Methodology

Every number on this site comes from NYC Open Data's Mayor's Management Report — Agency Performance Indicators dataset, not the printed Mayor's Management Report PDF. That PDF is used only as a secondary source for agency descriptions and definitional text that don't exist in the open dataset — never for the numbers themselves.

This site covers indicators the City itself flags as critical and that are not retired, restricted to citywide totals (the dataset's id = parentid rows — geographic breakdowns by community district, precinct, or school district are excluded). As of the last data refresh, that scope is 609 indicators.

Each indicator shows one value per fiscal year for roughly the last 5 fiscal years, including the current, partial year.

Where an indicator is reported more often than annually, the value shown for a fiscal year is the year-to-date figure from the most recently reported period in that year — the same figure the Mayor's Management Report itself presents as the year's "Actual," not a raw monthly snapshot.

For the current, partial fiscal year, some indicators also show a light dashed projectedmarker — a full-year estimate at the current year-to-date pace. This only appears where the indicator is a running cumulative count (e.g. a total number of complaints or incidents), determined per indicator from a complete prior fiscal year by checking whether that year's final year-to-date figure matches the sum of its reported periods rather than their average. Indicators that are themselves an average or a rate (most percentages, response times, and many "Average X" indicators) don't get one — their year-to-date figure is already the best available estimate, not something to scale up. The projection is shown for reference only: it never feeds the trend or target-status figures elsewhere on this site.

An indicator is on targetif its latest value meets the City's own numeric target in the direction the City itself defines as desirable (for a "lower is better" indicator, on target means at or below the target; for "higher is better," at or above).

Many indicators — including most of NYPD's — don't have a numeric target in this dataset at all; the City tracks a direction (e.g., "reduce crime") without committing to a specific number. Those are labeled no numeric target set, which is a real status, not a data gap.

Trend is fit across every fiscal year available in an indicator's lookback window using the median of all pairwise year-to-year slopes (a Theil–Sen estimator), not a simple least-squares line and not just the latest year compared to the one before it. This is deliberately resistant to a single unusual year dominating the label — a one-year spike or dip in an otherwise flat or declining series won't flip the trend on its own. The direction is judged relative to the City's own stated desired direction, so "improving" always means "getting better," never just "the number went up."

• Non-critical and retired indicators (roughly 1,000+ additional indicators in the full dataset).
• Geographic drill-down — community district, precinct, and school district breakdowns.
• The Mayor's Management Report's "equity indicator" flag — it isn't present in this open dataset, so this site makes no equity-specific claims.
• Historical data older than roughly 5 fiscal years.

NYC Open Data updates this dataset roughly monthly. This site's data was last generated on July 19, 2026.