Pattern Essays — Pattern is Everything https://patterniseverything.com/essays/ Short visual essays on patterns in machine learning, markets, and the world. en Sat, 18 Jul 2026 15:41:50 GMT The Deep Kalman Filter https://patterniseverything.com/essays/#essay-kalman https://patterniseverything.com/essays/#essay-kalman Wed, 10 Jun 2026 18:04:17 GMT You cannot measure everything directly. The Kalman filter estimates a hidden state by blending what it predicts with what it noisily measures — and the deep version learns the model from data. Survivorship Bias https://patterniseverything.com/essays/#essay-survivor https://patterniseverything.com/essays/#essay-survivor Fri, 29 May 2026 18:56:40 GMT We study what survived and forget what did not. The missing data is often where the real lesson hides. The Fractal https://patterniseverything.com/essays/#essay-fractal https://patterniseverything.com/essays/#essay-fractal Fri, 29 May 2026 18:56:40 GMT Zoom in and the pattern repeats. Self-similarity across scales is one of nature’s most common signatures. Simpson’s Paradox https://patterniseverything.com/essays/#essay-simpson https://patterniseverything.com/essays/#essay-simpson Fri, 29 May 2026 18:56:40 GMT A trend can point one way in every group and the opposite way when the groups are combined. Aggregation can lie. The Bell in Everything https://patterniseverything.com/essays/#essay-bell https://patterniseverything.com/essays/#essay-bell Fri, 29 May 2026 12:05:45 GMT Normal distributions emerge wherever many small, independent forces combine. Regression to the Mean https://patterniseverything.com/essays/#essay-mean https://patterniseverything.com/essays/#essay-mean Fri, 29 May 2026 12:05:45 GMT Extreme outcomes tend to be followed by less extreme ones — not a force, just arithmetic. The Long Tail https://patterniseverything.com/essays/#essay-tail https://patterniseverything.com/essays/#essay-tail Fri, 29 May 2026 12:05:45 GMT Most things are small. A few are enormous. The pattern repeats across domains that seem unrelated. Signal in the Noise https://patterniseverything.com/essays/#essay-signal https://patterniseverything.com/essays/#essay-signal Fri, 29 May 2026 12:05:45 GMT Every dataset is a mix of pattern and randomness. The hard part is not inventing signal where there is none. The Map and the Territory https://patterniseverything.com/essays/#essay-map https://patterniseverything.com/essays/#essay-map Fri, 29 May 2026 12:05:45 GMT A model is a deliberate simplification. The danger is forgetting what was left out. The Feedback Loop https://patterniseverything.com/essays/#essay-feedback https://patterniseverything.com/essays/#essay-feedback Fri, 29 May 2026 12:05:45 GMT When a system's output feeds back into its input, small nudges can cascade into enormous change — or freeze everything in place. The Random Walk https://patterniseverything.com/essays/#essay-walk https://patterniseverything.com/essays/#essay-walk Fri, 29 May 2026 12:05:45 GMT Each step is random, yet the path that emerges is not without structure. Distance grows — just not in the direction you expect. The Threshold https://patterniseverything.com/essays/#essay-threshold https://patterniseverything.com/essays/#essay-threshold Fri, 29 May 2026 12:05:45 GMT Many systems stay quiet for a long time, then change all at once. The threshold is the hidden line that separates gradual from sudden.