<< Back to Warzone Classic Forum   Search

Posts 1 - 20 of 30   1  2  Next >>   
The pinnacle of Metagaming: 12/29/2023 19:21:16

Level 64
A few months ago an unusual idea had been born in my mind: Can you benefit from giving away some information to your opponent for free?

Here are a few examples of the concept:

1) A direct threat to your opponent's bonus completion:

A very naive but still possible in theory method to frighten them with potential counter that causes to cancel the attempt.

2) Force their last order in your favor:

In this particular scenario the blue player has SEA+CA+Argentina starts with intel only on Ant. The plan is to go for +11 income in 2 turns and counter Ant at the same time. There is also a possibility for opponent to go for +12 income if they have EC+Aus as well. The problem is that move order is unknown and the counter may not work. In order to solve this particular problem you are giving them info that you border EC, therefore forcing them to have the last order on EC and not Ant.

3) Scout around for safety:

Probably the most complex concept yet to be found. If you have 2 relatively safe bonuses to take first and you are not sure if you can open your 3rd bonus right away (in this case CR), then you have this option to invite your opponent towards you as soon as possible before you open your last bonus for completion. This strategy can repeat itself for a few turns if you see no action until then.

Now the real question to all the big brain thinkers: Now that you know that these strategies/tactics/tricks exist, can you ignore them?

I myself have trouble with these shenanigans and not entirely sure what conclusion to make, but my intuition says that there is something really deep and special about this that can cause unnatural behaviour even when you are aware of what is happening. Furthermore, I may never be smart enough to exploit it.
The pinnacle of Metagaming: 12/29/2023 19:55:01

Level 63
Yes this is brilliant. Sometimes expanding funny or making weird picks can heavily effect enemy intel gathering. And intel wins games! For example, if I get my 6th pick I will win 9/10 times.

However, my favorite trick to mess with enemy intel is in team games to transfer armies to help a teammate expand. It causes intel gathering to be much more difficult, since you have to account for 2-3x the possibility

A trick to give off intel in order to be in better position — hard to think of tbh — with light fog it’s a lot more relevant, since enemies can see territories but not deployments so everything can be kinda meta

Edited 12/29/2023 20:57:10
The pinnacle of Metagaming: 12/29/2023 20:37:29

Level 60
Mind -> blown

Thanks so much Rufus for sharing your phenomenal tricks and tactics with the community!
The pinnacle of Metagaming: 12/29/2023 20:38:53

Photonic Symmetry
Level 60
I certainly don't understand this at all. To be honest, case 1 would have crossed my mind before but I dismissed it as too naive and risky to attempt. Case 2 disproves this assumption. It is a very good demonstration of where it can be potentially advantageous. Case 3 eludes me. Still can't quite wrap my mind around it.

There is another potential avenue for these sorts of "metagames" via chat, where you can reveal more information than just ghost attacks. I would imagine in a scenario where the result of a game is down to predictions, relaying all or selectively relaying some of your orders would very much pose a problem for your opponent. This avenue may become (or already is) taboo in the same way stalling is, but I have seldom seen it exploited or talked about in the same manner.
The pinnacle of Metagaming: 12/29/2023 20:42:33

Level 59
So if I've got this right, this looks like a way of playing mind games with your opponent to trick them into doing some bad moves. Am I correct in thinking that?
The pinnacle of Metagaming: 12/29/2023 20:46:30

Photonic Symmetry
Level 60
It can be ignored though. Case 2 backfires if you don't have last order T2 and opponent remains more wary about a counter in Antarctica.
The pinnacle of Metagaming: 12/29/2023 21:29:11

Level 54
Ռուֆուս, օկուլտիզմի թագավոր, Գնուֆոնի ստրուկը
The pinnacle of Metagaming: 12/29/2023 22:12:06

Level 68
I can see the first one working if you’re also completing a 5 bonus and know you won’t have last move to counter with a 3.

