No. You're thinking of the study described here: https://www.lesswrong.com/posts/LETXrTsGAqJrcik3T/fnord https://www.lesswrong.com/posts/LETXrTsGAqJrcik3T/fnord

No. You're thinking of the study described here: https://www.lesswrong.com/posts/LETXrTsGAqJrcik3T/fnord https://www.lesswrong.com/posts/LETXrTsGAqJrcik3T/fnord and those 90% aren't what you think. You might say the key quote is this: «Of course this example is me just making up numbers to show how counter-intuitive results can be from this kind of model.» If you say 50% instead of 90% then you describe the study fairly (and a little incompletely).