Aaron Sun

Sharper, Not Safer: The Direction of Human Deviations from Engine Play

Version v5 (September 2026). Corrected estimator for all mixed models: the player random intercept described in the text was missing from the fitting helper in earlier versions. All 23 models were refit under a correction specification locked before the refit; 24 printed values c...

Conditional by Design: Sex Differences in Chess Middlegame Decision Style Emerge at the Expert Tier

Studies of sex differences in strategic decision-making have produced inconsistent results across domains. I suggest that part of this inconsistency comes from the question being asked: most studies ask whether differences exist rather than when they appear. I used a title-matche...

Sharper, Not Safer: The Direction of Human Deviations from Engine Play

Version v5 (September 2026). Corrected estimator for all mixed models: the player random intercept described in the text was missing from the fitting helper in earlier versions. All 23 models were refit under a correction specification locked before the refit; 24 printed values c...

Where the Benchmark Can Err: Human Deviations from Model Forecasts and Their Return

Second study in a two-domain line on human-AI complementarity, following "Sharper, Not Safer" (chess, doi:10.5281/zenodo.22267000). It asks whether human deviation from a model forecast earns a return in a domain where the model itself can err, using the one ForecastBench round w...

Where the Benchmark Can Err: Human Deviations from Model Forecasts and Their Return

Second study in a two-domain line on human-AI complementarity, following "Sharper, Not Safer" (chess, doi:10.5281/zenodo.22267168). It asks whether human deviation from a model forecast earns a return in a domain where the model itself can err, using the one ForecastBench round w...

Same Axis, Different Place: How Middlegame Style Shifts Across the Expertise Gradient

A companion study found that elite male and female titled chess players differ in multivariate middlegame style, but it could not test how style changes with skill. Open FIDE titles are rare below 2000 Elo, so the skill-by-sex interaction was underpowered. This paper isolates the...

Opening Convention, Middlegame Freedom: What Elite Stylistic Variation Implies for Developmental Chess Pedagogy

Chess instruction for beginners has to settle a recurring question: how much of play should be taught as convention to be reproduced, and how much should be left for the learner to explore. This paper argues that the answer differs by game phase. Opening play is convergent. It re...

Sharper, Not Safer: The Direction of Human Deviations from Engine Play

When a human player departs from a chess engine's preferred move, is the departure noise, or does it have a direction? I analyze 1,970 rated online games by titled players: 50,021 middlegame position-moves; 120,723 engine evaluations in total, comprising 63,429 root and context e...

Sharper, Not Safer: The Direction of Human Deviations from Engine Play

When a human player departs from a chess engine's preferred move, is the departure noise, or does it have a direction? I analyze 1,970 rated online games by titled players: 50,021 middlegame position-moves; 120,723 engine evaluations in total, comprising 63,429 root and context e...

Conditional by Design: Sex Differences in Chess Middlegame Decision Style Emerge at the Expert Tier

Studies of sex differences in strategic decision-making have produced inconsistent results across domains. I suggest that part of this inconsistency comes from the question being asked: most studies ask whether differences exist rather than when they appear. I used a title-matche...