[MUD-Dev] [DGN] The psychology of random numbers
rayzam at travellingbard.com
Tue Jan 27 21:40:42 New Zealand Daylight Time 2004
From: "J C Lawrence" <claw at kanga.nu>
> The PR problem is the killer, not the distribution accuracy, and
> that's where the engineering problem definition comes in. In
> general we're not engineering for statistically accurate
> probability models, we're engineering for something that is
> perceived well, and in general that means a system which appears
> unpredictable to the untrained eye for almost all small sample
> sizes, and which, if anything, "fails" to the player advantage
> more than it should. "Appearing random" and "being truly random"
> are different problems with different solutions.
Yep. As an exercise, ask the person next to you to write out 100
digits in a random order. Compare to 100 digit numbers created
randomly [with the function of your choice]. You'll see that the
expected value of number of runs and length of runs generally falls
outside the 95% confidence interval for it occurring by
chance. People's expectation of random, as self-generated, is very
different than really random.
> I haven't known a player yet who would complain about being lucky.
That's the real rub. And part of that is how our memory works. It's
much easier to remember incongruent and negative situations. Part of
it is survivability. Rats very quickly learn how to avoid the parts
of a maze with electric shocks. It takes them longer to learn the
way to the cheese.
Players remember unlucky streaks more, as each builds on the last,
making it more and more negative.
So we have a luck factor, that is described in words, from You are
having a day best forgotten, to You are having a great day. And what
players would claim is that they fail more spells when they were
having a great day, than a day best forgotten. I pulled stats on it,
and it wasn't true, there actually was a slight difference in spell
success rate that coincided with more luck giving more success.
One of the reasons for the player belief is because when the same
negative streak occurs [say failing a well-trained spell 3 times in
a row] during a great day, there's no external cause to blame and it
has a greater negative emotional impact than when it occurs during a
day best forgotten, which provides an external cause. This probably
relates to the same attribution mechanisms that were manipulated by
psychologists [give everyone a pill that ups their heart rate a
bit. Tell half of them what it does, tell the other half that it
does something innocuous. Then have each of them judge an emotional
situation. Those that knew what the pill did, said the situation
didn't really affect them. Those that don't know the pill upped
their heart rate, misjudge the situation as having an effect on
Being lucky is internalized. *I* am lucky. Bad streaks is
externalized. The *game* isn't fair.
Fundamental attribution error.
MUD-Dev mailing list
MUD-Dev at kanga.nu
More information about the MUD-Dev