I’ll wait for you to post a game link where you do one of these tricks. It’s too big brain effort for me
The pinnacle of Metagaming: 12/29/2023 23:45:50

Kenny • apex 
Level 59
Case 3 is my favorite.
The pinnacle of Metagaming: 12/30/2023 03:59:09

Level 62
Does case 3 works like delay moves?
The pinnacle of Metagaming: 12/30/2023 04:07:13

Level 61
Even though you could influence your opponent with this kind of move and take a win, the correct answer to the puzzle is:

This move is never a good move to make.
Playing good in warzone is all about finding non-dominated moves. For optimal play, after you managed to weed out all the dominated moves, you would have to assign probabilites (based on possible positions after making the move, depending on enemy moves) to the non-dominated moves and then randomly choose one of those based on the probabilities given.
This basically ends up being an overly complex version of stone-paper-scissor (only that the probabilites likley arent evenly distributed as for stone-paper-scissor, where the optimal (as in not counterable) strategy is 1/3, 1/3, 1/3) and its not mirrored, as each side has different options to choose from. By giving away info as shown, you ll certainly change this gambling-game and you may still win it with a correct prediction, but you will always reduce your advantage by doing so.
The pinnacle of Metagaming: 12/30/2023 04:13:09

Level 65
I like option 3 for the reason that you don't require additional intel to perform it, rather the absence of intel. However, does it really provoke a good opponent to border you? Especially now that the thread is out, they might as well understand what you are doing :p
The pinnacle of Metagaming: 12/30/2023 05:07:08

Level 65
I'm not sure how to feel about this, I think case 2 is definitely the most useful, however for cases 1 and 3, I'd argue that they don't really change much. Case 1 if you've already committed to the bonus, you can't take the risk of leaving yourself behind on development because you know your opponent is there, I'll even sometimes go for a bonus on picks knowing my opponent is there because I don't want to be behind in development. As for case 3, a stronger opponent will know that they should try to hide as much information as possible from you, I almost always hesitate to reveal a pick I have even if I know my opponent is there, in the hopes he commits to it at some point. I can't help but also feel like these moves make it pretty obvious that you're trying to play mind games with your opponent, you see that and you know there is a good reason he is giving you information, don't play into the tricks and just do as you usually would do anyway. Against weaker players I think this is great to trick them into making dumb mistakes, but I think at the top level this could indeed just become rock-paper-scissors and give you no real advantage. If anything, all it does is give your opponent intel, which could be seen as disadvantageous if he tries to take advantage of the extra information.

Edited 12/30/2023 05:10:02
The pinnacle of Metagaming: 12/30/2023 05:24:13

Roi Joleil
Level 60
-Can you benefit from giving away some information to your opponent for free?-
Well, you arent doing it for free but expecting a greater return. Else it would simply be a no.

I dont see how you would gain anything in this example.
Either the opponent didnt intend to go for the bonus, in which case your threat does nothing and he either didnt see it (didnt move) or now knows to first secure the territory.
Or he planned on getting the bonus and now you saved him from taking a couple of neutrals and instantly losing if you would have just countered t2.

I like that one a lot because for 1 reason.
You arent 'really' giving away information and you never planned on countering it anyway.
Basicly the difference i see in scenario 1 and 2 is that in 1 there was an "either / or" situation where you either would counter like normally or give away information.
But here this isnt an "either / or" because you never intended on countering. So its only an addition to a plan you would do anyway.
One of 3 things would happen.
1: He just wasnt in EC in the first place which means you gave no information away (and if anything gained information that he definetly isnt there the next turn when he doesnt show up)
2: He simply ignores it and goes for the t2 12 inc anyway in which case your move didnt do any harm.
Or 3: He gets scared of the t2 12inc and secures the double border first. Which isnt a bad thing because the opponent would have gone there anyway so it isnt like the position go worse.

The move into ER makes sense if you dont know he is there. Pressure him into securing the territory.
The move into caucasus seems absurd tho. Considering the idea would be to not move turn 2 (and turn 3?) either, the opponent in WR could just move into CR to the double border and blockade that befor you can exploit it making the position much worse.

my take to the question is:
I think if the following is given it can make sense.
-You dont know if the opponent is there
-Its beneficial for the opponent (if he is there) to secure the territory immediatly
-You didnt plan on taking the territory for yourself anyway

Edited 12/30/2023 05:29:11
The pinnacle of Metagaming: 12/30/2023 09:04:37

Level 62
In theory I am with Giant Frog, as this sort of moves only give more information to your opp, thus helping to solve the puzzle.

Basically he is always free to ignore the extra intel, or he can play the game of trying to infer why he has given that intel and come to an even better deduction (or on the contrary be fooled...).

In practice, as this is a predict game to a certain extent, it creates new options and may influence games. Who just solves Nash equilibrium and randomize decisions each turn? (Which btw is just good for making an average move). I never do so, and always try to infer opp moves based on how I know him. So I'd basically be lured in the game and try to infer why he is giving to me the free intel.

