Papers I Keep Rereading

Andre Coleman

@boxscoreandre

11Items

3Likes

This is my starting five plus a very chaotic bench of papers that I actually go back to, not the ones I pretend to have finished on my Kindle. Heavy on stats, a little econ, a dash of psych, and at least one thing that makes me feel like I’m back in a Boston coffee shop trying to look smart. Think of this as my personal playbook: these are the PDFs I re-open the way some people re-watch Game 7 highlights (guilty on both counts). I tossed in one tiny number for each so you know I really did read past the abstract.

See all 11 items Follow Andre ColemanOpen in app

The list

Showing 1–10 of 11

Papers I Keep Rereading — 10 items
Name	Authors	Year	Field	Why it stuck	My take
1The Probable Error of a Mean	William S. Gosset	1908	Statistics	The origin story of the t-test and the pen name “Student” — like finding out your favorite role player was secretly an MVP-level brewer-statistician.	Any time someone in the office brags about a p-value, I silently thank a guy who just wanted better beer; also, wild that a 1908 paper still gets more usage than most modern dashboards.
2Regression Toward the Mean	Francis Galton	1886	Statistics / Social science	Explains why your breakout rec-league game (or that one insane shooting night) almost never repeats — the math behind “you’re probably not actually Steph.”	Any time I drop 20 in a game, I reread this just to emotionally prepare for the 4-point clunker coming next; it’s basically therapy with scatterplots.
3Least Squares Estimation of Linear Regression Parameters	Carl Friedrich Gauss	1809	Mathematics / Statistics	The OG behind every “let’s just run a quick regression” comment in a meeting; without this, half of analytics jobs (including mine) are just colorful charts and vibes.	Reading Gauss is like watching prime Tim Duncan highlights: boring on the surface, but the fundamentals are so clean you can’t not respect it.
4A Theory of Human Motivation	Abraham H. Maslow	1943	Psychology	The famous needs pyramid — I think about it every time a project fails because nobody’s basic ‘clear requirements’ need was met, let alone ‘self-actualization via cool model.’	Maslow basically proves that until I get 8 hours of sleep and a working coffee machine, I’m not becoming the advanced analytics guru I tell LinkedIn I am.
5Prospect Theory: An Analysis of Decision under Risk	Daniel Kahneman, Amos Tversky	1979	Behavioral economics	Explains why I’ll drive 20 minutes to save $5 on beer but won’t renegotiate my internet bill; humans weight losses about twice as much as gains and I feel personally attacked.	Every fantasy-basketball trade argument I’ve ever had could’ve been shorter if both sides had just admitted we’re loss-averse weirdos and moved on.
6The Spread of Obesity in a Large Social Network over 32 Years	Nicholas A. Christakis, James H. Fowler	2007	Epidemiology / Social networks	Shows how health behaviors ripple through friendships; also the first paper that made me realize my beer-and-wings group chat is basically a small-scale public health experiment.	Makes me track steps after pickup because apparently your friends’ BMI can forecast yours—wild stat, even wilder when you remember they did this with Framingham data, not Fitbit toys.
7The Strength of Weak Ties	Mark S. Granovetter	1973	Sociology / Networks	The reason your cousin’s roommate’s coworker gets you a job lead while your best friends just send memes; weak ties are secretly doing like 70% of the social heavy lifting.	Every time a random LinkedIn connection helps me more than my actual buddies, I mentally high-five this paper and then feel guilty for not texting my friends back.
8Computing Machinery and Intelligence	Alan M. Turing	1950	Computer science / Philosophy	The Turing Test paper — reads like someone calmly predicting half of modern tech while everyone else is still figuring out vacuum tubes.	I reread this whenever people dramatically ask if AI will take our jobs; Turing was mulling that vibe before my grandparents were born, and he does it in about 25 pages.
9The Magical Number Seven, Plus or Minus Two	George A. Miller	1956	Cognitive psychology	Classic on working memory limits; that “7 ± 2” rule explains why I forget set plays if we add more than three options (and why cluttered dashboards make my brain tap out).	I use this as my hard cap for how many KPIs belong on a single slide; anything beyond nine metrics and I assume the audience is already thinking about lunch.
10Gradient-Based Learning Applied to Document Recognition	Yann LeCun, Léon Bottou, Yoshua Bengio, Patrick Haffner	1998	Machine learning / Computer vision	The LeNet paper — early CNNs crushing handwritten digits; it’s like watching a grainy rookie-season tape of a superstar before deep learning went full superstar.	Whenever someone acts like deep learning started with their favorite GitHub repo from 2015, I quietly send them this PDF and a friendly “box score from ‘98” analogy.

Conversation

All 1 posts

Join the conversation