I disagree with this line of reasoning, because exaggerated numbers are more likely to be make it to the front page of HN. In fact, the optimal lie (wrt the pro

I disagree with this line of reasoning, because exaggerated numbers are more likely to be make it to the front page of HN. In fact, the optimal lie (wrt the probability of it being upvoted) contains compensation claims that are just on the border of unbelievable. If such claims were non-anonymous, they'd carry much more weight.