statistics

Jake Vanderplas's PyCon 2016 talk shows how sampling-based methods such as simulation, shuffling and bootstrapping make statistics intuitive. His pitch is that if you can write a for-loop, you can do statistics.

Lecture 1 of Harvard's Statistics 110 with Joe Blitzstein, introducing sample spaces, the naive definition of probability, counting, binomial coefficients and the sampling table.

Hans Rosling uses data and animated statistics to show how world population is changing, why birth rates fall as child survival and wealth rise, and why population is expected to peak around 11 billion. It also tackles common misconceptions about Africa and family planning.

In this TED talk, Hans Rosling uses animated statistics to debunk myths about the developing world, showing how health and income have improved across countries.

Veritasium explains why many real-world systems follow power laws rather than normal distributions. It covers self-organized criticality, with simulations of sandpiles, magnetism and forest fires.

Physicist Angela Collier rants for 42 minutes about why violin plots are a poor way to show data, citing papers on violin and bean plots. It is part lecture on how to read distributions and part opinionated critique.

Matt Parker analyzes the Minecraft speedrunning controversy in which Dream was accused of cheating via improbably lucky drops. He walks through the probability and the statistical arguments in the original accusation and the reply.

Joel Grus's JupyterCon talk explaining why he dislikes Jupyter notebooks. He covers hidden state, trouble with reproducibility and testing, and how notebooks frustrate teaching. He shows his preferred workflow and what Jupyter could change.

John Kitchin shows how Emacs org-mode with Python lets researchers combine narrative, math, figures and executable code in one document. It exports to LaTeX, HTML and PDF, and he used it for papers and e-books.

Jeremy Howard defends Jupyter notebooks as a serious development environment, especially when paired with nbdev. He responds directly to Joel Grus's talk about why he dislikes notebooks, and contrasts REPL-style and notebook-style work.

A reupload of Joel Spolsky's talk teaching spreadsheet users the Excel features they are probably missing, such as keyboard navigation, named ranges, tables and lookups. The original disappeared and this copy comes from the Wayback Machine.

Database pioneer Michael Stonebraker argues that 'big data' is really at least four different problems: big volumes with small analytics, big analytics, velocity, and variety. He comments on where each market is heading.

Fons van der Plas introduces Pluto.jl, a reactive notebook for Julia that tracks references between cells and re-runs affected cells when code changes. The talk was given at JuliaCon 2020.

Dan Robinson, then CTO of Heap, explains how Heap runs its product analytics backend on PostgreSQL and CitusDB. It covers conversion funnels, retention and cohort queries across tens of billions of events and 50 TB of data, plus Postgres and Kafka together and the indexing strategy.

David Beazley's PyData Global 2021 keynote, an oral history of early scientific Python. He tells how Python ended up running on a Connection Machine supercomputer at Los Alamos.

Sandy Pentland of MIT presents his book on social physics, which uses big-data studies of how people interact to explain how ideas spread. He describes findings on collective intelligence and social influence.

Robert Erdmann, a materials scientist at the Rijksmuseum, gives a PyCon keynote on the ultra-high-resolution imaging of Rembrandt's Night Watch. He covers the automated gantry camera system, the Python software that drives it, and the large volumes of image data involved.

Peter Norvig's 2007 Google talk on theorizing from data, covering what huge datasets can do for language understanding, translation, information extraction and inference. It argues that more data often beats cleverer models.

Bryan Cantrill presents Manta, an object store with in-situ compute that lets tools like grep, awk, and sed run map-reduce style over huge data. He frames it as bringing the Unix philosophy to big data.

A 49-minute demo of Palantir Foundry by its chief architect. It walks through data integration, ML and AI workflows, analytics, and application building across the platform.

Veritasium explores the Prisoner's Dilemma and Axelrod's tournaments, showing why tit-for-tat and forgiving strategies win in repeated games. It extends the ideas to cooperation in evolution and conflict.

3Blue1Brown explains why positive results on accurate medical tests are often false positives, and reframes Bayes' rule using likelihood ratios (Bayes factors) to update odds. It uses visual examples to make the reasoning intuitive.

Derren Brown's TV special in which he convinces a person he has a system for predicting horse race winners, demonstrating how survivorship bias and selective outcomes can fake predictive powers.

Owen Maitzen's animated hour-long introduction to Hackenbush, a simple combinatorial game, and how its strategy leads to surreal numbers and the roots of the number system.

Sid Meier's GDC 2010 keynote on how players perceive probability in games, including how Civilization's odds were adjusted because players misjudged them.

Veritasium explores why the number 37 turns up so often, drawing on Tom Magliery's collection of 37 sightings and a poll of about 200,000 people asked to pick a random number. It covers how humans choose numbers and why 37 stands out.

Chris Stucchio presents the ethical principles behind algorithmic decision making and shows with examples from lending, criminology and education that they cannot all be satisfied at once. The talk frames AI fairness as a set of quantitative tradeoffs.

Numberphile talks with Grant Sanderson of 3Blue1Brown about a darts puzzle with a shrinking bullseye. Solving it takes the volumes of high-dimensional spheres, and the answer turns up Euler's number.

A Veritasium video on how a feud between Russian mathematicians led to Markov chains and the modern prediction algorithms built on them. It covers the history and the math behind them.

Stephan T. Lavavej's GoingNative 2013 talk explaining why calling rand() and using rand() % 100 is bad in C++, and how the C++11 <random> header does the job properly.

Nassim Nicholas Taleb talks at Google about his book Antifragile, on systems that gain from disorder and volatility. It draws on themes from his earlier books.

Veritasium explains why no voting system can be perfectly fair, drawing on Arrow's theorem and the work of Eric Maskin. It looks at problems with systems like first-past-the-post and ranked-choice voting.

Hans Rosling uses the Gapminder animated-data tool to debunk myths about world development, with sportscaster-like energy. It shows trends in health, income and child survival across countries.

Santa Clara County District Attorney Jeff Rosen explains in a TEDx talk why Germany has lower crime and far fewer prisoners than the US, and what American criminal justice could learn from German prisons.

A cardiologist runs a detailed, skeptical review of the Wim Hof Method, covering the research on cold exposure and breathing techniques. It also teaches how to read scientific statistics critically.

Tom Nicholas critiques Veritasium's sponsored video on driverless cars, using it to examine YouTube sponsorships, science-channel credibility and how statistics and misinformation can be used in branded content.

An animated short on limits to growth, covering oil formation, peak oil, energy, economic growth and resource depletion. Its central section deals with exponential growth, and the author notes some statistics are dated.