The hottest Substack posts of Mike’s Blog

And their main takeaways
255 implied HN points 15 Apr 23
  1. Sometimes policies aimed at solving a problem may backfire and create more issues.
  2. Predicting outcomes of complex systems can be challenging even with set rules in place.
  3. Unintended consequences can arise from misaligned policies, impacting those they were meant to protect.
39 implied HN points 07 Nov 23
  1. Calibration plots show how accurate forecasts are by comparing predicted probabilities to actual outcomes.
  2. Sports betting markets are remarkably well-calibrated based on analysis of MLB, NFL, NBA, NHL, and NCAA data from 2016-2020.
  3. Implied probabilities in sports betting are normalized for analysis, where they sum to 1, to compare prediction accuracy.
216 implied HN points 12 Nov 22
  1. The post evaluates and compares the performance of forecasting sites like PredictIt, FiveThirtyEight, and Manifold Markets for the November 2022 midterm elections.
  2. Data collection was done manually by checking the sites before the election date, and the analysis includes calibration plots and Brier scores to assess forecast accuracy.
  3. Results show the importance of not solely relying on calibration plots, as Brier scores may provide a more accurate measure of forecast performance.
Get a weekly roundup of the best Substack posts, by hacker news affinity:
137 implied HN points 26 Dec 22
  1. Sports betting markets can provide insights into prediction market efficiency
  2. Efficient Market Hypothesis suggests asset prices reflect available information
  3. Data analysis comparing sports betting odds and forecasts from FiveThirtyEight showed similar performance
78 implied HN points 07 Apr 23
  1. Betting markets slightly outperformed FiveThirtyEight in predicting NBA, NFL, and MLB games.
  2. New data collected for March Madness shows both FiveThirtyEight and betting markets performed similarly, and neither significantly outperformed.
  3. Hypothesis: Both betting markets and experts may have worse accuracy in playoffs and tournaments compared to regular season games.
58 implied HN points 12 Feb 23
  1. There are 2 types of forecast accuracy: Correct Prediction Accuracy and Uncertainty Judgment Accuracy.
  2. Brier scores are a common metric to evaluate forecasts, balancing Correct Prediction Accuracy and Uncertainty Judgment Accuracy.
  3. Brier scores can show how a partially-psychic forecaster with high correct predictions may still have a worse score than a forecaster with good judgment.
2 HN points 06 Feb 23
  1. The correlation between land use regulations and homelessness is interesting.
  2. There might be a potential link between NIMBY regulations, lower housing supply, expensive housing, and homelessness.
  3. Further investigation is needed to determine if overregulation of housing markets causes homelessness.