In the answers I've read so far, nobody mentioned the income/territories ratio of the bonuses.

I think maps in which all the bonuses with the same number of territories have the same ratio (maps that follow the rule i=t-1 or i=t-2) have a MUCH LOWER STRATEGIC VALUE than those in which there are bonuses with better and worse ratios.

This is why I think ME and China for 1v1 and Europe for 3v3 are probably the best maps I've tried so far, while other maps such as RoR are very poor under this point of view

Edited 6/22/2015 13:04:12

^ Seph, I'm wondering if you could give me a few maps that do this well and a few maps that do this badly.

I haven't read the topic thoroughly but saw a comment where Shyb talked about the amount of territories will increase strategic value. Consider fog luck then. You have 0 information where your opponent is, which would be likely on a large map. Now you have a couple dozen choices to make, but only a handful could result in a win in which case the others allow your opponent to reach your income first. A good map for a strategic game is big enough that information in pick phase is crucial, yet small enough that opponents reach each other rather quickly.

Seems I need to clarify my statements.

- Chess is a game where, if you have enough time and computing power, you can work out the perfect game. That requires no skill, just time and computing power. A very simple example is tic-tac-toe. A "skilled" player will beat an "unskilled" player every time. But anybody can work out all the possibilities for the perfect game, it just takes the effort to do so, which leaves the game uninteresting. The same goes for chess: If you have enough time and computing power, it is ultimately an uninteresting game (and there is no strategy involved). Same goes for 0% SR games (ignoring how picks affect things). Try 0% SR on Silly Isles, for example.

- By the definition starting this thread, a strategic game is one where the more skilled player has a greater probability to win the game, in proportion to the skill difference. That holds true whatever the luck percentage applied to the game, the difference will only affect the variance around the mean probability, not the "strategicness" of the game. So for 0%SR you have practically no variance, i.e. the one with the greater skill (or more computing power available) will always win (again ignoring the effect of picks which possibly introduce some randomness). With 100% luck WR, the variance is much larger, there are big chances for an upset by luck, but the the probability of the more skilled player winning can still be proportionate to the difference in skill.

Ultimately, the chess-like games end up being the field of people who exclusively do nothing but play chess and it becomes uninteresting for those who do not want to invest the time to play outside their own playing level. While a game like backgammon can remain interesting for very diverse populations. Backgammon is still a strategic game, because the more skilled player will win proportionately more games, but the underdog still feels he has a chance.

@tobe:

In human competitions in the chess world, computer AI are banned (with very bad consequences if used), and both players are given a set amount of time to calculate during the game. Therefore, both players have equal amount of time and no advantage in computing power, refuting your argument [edit] that chess is not a game of skill [/edit].

Edited 6/23/2015 21:08:45

@Fallen Angel

You could memorize Chess, and all possible variations. Thus you could bring the computer analysis with you to the game. No actual skill in chess required. Oh, sure you would need a superhuman brain to do so but the point is still valid.

FIY: The best chess players knows the first 20-30 moves in most openings, that is 40-60 half moves. They have also memorized several common endgames and the middle game. When it comes to cheating, banning computers are easier said then done. There have been incidents in the past year were a player was strongly suspected of cheating, but no evidence was found. This was at top level chess.

You could memorize Chess, and all possible variations. Thus you could bring the computer analysis with you to the game. No actual skill in chess required. Oh, sure you would need a superhuman brain to do so but the point is still valid.

How is a logical impossibility at all relevant to a discussion of skill? I defy you to study the world championship matches and top-level tournaments. Tell me, if no skill is required, then why is it that Magnus Carlsen slaughtered Anand two WC cycles in a row?

Edited 6/23/2015 21:52:42

in chess ONLY skill is required, and the skill in chess IS being able to foresee X turns ahead in a given amount of time without a machine doing it for you. Also a wide set of valid openings has been studied because there would be just too many possibilities to take into account in the first turns

what tobe and tbest say just doesn't make sense, I would like to hear what do they define as "skill"

and the skill in chess IS being able to foresee X turns ahead in a given amount of time without a machine doing it for you

This is another misconception, but a less grievous one. :)

For the record, you can get by in chess without having to look more than a move or two ahead, in most cases. Obviously in some complicated lines you'll need to be able to plan things out move by move (ex. piece a takes b, leaving c vulnerable next move etc), but you can 'see' long-term just by describing positions in your head and deciding how a move will change the position; most games, you don't need to calculate 15 moves deep to determine that a move is good.

that's what i meant, of course it is not needed to think more than a couple turns ahead in most cases (it also derives from experience); but in some cases you need to think many more turns than that to figure out which move gives the best long-term outcome: so the more you're able to compute, the better chances you have to find the best move.

