did not cooperate.

This is the chronicle of that journey—a tale of data cleaning, pattern discovery, and the truths that surface when you stare at a CSV long enough for it to stare back.

Chapter I: The Purification Ritual

Before any wizard divines the future, the dishes must be done. The raw data arrived like an ancient scroll—467,928 entries across 19 columns—and, like most ancient scrolls, it was a mess. Each offense carried an ID built from incident numbers, offense codes, and extensions. Some bore precise timestamps; others spoke only in vague ranges, like witnesses who would rather not get involved.

Anomalies lurked within: 41 entries where the crime ended before it began—actual time travel, per the spreadsheet—which had to be banished. Coordinates vanished for privacy; certain dates simply refused to align, on principle. Each error was cataloged, corrected, or excised with surgical precision and only moderate muttering.

Purification complete, I filtered out traffic incidents to focus on true crimes. What remained was a pristine dataset—clean, obedient, and ready to confess everything.

The Grimoire's Contents