Application of Game Theory in daily life

[BurnedOut]

Post some practical applications of Game Theory that you (can) use to enhance your life.

---

Game: Gain an upper hand in online chats

Context: Strategizing your chatting patterns so that you are not left looking like an attention whore which can have a really negative impact on your self-esteem.

Players: 2

Choices: Chat ; Ignore

A (down) B (right)	Chat	Ignore
Chat	Both gain info, bond, etc (+1, +1)	A loses self-esteem, B gains dominance (-1, +1)
Ignore	A gains dominance, B loses self-esteem (+1, -1)	Both get nothing (0, 0) ​

Solution: Ignorance is a better strategy to pursue. Contrary to the commonly thought belief that being more available increases your chances of being taken seriously, seems right the opposite from a game theory point of view. Chatting has a better chances of yielding a worse outcome than ignoring.

Win-Win strategy: Set a schedule/appointment and hold the opposite party responsible for losing out. Since there is personal incentive involved in committing, skipping commitments comes with loss of dominance than gain.

Practical implications: I am surprised that such basic application of game theory works quite well in settings where you are dealing with people of your status.

 

[EndogenousRebel]

I see the logic here, but unless it measures actual standing the world it's missing a huge piece. Not to mention it's only remotely true with a very shallow and myopic lens. It's what the impulsive "reptilian brain" would follow along with, which I guess is what informs most of people perceptions.

Don't know how you derive that win-win strategy from that game, but you can do that without ignoring people. It doesn't work on people who are shameless either, they might care what a number of people think, but if they can make up for it or semantically shift blame on other things. If you try to "chat" with them to put them in a accountability judo hold they can just ignore you and pretend like your overstepping your bounds.

If you're trying to get a someones self-esteem just condense truths they can't handle into a quippy short jabs that can be seen as playful. If you know they are shameless do it like rapid fire in front people whom they the respect of. I promise I just know a lot of toxic people and am not toxic myself.

Of course like I said, if know peoples standing in the world you'll know it's often not worth it to engage in affairs like this. Some people live in a really small worlds and drag you into it to use you as a stepping stone. Never let someone disrespect you without recourse, just know when it's not worth the energy, energy better spent somewhere else, getting things you want and need. Doing that is always the dominant strategy.

 

[Daddy]

Interesting, I guess this could work, if (like Rebel says) they aren't shameless, and there are only two people chatting. If it's a group, people appeal to the group over the individual and the individuals gets drowned out, sadly (which is another reason why Facebook-liking posts is so gross to me).

 

[Ex-User (9086)]

How are you going to win if you don't play the game though? Ignoring is basically choosing to "pause" the game before you have a chance to lose or win. To me that's worse than losing.

If you use the same ruleset, but assume that the +1/-1 score is global and adds up with each opponent then it's actually worth losing to some opponents if it means finding people who don't ignore you so that you can keep infinitely milking that sweet +1/+1 bonus points to your life happiness score.

Let's say it takes you 200 attempts to find a player who doesn't ignore you and you try chat 3 times before you give up on the opponent. So you stand to lose up to 600 points and then you find a player who always responds to you so you get infinity points for losing only 600.

I wouldn't call that a self-esteem loss, more like an energy and time loss. If I have a friend that gives me 1 energy for every 3 energy I spend then it may be beneficial to look for a relationship where the energy ratio is closer to 1:1. Anyway if you have no friends then even a 1:100 energy ratio is good to ward off depression until you improve your social (energy exctraction skills) and find more active friends who are willing to give you more of their energy.

Though I wouldn't advocate for the opposite. Being the one who gets more attention and energy without giving back similar amounts is not ethical in my view. If the relationship is to continue then both parties should aim for the 1:1 ratio with some exceptions, otherwise the player who gets a lot of attention should "ignore" all the "chats" to end the relation which is more ethical than perpetuating an unequal energy distribution by for example using "chat" and "ignoring" 99 times and then repeating.

 

[BurnedOut]

Not to mention it's only remotely true with a very shallow and myopic lens. It's what the impulsive "reptilian brain" would follow along with, which I guess is what informs most of people perceptions.

That is the whole point of game theory. It assumes that your best decision is based on the best decision of somebody else. Even if this seems shallow, inane and utterly reptilian in nature, I have had too many first hand experiences of such a phenomenon. Since everybody is going to think of ignoring, you are forced to ignore as well because that is how you'll not play a dominatED strategy.

It doesn't work on people who are shameless either, they might care what a number of people think, but if they can make up for it or semantically shift blame on other things. If you try to "chat" with them to put them in a accountability judo hold they can just ignore you and pretend like your overstepping your bounds.

Trying to chat with someone shameless after observing their shamelessness makes it clear that your best strategy is to ignore.

Some people live in a really small worlds and drag you into it to use you as a stepping stone. Never let someone disrespect you without recourse, just know when it's not worth the energy, energy better spent somewhere else, getting things you want and need. Doing that is always the dominant strategy.