But above all, as I speak Armenian, I think QH is the one that I fully agree with and centered the point.
The pinnacle of Metagaming: 12/30/2023 09:15:22

Level 64
To clarify some details for each case scenario:

1) Maybe I should have been more clear, but the important thing there is that you either don't have intel on that bonus or you can't afford to counter it.

2) As Roi pointed out, it is almost like a no loss situation, since you do not plan to go there (EC) anyway.

3) First of all, Caucasus move might be misleading, because on this board WR was wastelanded, so my bad for not showing this. Secondly, this is more complicated than it seems at first. The purpose is not only to scout around, but encourage a different behaviour from your opponent as well, e.g. they might overdeploy going towards you and lose tempo.

So overall, it is not as simple as "mind games", in my opinion. But it has potential to disrupt your opponent's natural flow, that's why at the end of it I asked can you really really ignore it? After so many games being played my take is that I am not sure if our understanding is deep enough to see the real advantages in this, and something like a perfect AI maybe could exploit it.

Edited 12/30/2023 12:14:03
The pinnacle of Metagaming: 12/30/2023 13:47:56

Level 61
Who just solves Nash equilibrium and randomize decisions each turn? (Which btw is just good for making an average move)

Noone does it, but also i dont think there is anyone capable of doing that. However, i think we do intuitively try to do the same thing, except for the randomisation part, where most ppl try to out-predict their opponent instead. But figuring out which moves are being dominated by others and which arent is pretty much what makes a good player i think.
Not sure why you say you d only make average moves with Nash equilibrium. You d be playing perfect. I think what you are thinking of is a game where no action is dominated by others, where nash equilibrium basically only makes it so that you cant be countered, but there could be plenty of strats that you only draw against, while someone who manages to out-predict the opponent may get better results (nash equi would still be unbeatable (as in, you can never have >50% chance to win against it), just not beating some other strats hard enough)

Edited 12/30/2023 13:55:02
The pinnacle of Metagaming: 12/30/2023 14:25:30

Roi Joleil
Level 60
Consideration the clarification to 1) and 3), i would be more willing to say "yea you could definetly do it" as it aligns more with the 3 points i would consider need to be fulfilled.
Obviously tho the question willl remain, this being only some mindgames, if it will even get a desired effect and not backfire.

As for the AI, an AI would never do that. Youd neither be able to handcraft it nor would it ever learn that through self play.
Handcrafting is simply to difficult and there are a ton of other problems much more relevant in solving anyhow.
And for self play to pick it up, its simply to ambiguous and miniscule to do it.
And if we would just say "well hypothetically if wed say it just happened to exist tomorrow" then id question "ok then we might aswell imagine an AI with unlimited computing power who can just simulate all scenarios and play thhe statistically most precise move" making the mindgame redundant.

Edited 12/30/2023 14:29:22
The pinnacle of Metagaming: 12/30/2023 15:23:17

Level 58
Some mavelous explanations here by GiantFrog.

Playing good in warzone is all about finding non-dominated moves. For optimal play, after you managed to weed out all the dominated moves, you would have to assign probabilites (based on possible positions after making the move, depending on enemy moves) to the non-dominated moves and then randomly choose one of those based on the probabilities given.

I recently tend to get a bit frustrated seeing picks which just outright lose against the most basic meta pickset (2 turn 12 income; picking the FTB,...).

I believe that the notion that perfect play is a mixed strategy is a pretty important one. Back in 2013, when I was learning the game, I did so by looking at picks from people like dead piggy or szeweningen. After some time I understood their picks pretty well but usually they didn't pick exactly what I would have gone for. I found the thought that two players can pick a board flawlessly without picking the exact same picks very helpful. It also helps when having a tolerant discussion where you have to distinguish between a clear error in the thought process and some picks which also are reasonable but just not something I would have personally gone for.

Edited 12/30/2023 15:25:12
The pinnacle of Metagaming: 12/30/2023 20:01:45

Beep Beep I'm A Jeep 
Level 64
I like this, and even just for confusion in some cases.
It can maybe be compared to a MA-LF game, where you can make random attacks all over the map and your opponent will see them in the order summary and history. It's annoying as heck, and will cost your opponent time and nerves, but technically it's completely pointless.

In reality, we can say here that this technically is a useless behavior in many cases blah blah blah but the truth is we get confused and make mistakes all the time, and I can see this being part of it.
And I'm not even sure about the useless behavior part.
Now maybe a key question is to find the difference when it adds to confusion and when it just gives away free info.

Thanks for sharing Rufus

Edited 12/30/2023 20:08:47
Posts 1 - 20 of 30   1  2  Next >>