Monday, July 20, 2009

Non Decisions: Calculation and Guessing

Game play is either calculation or guessing. Neither of which are decisions.

Calculation is not really a decision: you calculate correctly and succeed, or you calculate incorrectly and fail, or you don't to calculate - or recognize when you can't, because the information is not available to you - and you guess.

Guessing is also not a decision: you guess with the odds, or against them which is foolish.

Calculation is made difficult by time constraints, limited mental capacity, and false beliefs. Time constraints may be explicitly enforced, such as in Chess or Scrabble, or implicitly enforced by social necessity: your friends will leave if you don't make a move already. Mental capacity limits how much information you can juggle in your head at any one time. False beliefs are those which skew your ability to calculate or guess properly, such as believing in false patterns or that something will happen because you want it to, very badly.

Calculating odds is not guessing. Calculation is the opposite of guessing. Anything you can't calculate you guess, and since you have no information, no guess is better than any other. Deciding whether something really is a guess or can be calculated, is also a calculation. More specifically, in deciding when a pattern is a pattern, and when it's just random noise.

Determining what your opponent will do is often guessing. Game Theory has a lot to say about it, often on the assumption that your opponent is rational and intelligent. But, in essence, your opponent's moves are just as much odds calculations and guesswork as any roll of multiple dice with constraints can be. Which means that even classic abstracts with perfect information - other than your opponent's future moves - are luck, on a certain level. Your opponent may not find the right move, or he might, by accident, when he guesses a play that is beyond his calculation abilities. Over the course of a long game, the odds of his repeatedly finding the right play by accident become slim to non-existent.

In some games, theme, role play, and humor also play a role. Setting this aside, where are the decisions in a game, if they not in the calculation and not in the guessing? Is decision-making an illusion?



Aaron said...

This seems like a similar discussion point to something I recently blogged about regarding choice in games, in response to a blog by James Portnow making a delineation between choices and problems:

Dug said...

Actually, calculation and guessing are part of the decision-making process. Whenever you are faced with a choice (do A, B, or C) you make a decision. Sometimes the tree of possible outcomes is very complex (like in many board wargames), sometimes it's moderately complex (place a tile in a specific place on your board in Galaxy Trucker), and sometimes it's pretty simple (buy a property or not in Monopoly).

Thus, calculation and guesswork are actually two poles on a decision-making axis. Do you have information, or not? Usually it's somewhere in the middle. You may be able to make a calculation that helps you make the decision, but if that calculation is only one small part of the information you need to make a good decision, then it may actually hurt the decision making process if it becomes your main rationale.

Thus, calculation may be important, but the *really* important skill in gaming is actually estimation - understanding what you know, what you don't, how important the various elements are, and coming to a decision as a result. The opposite of estimation is intuition (in it's purest form), which many people confuse with estimation because it actually relies on information they have but aren't thinking actively about.

So no, calculation and guesswork aren't decisions because then they would be redundant, as they are part *of* decisions.

That's my 'geek wisdom for the day.

David Klein said...

Though you bring up interesting points, the title is really just a question of semantics. The calculation or guessing is not a decision. Making a move based on that calculation or guess is. See:

Yehuda said...

Aaron: looks interesting, I'll check it out.

Dug, David: I think you're right. In my own words, my article is highly theoretical; in practice, you have to decide.

And the decision is usually right at that point between calculation and guessing: evaluating whether the information gained by calculation is right, and whether the guesswork is truly random or not.


Poet said...

Your thesis can be easily applied to life, and after that you may as well proclaim humans have no free will :)

thk123 said...

I think assuming that your game isn't entirely luck based you are basically just calculating. However, because most games feature either a too greater scope (Civilization) or to complex a system (Dawn of War) you must guestimate, which is effectively making a decision. You use past information and a rough understanding of the system to try and predict what the effects will be.

It is also made more interesting by the person you are playing. As you do not know what they are going to do, most games won't have a calculable correct move (some do, for example, Tic tac toe, there is a winning strategy and in chess too, we just haven't worked it out yet)

Mikey said...

I think the real issue is what you consider a decision. How do you define the term. Assuming a fairly colquial use of the term I have to agree with the comments above. Along those lines, high level magic players are fond of saying there is no such thing as a good play, only the right one. (I believe this is atributed to Jon Finkel) Mathematically there is only one of the myriad of options in a game that is most likely to win you the game, but that doesn't mean it is easy to figure out or that you might not choose it. Say your goal is not to win but rather to make sure someone else in the game has fun. what was the "right" move to make before is no longer the right move.

On a side note your post reminded me of the chinese room argument made against functionalism by John Searle. Was it at all inspired by that?

Chris said...

I'm going to side with Dug here - calculation and guesswork are part of the decision process. :)

(And, as an aside, all of these processes take place in the orbito-frontal cortex, the "decision centre" of the brain).

For me, a third process is important: intuition. You could choose to roll this into "guesswork", I suppose. But whereas guesswork only happens in the context of incomplete information, intuition can be applied in any situation - as a quicker alternative to accurate calculation, for instance.

I often use my intuitive judgement in preference to calculation (even when the information is total and calculation is attainable) because I find it more fun to intuit than to calculate.

Can this be separated from guesswork? Perhaps it depends on how one chooses to define a 'guess'. :)

Best wishes!