Re: Bag of Tricks: Choosing actions via the matching law

> humans the text gave was basketball players and how they decide
> whether to shoot 2-pt or 3-pt shots. People don't always shoot 2 point
> shots or always shoot 3 point shots. They mix it up. You'd think they
> mix it up for all sorts of complex reasons like range to defenders,
> that sort of thing, but empircally over a large enough sample size
> (whatever that is), it follows the matching law. Freaked a lot of
> people out when they published that

I'm still not convinced by this basketball thingy. Let's say there
were 2-pt and 20-pt shots instead of 2-pt and 3-pt. People would be
trying a lot more 20-pt shots than they used to try 3-pt ones, because
it's worth much more. And they would have the same rate of success
than before with 3-pt shots.
I'd say this example would be relevant if all shots were worth the
same, and the conclusion was that a player makes more tries from an
area where he has a higher chance of success. Because with the
distinction 2-pt / 3-pt, the evaluation of an action is not only
"success" or "not success", but the number of points acquired.
Doesn't mean the wizard decision-making is not valid. But let's say
with Spell1, the orcs' bodies are completely destroyed and the wizard
can't loot them, while with Spell2 he can - and expects to find lots
of nice diamonds on them. Then the evaluation of his action is not
only "will the spell kill my ennemy", but also "how much money can I
make from this kill". Which you could incorporate into the engine, by
evaluating the spell value with something like : ennemy_killed (=0 or
1) + alpha * money_gained. A greedy wizard would have a high alpha
while one uninterested in money would have alpha = 0. And you could
refine the system more...
Eeeerrrr... I kinda forgot what I was trying to say while writing
this. Bleh, shower time.

Olivier
PS: ah, yes. The fact is I'm not convinced about this matching law.
When someone has a solution to a problem that has proved to be the
best with the experience, he's going to apply it 100% of times. Only
if he thinks he can get a better outcome with another tried solution
(but let's say that other solution has a higher chance of failure), he
may try it. Like the 2 pt - 3 pt shots in basketball. If they were all
2-pt shots nobody would ever shoot behind the line (except in some
exceptional cases).