The Axiom System - Part 3: How Do We Make Decisions

@Craze said in #3:
> Thus far, the framework reminds me of step two of my own '3-Step Process For Game Analysis' - which states that we should figure out what moves we had to see or what we had to appreciate in the position (evaluation) to find the best move.
>
> I had just been thinking of writing a 'First Principles' book on chess. I've found that the engine's 'logic' behind moves often deviates from more 'conventional' principles from the pre-computer era (as Willy Hendriks already started to indicate in 'Move First, Think Later'). It will be interesting to see how the first principles approach evolves from here.

Thanks for reading Max! In a later post I'll discuss how to use these ideas to review games/categorize one's mistakes - which sounds like it will have some overlap with your 3-Step process.

I'd encourage you to write that book :) I'd definitely enjoy reading some chess content along those lines.

@wpee said in #4:
> Interesting articles and I do follow that practical chess is all about decision making, but following questions came to mind when reading :

Firstly thank you for your comment and questions! Let me try to address them.

> * how do you evaluate positions if you have no framework of concepts to judge a move by? Even values of the pieces is a concept in chess and no universal law (given value of pieces can change given position at hand)

That's a great question and one that I'll deal with mostly in the next 2 parts. The next part will focus on the idea of using common chess principles to justify our evaluations and why I think that idea is flawed. Then, part 5 will focus on given these flaws, how can/do we evaluate positions. I'd try to give a summary but perhaps it is better if I explain it more fully in the subsequent posts.

> * Do we not use chess concepts to link known and unknown positions. e.g. famous mating patterns seems to fall outside the known positions, since they are no theoretical endgames. If you know smothered mate pattern, you can try to apply it in unknown positions. How would this fit in your current framework.

I think what you're getting at here is the idea that certain concepts that we can find through this decision making process also happen to have specific names (like smothered mate) in modern chess. In that case, I would say that that's not much of an issue for the framework. It's fine for certain things to have 'modern chess names', although what I would argue is that those names and explicit knowledge of those concepts is not necessary to find them/play good chess. When it comes to being able to find a smothered checkmate, it doesn't matter whether someone knows the name or has explicitly been taught it by a book or anything. Instead - at least in this specific case - I would say it's about their previous exposure to that tactical pattern. You can check out my youtube vid via my profile for more info on this particular case (acquiring/learning tactical patterns).

> * There are many interesting books existing about decision making at the board. Two authors that came immediately to mind are Dan Heisman (e.g. in his Best of Novice nooks book from Everyman he talks about decision making process and it being analyzing and evaluating) Jacob Aagaard considers seeing the non-obvious move as his definition of calculation in chess. (e.g. Excelling at chess calculation)
> So I feel that your axioms are maybe a new way of bringing things, but they seem to merely support at the moment the importance of concepts and chess culture, since they seem to presuppose concepts and bring existing concepts in a new manner, but the underlying idea seem the same to me.
> Anyway nice articles, and looking forward to the rest

Thank you!

@thefrickouttaherelol said in #8:
> You are severeeeeeely abusing the word "axiom" here.

I'm aware of the fact that 'axiom' is typically used in more abstract contexts such as in maths and logic, and has slightly different connotations than 'assumptions' or 'premise' for those 'in the know' so to speak - but The Premise System didn't sound as catchy :)

This seems like a very minor point that has no bearing on the main content of the article. The idea is that those statements are taken to be self-evident truths that will serve as the foundations for subsequent arguments. If it bothers you that much, then please just replace 'axiom' with 'premise' throughout the post ;)

@chessTX9 said in #9:
>

Yes you're right they are very similar. I think it was important to have some axiom that, in some way, established the equivalence of better moves = better decision making. To most people this is of course obvious, but it's important to explicitly do this if we're being thorough.

Regarding your last question, yes they are basically the same thing. The caveat would be that decisions (in the sense used throughout the post) = moves only in unknown positions. In known positions, since the move is pre-decided, it's not quite right to say that moves = decisions (in the sense of decisions = seeing + evaluating).

While it will take me a long time to get through all you wrote. From what I gather I think it is good that you are doing this, as at least one effort to bring in the reader's reasoning abilities, and the structuring of reasoning in communication about chess improvement or understanding (my slant is that improvement could exist without understanding, but that is not important).

> large number of positions.

That caught my eye. (well the things before). And it seems to develop further. I do like axiom one. As it does lead to the notoin of generalization from previous experience set to what is the possibly wilderness of chess positoins one can be expecting in the remaining of their lifetime (since we are talking about "practical" chess understood as what is for one person interest to read your treatise). I can work with that set of caveat about chess. It is making explicit a lot of previous blogs, taking that for granted.

I also approve of your non-pedantic need to clarify the meta level of any of your statements. This is something I have always appreciated in the presentation of mathematics where reasoning was expected of the audience, not imitation learning from examples only. So it means you respect this ability in potential chess player/improver/learner/me. whatever their level (or skillset spectrum).

