Hi, Michael. Thanks for taking the time to do an AMA. Where do you fall on the spectrum of looking for high probability wins vs black swan hunting? To put it a

Hi, Michael. Thanks for taking the time to do an AMA. Where do you fall on the spectrum of looking for high probability wins vs black swan hunting? To put it another way, how high of a failure percentage would you accept so long as it maximized total cohort value? E.g. would you accept see it as a loss if you funded the best single startup of the decade and all other YC funded companies failed? How about a 90% failture rate, but huge winners bringing a bigger return than YC has historically had? How do you balance the desire for a good hit rate vs a good total return?