If the question was posed "Who is the best WarLight player?" And assume I have the two best players before me. How would we decide in one game?

I Imagine these maps might work:
https://www.warlight.net/Map/1173-Imperium-Romanum
https://www.warlight.net/Map/5745-Battle-Islands-V (for some reason hyperlink doesn't work)
https://www.warlight.net/Map/4179-Macro-Land (possibily the best one made for small 1v1 games imo)

Though it's important to note that you might need to adjust some bonuses etc. Also if the players start next to each other (assuming manual dist. custom will never work, and radom isn't skill) it'll be many unfun turns of building a huge stack before the game can progress. (a situation similar to this game : https://www.warlight.net/MultiPlayer?GameID=2400706 After building for many turns (talking easily more than 100) I'd eventually be able conquer the rest of the map while keeping my opponent blocked), not very fun so to speak.
I'd heavily suggest that you make it 2 starting spots, as even if you were able to create a good template I'd doubt it would be fun (especially in the long run).

"If the question was posed "Who is the best WarLight player?" And assume I have the two best players before me. How would we decide in one game?"

Well as WL have infinite number of templates, deciding it on one game wouldn't make much sense (not to mention that players are strong on different settings:p) Anyway that's not the point, I'd think a small map (like macro Land), no-luck settings and cards (reinforcment, sanctions (both positive and negative would work, blockade/abandon/gift etc ) don't have to be many, but need to be a significant reward for capturing territories)

Edited 1/22/2015 20:51:17

@Mercer that would be a stupid question, just slap anyone who made it or blacklist him

Edited 1/22/2015 20:44:25

fairest map for 1v1 where there is only two starting spots (one for each side) and neither spot has an advantage over the other:
https://www.warlight.net/MultiPlayer?GameID=5992602

How would we decide in one game?

You cannot, assessing players skill based on 1 game is frankly impossible. At the very least there would remain a great deal of uncertainty. To a trained eye it might be more apparent who is the better player, but not based on the result, but rather moves and decisions.

Just to be clear, even the new "no luck" template has luck involvement. Not in terms of calculation, but the result. For example if we compare strategies taken by a both players, they all think forward and anticipate multiple scenarios. Based on all of them, we could judge that, for example, one player would win 60% of the time long-run, but as it happened he lost in this game.

If you really want a single game with the least amount of luck that gets in the way, Europe 3v3 is very close to being perfect.

@szew

What about a no-fog template; it would take away a lot of the uncertainty.. (though I'd say a heavy -fog would be a better measure of skill-requiring advanced guessing regarding your opponents status). You need to think outside the box for something like this. (that being said, you're ofc correct, except for the Europe 3v3 part :p)

No-fog template would remove a lot of guesswork early game (direction of attack/expansion), still guesswork when trying to break opponents bonuses is required. Some judgment calls are tricky and can backfire, double-bordering your opponent with an inferior stack is a good example, you'd be able to asess if you can force last order etc., but still some luck will remain. Europe 3v3 part I'm certainly not wrong about, I've played competetively for a long while and I can safely say it's the most skill based template so far. Also there is a general consensus among best players that team games in general with good settings are more skill-based than 1v1's. Still, your proposition of no-luck no-fog template variation of strat 1v1 seems to be a decent contender when it comes to skill-based 1v1.

If the starting spots are predetermined and known to both sides, then the perfect strategy will be easy to find out by trial and error within a few games. If the perfect strategy is easy to determine, then all the games of that template will be very similar to each other and quite predictable, regardless of how many starting points each side has.

And if the game is simple, then the importance of skill decreases whereas the importance of luck increases.

If, on the other hand, the starting positions will be picked out of a large pool of possibilities, each side having at least 2 picks, then a perfect strategy is very difficult to determine. In such cases, overall skill and understanding of the game play a big role, whereas luck plays a smaller role.

Nevertheless, if the opponent's picks are not known beforehand, there will be a considerable amount of luck at play as to how one's picks do against the opponent's picks. Therefore, to determine the "best player of WarLight", there would need to be a best of 9+ games challenge. Even chess world championships are played in the best of 12 format, despite chess being much less luck-based than the so-called "no-luck money games" of WarLight. There is too much luck involved for one game to be enough to decide.