I say: do not get distracted too much by impatience in responses. Although, it seems you are able to adress those. I am sorry It will take me a long time to be able to respond from the content point of view. But I will keep an eye, and keep an index of the things you put here. I am not saying I will agree or already agree with all the things.. But I agree with trying.. And find questions more important than answers, at least making them explicit. Answers can be better questions.

How large is that set of positoins. Should we count it? or measure another type of size?
Edit: There is no kidding in this post. I just realized given how big lichess might, I might have no clue about its wilderness of point of views (I often realize that, whenever I think I am making a joke or not). And not here.
The joke would be on those that think that chess is a done deal. And its teaching as well.

I don't think there's any abuse with regard to the term axiom. In the first place, these are unprovable things which we can then use to build more things upon.

Now whether these axioms are useful or not, we'll just see in the next parts

As long as these [unprovable starting point assumptions] /axioms don't contradict each other, and end up being useful in the end, I'm all for it.

@DailyInsanity said in #13:
> This seems like a very minor point that has no bearing on the main content of the article. The idea is that those statements are taken to be self-evident truths that will serve as the foundations for subsequent arguments. If it bothers you that much, then please just replace 'axiom' with 'premise' throughout the post ;)

I'd say it's a pretty major point and especially the discrete logic diagram you use for the banner images makes the initial catch misleading - for example, I clicked on the article thinking you had actually some sort of formal, rigorous system for evaluating chess moves.

Great articles, I would coelesce this approach with the rule of chess themselves.
We talk about vision, tactics, strategy, etc. but they all emerge from these, hence the blurred distinction and confusion.

ie. control the center? why? because less squares are accessible on the side (fundamental rule) -> less movements / more cramped = less potential moves + because of more synergy with checks (fundamental rule) -> less available tactics = less potential moves, considering chess is a __run__ to checkmate (the first to checkmate wins)

this could be reverted to,
1. to win the game, checkmate matters most
2. the piece synergy to attain that before the opponent does, considering the system is dynamic, is thus what matter most
3. maximizing piece movement and synergy is the way to attain checkmate (we could elaborate on why and how, then develop on each individual tactics/piece synergies etc.)
4. controlling the center do that, developing each pieces do that, etc. but alternatives also. (ie early game tactics, traps, etc.)

this way of thinking is better because it explains both the "standard" way to play and the exception.
it provides both a tactical and strategic framework, and most importantly bridges the distinction between both.

Personally, I think modern chess concepts (words) like "initiative, space, development, material" have real value in the see and evaluate skills. Often when I am weighing one move over another, I will consider what these moves do for me, and how important that specific concept currently is in the position. Also, when identifying candidates, I might ask questions like "What can I do to keep the initiative?" or "What are good candidates for continuing my development?" or even better: "Do I have any moves that accomplish or initiate several ideas at once."

For example when playing the white side of a Yugoslav Attack in the Dragon, often material is less important than the other three. So in my move choices I will be less inclined to preserve my h-pawn if losing it advances one or more of the other three significantly. Mind you, concrete variations are still important, but when I am making a "good decision" I tend to allow these modern chess concepts to help direct where I will spend the most time evaluating.

Still, I like where you are going with this. Your main point is not that modern chess concepts are wrong per se, but that they may not be a good basis for improvement. Learning them, while useful, may have poor returns on improvement, if you haven't refined your mental processes for move selection. I agree! But they are still ultimately useful, IMO.

Oh...

You wrote:

"Jack, duh! Everybody already knows that, and you're just making things sound more complex than necessary."

But you don't understand that it is possible to draw the following conclusion from what has been said?

Everybody already knows that. So it is pointless to write it. It is even more pointless to read it. It is just nonsense to use your time that way, if you like to improve your chess.

And on the top you have the attitude of everything else we ever learned about chess is wrong and useless.

And of course this is my opinion and maybe Max Euwe was iust following a similar idea in "Judgement and Planning in Chess" (my favourite chess book), but, and that is the difference to your idea, Max Euwe said it is important to know and understand the characteristics of a position.

So maybe there is an apple tree with red apples. You learned before about "apple tree with red apples" and know that it is time to harvest the apples. So judgement: "apple tree with red apples", planning: "how can I harvest that apples?".

Do you think it is better to learn to judge if a knight is strong or weak against a bishop in a position, or to learn what a known or unknown position is? What makes sense? What is nonsense?

I have a opnion on that. And that's it.

I don't want to be rude. But I want to share my opnion. You can call this upfront, frank, rude, cheeky, whatever. I just want to be 100% honest, there is zero intentions inside of me to hurt you or whatever word fits better. But I am clumsy, I don't know how to do better. And I am sorry for that.