This is exactly how chess AIs work, and that's why they have become so strong together with processors computing power

Important aspect that has not been mentioned much -- Warlight is a simultaneous move game. As such, unexploitable Nash equilibrium strategy for Warlight is generally a mixed strategy (pick option A 80% of the time, B - 15%, C - 5%). For example, when it comes to starting picks - if you know opponent's picks, you can usually come up with different picks that gives you an advantage. It is very rare that the best you can do is repeat your opponent's picks, and this is the only scenario where these picks can be qualified as the "best move". Same applies in every contact situation. Option A > Option B > Option C > Option A.

Therefore, artificially increasing luck through settings is unnecessary, there is always a significant amount of irreducible luck in Warlight games, on any settings.

totally agree with teddy - on most templates, if you get to know your opponent's picks, you can almost always find picks that counter them, no matter how bad they are against any other set of picks.



take this configuration as an example: triple picking West Africa is very bad due to the wastelands in North Africa and East Africa, but it is very strong against CA+SA.

rock/paper/scissors at picking stage cannot be avoided

So this kind of claims that warlight games are entirely about picking, and mostly about luck in guessing your opponents picks, the rest is really uninteresting ;-)

Picking is half the fight. Then winning after picks, is the next half.

Unless you know what I am responding to, just skip this post as it is a bit off-topic

what tobe and tbest say just doesn't make sense, I would like to hear what do they define as "skill"

I think what we said made sense, but maybe it was presented in a confusing way, I am going to try again.

The point was that Chess and tic-tac-toe is the same kind of game. The ones that are solvable, and a player can memorize the perfect way to play it. They can know how to DO something, while not knowing WHY it works. (In chess they know how to play an opening, but not why the other player can't just take the "free" pawn.)

An unskilled (or less skilled) player can therefore use chessbase* and find a weakness in the GM opening, and thus beating him, while the GM clearly is the most skilled chess player. This happens regularly.

Back to what skill is. Generally skill is the ability to excel at something. Therefore, simplified, it is whoever wins the game that has the most skill. In realty it is more complicated. Broken down skill relates to WHY the player made the moves he did. This can be very hard to measure, because you need the player to explain his reasoning. To illustrate here are a few examples.

1. A player sacrifices a Knight on h6. Either he is skilled a has seen a forced mate. Or he is unskilled and forgot about the pawn on g7.

2. Carlsen v. Topalov in Norway chess. Carlsen lost on time in a won position (he din't know the time was non- default.) In the game Carlsen has the most skill, but he still lost.

If you can't tell about a players skill over moves made or points scored then how do you do it? You could look on more games, more moves and investigate a long term trend/result. But that still doesn't tell you what skill is, only who has skill. You could say skill in chess is about tactics, calculations, evaluation of position like king safety, activity, pawn structure, knowledge of openings , middelgames and endgames. But at the same time skill somehow includes so much more. So I can't really find a good definition... and I have to stick to one that I know is wrong . Skill is the ability to excel at something. Or even simpler, Skill= Winning.


*Chessbase, if you are not familiar with it, is a program that contains many millions of games from many players, tourneys etc. and the tools to analyze them.


EDIT: Another point is that since chess is a solvable game, and starts in the same position evrytime any player can just copy any other players. That is anyone can play Carlsens opening, and moves while not having his skill. Or what a computer would do.

Edited 6/24/2015 19:10:11

An unskilled (or less skilled) player can therefore use chessbase* and find a weakness in the GM opening, and thus beating him, while the GM clearly is the most skilled chess player. This happens regularly.

For one thing, chess is theoretically solvable, but it won't happen anywhere near the next hundred years.

Next, beating a GM is not so easy as memorizing a single opening line. You clearly have no idea what you're talking about. gg.

Edited 6/24/2015 21:32:30

So this kind of claims that warlight games are entirely about picking, and mostly about luck in guessing your opponents picks, the rest is really uninteresting ;-)

No. This claims that you can get an immediate advantage over your opponent in picking.

@FK This is all theoretical. And yes, it is ages until chess is solved. So far chess has been solved with any 7 pieces. A climb until 32.

Oh, and the GM thing was an example taken from my life. One of my friends did that. Clearly I know what I am talking about here. GG. WP.

Edited 6/24/2015 22:47:04

Lol, so, anecdotal evidence of one person defeating a GM a single time refutes an argument. I'm done with this. XD

All games with complete information (like chess or tic-tac-toe) are solvable. All games period are solvable if you include mixed strategies. That's kind of irrelevant to the discussion though. People don't play chess by mapping out the entire gamespace in their heads, so it is a skillful game. People do do that when playing tic-tac-toe, so it is not.

If it got to the point where people were mapping out the entire gamespace of chess (or at least the parts that are likely to occur), then it would cease to be an interesting game, and people would stop playing it.