I would like to make sure Coin Games are as much skill as possible. So the question is broad. Small advantages not related to skill should be eliminated as much as possible.

We are working on a few updates that should make coin games more strategic and less about first pick advantage.

Any ideas are welcome.

I don't know how you can get around the first move on the first turn issue. If both starting positions are equally good then what matters is who gets first move, which might be an advantage or a disadvantage.

With a custom senario there is no first pick luck at all.
Making a ballance custom senario on symmetrical map is easy, only like Odin dit say, the perfect strategy will be easy to find out.

I think this game has custom senario where the perfect strategy will not be easy to find out:
https://www.warlight.net/MultiPlayer?GameID=2681068

Here Slot B can get a income advantage at the start and has a better position in africa, but Slot A has Greenland as safe bonus. If this game would be with no-luck, i think it would be even more ballanced.

You cannot, assessing players skill based on 1 game is frankly impossible. At the very least there would remain a great deal of uncertainty. To a trained eye it might be more apparent who is the better player, but not based on the result, but rather moves and decisions.

Just to be clear, even the new "no luck" template has luck involvement. Not in terms of calculation, but the result. For example if we compare strategies taken by a both players, they all think forward and anticipate multiple scenarios. Based on all of them, we could judge that, for example, one player would win 60% of the time long-run, but as it happened he lost in this game.

If you really want a single game with the least amount of luck that gets in the way, Europe 3v3 is very close to being perfect.

Someone is getting a little too technical about word choice. I think what Mercer is trying to figure out is what template, what map, etc. to use in order to have as competitive as a game as possible for 1v1.

For 1v1 there's all types of templates that test different points of skill, but just all-around skill would probably be the standard ladder template 1v1 strat me 0% WR. The reason is not because it is a good test of skill, but because it is so well-known and widely played that it is a good competitive template.

Other than Strat ME you could make cases for Greece ( http://warlight.net/MultiPlayer?TemplateID=388092 ), Guiroma ( http://warlight.net/MultiPlayer?TemplateID=435109 ), and Battle Islands V ( http://warlight.net/MultiPlayer?TemplateID=526636 ).

The templates that test more specific abilities would be like Poland ( http://warlight.net/MultiPlayer?TemplateID=526624 )(calculation), Turkey ( http://warlight.net/MultiPlayer?TemplateID=248974 ) (brawl as I like to think of it that way), and templates that use multiattack ( https://www.warlight.net/MultiPlayer?TemplateID=496851 and Ruthless Bastard's Blitzkreig tournaments), local deployment ( http://warlight.net/MultiPlayer?TemplateID=475503 and http://warlight.net/MultiPlayer?TemplateID=475488 are good examples), and no-split (don't have any saved).

Unfortunately, Mercer, there is no easy answer to your question. If we had the 2 best players of Warlight and wanted to test their skills we could argue endlessly on what template would be the best to do that. That's why all of our player-created leagues, tournaments, and clan wars tend to have multiple templates involved. That's why WSOW runs many tournaments. That's why in the WGL we tried to do different templates, and tried out many different things.

If I had the 2 best players of Warlight, and I wanted to determine who of those 2 was the best at 1v1... I would have them play a Strat ME game, and then allow the winner of that game to pick from a list of various templates which one he wanted to play 3 times. Then allow the loser of that game to pick from a list of various templates which one he wanted to play 2 times, and I'd alternate those 2 templates in a best of 5.

Even then the community wouldn't be satisfied to call that player the best. 2v2 and 3v3 make you utilize other skills not found in 1v1. So yeah, Warlight isn't a straightforward game like Chess, League of Legends, and the other games you're trying to emulate in terms of standardizing a competitive template.

The reason why we've all been trying to get Warlight to support a 3v3 Europe ladder is because that template is hands down the best 3v3 template we can think of. Unlike 1v1 and 2v2, there's not a huge argument going on about what template is the best way to test your skill in 3v3. So if you're trying to find a standard competitive template with little argument on test of skills.. well here you are:

https://www.warlight.net/MultiPlayer?TemplateID=583714

If you keep asking about 1v1 you'll constantly get different answers, but with 3v3 there really isn't a better 3v3 template.

