Reply to thread

Message

<blockquote data-quote="Thotas" data-source="post: 2133011" data-attributes="member: 18974">As completely wrong as it was of the prosecutor to make that remark, at least he did know the suspect was a game player.&nbsp; Thinking that this was meaningful is indicator that he needs a refresher on the rules of inference, true 'nuff.&nbsp; But it's not nearly as bad as the ignorance shown by an investigator of the Washington sniper a few years ago.&nbsp; All it took was a couple of Tarot cards laying around for this guy to say the crimes might have a link to Dungeons and Dragons!&nbsp; -- a connection that as we all know did not exist in the least.&nbsp; Mathematician John Allen Pallos (books not handy, that might only be an approximation of the name) pointed out the flaw in this logic in his book &quot;Innumeracy&quot;, without even knowing anything about D&amp;D, in connection to the teenage suicide rumors.&nbsp; Basicly, the expected number of teen suicides who play D&amp;D assuming that there is no connection would be = (% of teens who play D&amp;D) x (% of teens who commit suicide) x (actual number of teens total).&nbsp; We find that the actual number is less than the expectation, meaning that D&amp;D inhibits suicide, suicidal types don't play D&amp;D, and /or other factors that encourage D&amp;D playing discourage suicide and vice versa.&nbsp; I'm betting Pallos' math applies to crime as well.</blockquote>

[QUOTE="Thotas, post: 2133011, member: 18974"] As completely wrong as it was of the prosecutor to make that remark, at least he did know the suspect was a game player. Thinking that this was meaningful is indicator that he needs a refresher on the rules of inference, true 'nuff. But it's not nearly as bad as the ignorance shown by an investigator of the Washington sniper a few years ago. All it took was a couple of Tarot cards laying around for this guy to say the crimes might have a link to Dungeons and Dragons! -- a connection that as we all know did not exist in the least. Mathematician John Allen Pallos (books not handy, that might only be an approximation of the name) pointed out the flaw in this logic in his book "Innumeracy", without even knowing anything about D&D, in connection to the teenage suicide rumors. Basicly, the expected number of teen suicides who play D&D assuming that there is no connection would be = (% of teens who play D&D) x (% of teens who commit suicide) x (actual number of teens total). We find that the actual number is less than the expectation, meaning that D&D inhibits suicide, suicidal types don't play D&D, and /or other factors that encourage D&D playing discourage suicide and vice versa. I'm betting Pallos' math applies to crime as well. [/QUOTE]

Verification