Mike Bishop 🇨🇦🇺🇸
@thatmikebishop.bsky.social
AI, forecasting, policy, social science
Loves: humility, kindness, high decouplers
Loves: humility, kindness, high decouplers
Also note that internet poker pros earned a steady income playing 12 tables simultaneously… they weren’t deeply analyzing their opponents, they steadily skim money off players that are playing far from game-theory-optimal.
November 4, 2025 at 2:19 PM
Also note that internet poker pros earned a steady income playing 12 tables simultaneously… they weren’t deeply analyzing their opponents, they steadily skim money off players that are playing far from game-theory-optimal.
Realistically, either A.I. solves this for posters or it doesn’t get solved. Almost nobody gonna put effort into it.
July 15, 2025 at 4:32 PM
Realistically, either A.I. solves this for posters or it doesn’t get solved. Almost nobody gonna put effort into it.
I greatly respect your work but I disagree.
When normal people disagree with the scientific consensus it’s mostly because they don’t care much about the topic, or worse, because they don’t trust scientists.
When normal people disagree with the scientific consensus it’s mostly because they don’t care much about the topic, or worse, because they don’t trust scientists.
July 15, 2025 at 4:29 PM
I greatly respect your work but I disagree.
When normal people disagree with the scientific consensus it’s mostly because they don’t care much about the topic, or worse, because they don’t trust scientists.
When normal people disagree with the scientific consensus it’s mostly because they don’t care much about the topic, or worse, because they don’t trust scientists.
Matt, I have to disagree with the gist of this.
The problem of academics overclaiming and overgeneralizing from their work is serious and confuses the public (and even other academics).
The problem of academics overclaiming and overgeneralizing from their work is serious and confuses the public (and even other academics).
July 15, 2025 at 3:46 PM
Matt, I have to disagree with the gist of this.
The problem of academics overclaiming and overgeneralizing from their work is serious and confuses the public (and even other academics).
The problem of academics overclaiming and overgeneralizing from their work is serious and confuses the public (and even other academics).
Preventing fraud is best but bigger penalties would help too. At a certain level it should be literally criminal.
July 14, 2025 at 2:42 PM
Preventing fraud is best but bigger penalties would help too. At a certain level it should be literally criminal.
See link to paper in this thread x.com/alexwcohen/s...
Alex Cohen on X: "Do effect sizes/standard errors from econ studies predict future results? This paper finds they often don’t (CIs don’t contain future results) unless we allow for real diffs across studies (ie, low external validity). Requires assumptions but cool mix of empirical approaches. https://t.co/JURGQLCOty" / X
Do effect sizes/standard errors from econ studies predict future results? This paper finds they often don’t (CIs don’t contain future results) unless we allow for real diffs across studies (ie, low external validity). Requires assumptions but cool mix of empirical approaches. https://t.co/JURGQLCOty
x.com
July 7, 2025 at 9:59 PM
See link to paper in this thread x.com/alexwcohen/s...
Even social scientists that "get it" mostly abide by a norm against forecasting.
Including subjective probabilistic forecasts in serious scholarship might receive pushback from reviewers or editors even though it would help readers interpret research implications.
Including subjective probabilistic forecasts in serious scholarship might receive pushback from reviewers or editors even though it would help readers interpret research implications.
July 7, 2025 at 9:59 PM
Even social scientists that "get it" mostly abide by a norm against forecasting.
Including subjective probabilistic forecasts in serious scholarship might receive pushback from reviewers or editors even though it would help readers interpret research implications.
Including subjective probabilistic forecasts in serious scholarship might receive pushback from reviewers or editors even though it would help readers interpret research implications.
I wonder what you mean by “we” and “allowed” … there was a resistance, both obvious and invisible. And those resisting were themselves diverse and sometimes in sharp disagreement about strategy.
July 6, 2025 at 2:06 AM
I wonder what you mean by “we” and “allowed” … there was a resistance, both obvious and invisible. And those resisting were themselves diverse and sometimes in sharp disagreement about strategy.