Edited 1/23/2015 04:56:19

This one.

https://www.warlight.net/MultiPlayer?TemplateID=303427

You're welcome.

Tic Tac Toe: Reverse Income!

https://www.warlight.net/MultiPlayer?TemplateID=347566

no fog make less interesting the game, and require less skill: you don't have to calculate where your opponent is, his income, and when there are different chance of expansion, you have to calculate the best and find a way to win if opponent pick it (you will win anyway if the opponent picked the best expansion possible); remove fog will make the games decided by picks mostly.

Make 2 start, one for player, kill one of the hardest part of the game: make picks. usually, in medium size map, 3 or 4 picks are the best choice for make competitive games.

@lolowut about the 3v3 Europe, lately we used the 4x4 0% SR as random leftovers could influence game in short term. I also think no luck cycle moves are good too.

We are working on a few updates that should make coin games more strategic and less about first pick advantage.

I am answering about Ror and Medium Eart coin games as i didn't played other games, but i think in both of them, 1st pick isn't important: i often decide to wait till last second to commit bc i'd rather have 2nd and 3rd pick then 1st. As luck is 0%, and 3v2 always work, 1st pick is less important than 1v1 ladder (example: for make a +4 bonus of 5 territory, in 0% WR is better have the raondom warlord in the centre of the bonus, so you can utilise both leftovers in the next turn and increase your chance to make the bonus; in 0% SR, you can make the same bonus 100% without utlising one of the leftovers of the 1st turn).

Edited 1/23/2015 13:36:50

coin maps don't even have *transfer only/attack only/percentage attack* checked as on..
and only strat earth has order priority and order delay cards...
these are subtle skill features... features that are wasted on low skill people and can be deadly in the hands of high skilled people..

honestly, I find that new no-luck features tend to take alot of the skill out of the game...
yeah.. it's more straight to the point, but good players could compensate and calculate for the slight luck involved, instead of using a standard attack everytime as we do now.

What youre looking for is a game like rock/paper/scissors where your actions dictate the outcome. That doesnt mean the game requires much skill.

General rule when you take options out you are lowering the skill level.

Taking out the fog lowers the skill level
Taking out cards lowers the skill level
Taking the option to attack with 3v2 or 4v2 out lowers the skill level
Making the map smaller lowers the skill level, that is part of the reason why team games require more skill. The other reason is 3 teams with 3 starting points requires more skill than 1 team with 9 starting points.
Taking out attack/transfer only and attack by % lowers the skill level.
Taking out cyclic lowers the skill level, because with cyclic you have to plan for the turn order and use your cards appropriately.

So thats a starting point to creating a template that requires the most skill.

If you want to create a template that requires the least luck. Id go with custom setting where you start next to your opponent in each bonus. Now the game is 100% reliant on predicting your opponent, it becomes a game of rock/paper/scissors.

Edited 1/25/2015 07:34:03

General rule when you take options out you are lowering the skill level.

How eloquently put! As you reduce the variables you actually reduce the amount of experience required to calculate possible outcomes. In reality we need to balance versatility by how enjoyable it is to play, which is what WarLight achieves so well. We have to deal with so many variables in our day to day existence that WarLight tends towards a pleasing simplicity, but the slippery slope effect of that will indeed lead to "rock, paper, scissors."

I would suggest that the ladders are close enough to the perfect expression of what you are trying to achieve.

See Nostance, that's where most people will disagree. Strategic ME is really not the best representation of skill with lowest amount of luck involved. The problem here is that there are about a dozen templates we've made that qualify in testing someone's skill with low amount of luck involved.


However when you get to 3v3, there's very few templates. Gnuffles and I are have only mentioned 2, and he's probably right about the leftovers issue on 4x5 0% WR Europe. Personally I always preferred 0% SR Europe too. So if we're going to make a standardized template, then the most standard test of skill with lowest amount of luck would be that 3v3 template, not any of the aforementioned 1v1 templates.