Sadly chatting is one of the things that you cannot run away from because not being on it much more lossy than being on it. This is not because of pop social media but basic things of life being transferred online.

How are you going to win if you don't play the game though? Ignoring is basically choosing to "pause" the game before you have a chance to lose or win. To me that's worse than losing.

It is not pausing the game but rather playing a game of chicken to see who caves in first. Imagine yourself chatting with your best friend. What if you keep getting extremely curt replies (one-two worders) for each of your elaborate paragraphs. Do you think your relationship with her is going to matter when your self-esteem is hurt due to lack of reciprocity? We like reciprocity and we are wired for it.

If I have a friend that gives me 1 energy for every 3 energy I spend then it may be beneficial to look for a relationship where the energy ratio is closer to 1:1. Anyway if you have no friends then even a 1:100 energy ratio is good to ward off depression until you improve your social (energy exctraction skills) and find more active friends who are willing to give you more of their energy.

That is another solution to the game. It counts under "cutting losses" where you lose a little while judging baseline behaviour and scoot your arse off to something better. The second proposition is a very dangerous solution to sadness. In TA, this is known as 'stroke starvation' or something along those lines. When you have a 'stroke (social attention) deficit', it starts affecting your self-esteem in definite ways.

If the relationship is to continue then both parties should aim for the 1:1 ratio with some exceptions, otherwise the player who gets a lot of attention should "ignore" all the "chats" to end the relation which is more ethical than perpetuating an unequal energy distribution by for example using "chat" and "ignoring" 99 times and then repeating.

1:1 ratio is sadly not achieved in life in all relationships when we take only one aspect into consideration. These are more things to consider and granted my analysis is extremely simplistic but that is the point. But every relationship should try to ensure that the sum of all games' payoff should not be nonzero.

 

[EndogenousRebel]

I'll give you that it in theory it could be a good defensive strategy. The only issue with this, according to conjecture from tacticians of war anyways, is that defensive strategies are rarely "winning" ones. If someone is throwing punches at you and you are only blocking them and never throw your own punch, then you risk not being able to block some punches and not putting in any damage to the opponent yourself, due to not making attempts. You get nothing from being strictly defensive.

Help me understand what is offensive and defensive in your matrix here. It's missing circumstance such as previous encounters and multiple agents.

• Agent A always trys to communicate, with everyone. Say they succeed 50% of the time.
• Agent X always ignore people and succeeds 100%, though lets say X only gets dominance points 75% of the time, meaning that 25% of the people that he ignores ignore him too (?).

The reality of the environment the agents are in, and what all agents in the game can derive from the information they have dictate the multiplicity of stories that could be told here. I'm going to substitute these dominance points, for just DP.

Dogs are smart enough to derive some sort of sociological structure, and while it's not unimpressive, most will simply mimic the behavior of the entity they think is the most dominant, or rather, the one that the dog thinks is best at doing things it values. Agressive owners make aggressive dogs.

Seeing as adult people reason in comfortable stories, I don't think it favors any of the agents playing the game to collect DP unless they are trying to achieve a specific thing over a certain period. Seeing as they don't have all the information I doubt they will even acknowledge DP and will attribute different points to this phenomena, points that they find sensible in the context of themselves and the environment.

I don't see how DP would necessarily favor Agent X. They are either very busy, have some sort of superiority complex, or are clinically depressed.

So yeah, unless Agent X puts information into the environment visa communicating, they forfeit complete control of narrative and ability to influence perception. But I guess since they don't have any information or bonds, they don't care and feel alpha as fuck?

 

[BurnedOut]

@EndogenousRebel

So you are questioning the whole premise by stating that people don't really care much about being ignored and being chatty online because the ones ignoring do not have a tangible payoff.

In Game Theory, you don't consider the words 'offensive' and 'defensive', you consider the words 'dominated' and 'dominant'. Sometimes an offensive strategy ends in vain as well.

Let us consider a hypothetical boxing match like the one you talked about

A (down) B (right)	Uppercut	Block
Hook	10,5	-1, 0
Jab	0,-3	0,0

Here, the best strategy for B is to Block because it yields no loss. If A gives a hook and if B blocks, it ends up yielding a negative payoff (let us assume that A is vulnerable for a few milliseconds enough for B to counter with a jab after the exhausting hook A throws at B). Therefore, A eliminates the possibility of hooking because he knows that B will block. Similar B cannot risk throwing an uppercut because if A jabs, it ends up negatively affecting her payoff. Therefore, we remain with two combos (hook block) and (jab block). A will choose to jab because it is the safest again.

The point is offense can be defense and vice versa so we need to eliminate the usage of these terms altogether.

We consider cases here which have A and B in a similar setting. Your point is valid when circumstances are very different:

If an extremely busy person (assume A is a social worker in Kabul) does not find time to text back and forth, she is also losing DP in losing out on her socializing time which in return eliminates the DP she can gain against her chatty writer of a husband B who works from home.

However, assume that A and B work in the same office and work similar jobs which roughly gives them the same circumstances. We will also assume that due to the usual occurence of homophily, A and B share similar social lives and mutual friends. We also consider chatting as the only game in play notwithstanding anything else. Despite all the restrictions, it's veracity still holds quite nicely in real life scenarios.

If A chooses to play chicken with B regarding chats, B is on the losing end. A gains control of B by having B service A even when not required which makes B stroke starved and particularly vulnerable to B's manipulation.

You can easily observe these phenomena between office colleagues and young couples and even high-schoolers and children as well. The one who receives more attention ends up acting smug notwithstanding her personality.

As far as considering multiple people, I don't know because I have just started out with the theory.

Can you provide more real life evidence of your propositions with the aforementioned restrictions?

 

[Ex-User (9086)]

That is another solution to the game. It counts under "cutting losses" where you lose a little while judging baseline behaviour and scoot your arse off to something better.

It's basically breadth-first search vs depth-first search class of algorithms. Or in plain English an extensive search vs an intensive search.

If you have 100 energy units or time units to do a search then it pays off to spend some % of it on a broad search to identify places where a deep search will be worth the effort.

 

[EndogenousRebel]

Dismissing the terms Offensive and Defensive because of the net outcome is not a good reason.
Dominance is a state. Offensive and defensive refers to tactics, and either (usually a combination of the tactics) can lead to dominance.

That boxing match device was just there to illustrate why defensive strategies are loosing strategies. Like seriously, there are books about war that people have been writing for hundreds of years, it's pretty well known.

Case Study
The Germans in WW2 were violently offensive with blitzkriegs and scared a lot of allied forces into strictly defensive strategies. It was possible to attack German bases directly, and in hindsight that would've been the right move, but they instead used those possible attackers to defend their own bases during blitzkriegs to minimize damage. Of course, it's the 1940s, so they hardly minimized anything at all. This is how they were able to get Poland and France to surrender very quickly (I don't trust this argument, but supposedly the same would've happened to Britain if the Germans were left to be uncontested by other international forces)

I supposed if you think offense and defense are interchangeable with each other because of net outcome, you very well would want to remove the extra noise. But you'd miss out on understanding the specific reasons for success. If you're playing a game to not lose the game, you will never win it. Higher risks = higher rewards, and if your enemy is gaining more rewards than you, and you give it to them freely, then you cannot really say you're winning.

I don't understand your boxing matrix at all. I'm to use to the binary prisoners dileman-eque matrix. You can try to explain it to me, but I recommend sticking to the original case study you put forward. Change that if you want.

Clarification of Dominant Strategy
A dominant strategy is just a strategy that plays against the opponents best move possible. Logically they will always do the best move possible, so logically you should always use dominant strategy. This doesn't mean that a dominant strategy won't end with outcomes where you are assuming more losses, it just means you're literally doing the best thing you could possibly be doing. If you dominant strategy trumps the other person, then congrats, you are most dominant in said encounter.

That boxing device was just to convey why strictly defensive strategies are bad. Let's not jump away from the original example you posited.

Issue of vocabulary?
DP, as I understood from your posts, isn't measuring Favor in a environment. I don't know, it might be measuring serotonin or something.

The most dominant is in the most favorable position. You don't gain any favor just because you ignore someone. People can derive whatever they want from internal and external narratives. If someone strongly favors you, they might just assume that you're busy, or that they made you mad, or are in a bad mood. If they don't favor you at all they will concoct some other narrative, like that you're rude or self-involved.

This idea of favor can be integrated in multiple ways, such as economic favor, social favor- between individuals and groups. These days- at least in developed countries, everyone is so independent that they see no reason to favor any one person too much, and won't give a second thought to someone ignoring them. Getting DP only actually gives you dominance against insecure people who value your opinion, and even then, there's social media and Google to validate people's choices.

 

[BurnedOut]

But you'd miss out on understanding the specific reasons for success. If you're playing a game to not lose the game, you will never win it. Higher risks = higher rewards, and if your enemy is gaining more rewards than you, and you give it to them freely, then you cannot really say you're winning.

The point of GT is to avoid taking these foolish risks altogether which lead to a bad outcome nevertheless. One of the extreme risk-taking events would include shutting up in prisoner's dilemma and expecting that your friend will shut up too. I am quoting this not because I have a different stance than you on the philosophical timbre but because I feel that your understanding of strategies within the GT are not too clear.


Re: Boxing Match example
Playing a game not to win can also get you a win, for eg. War of Attrition. This game is also popular in wars and sports but this is out of the purview of the discussion.

About my boxing match example, I wanted to simply portray that 'offensive' strategies may also lead to poor outcomes. In patois, we call such players being 'a bull in the china shop'.

Again, this example highlights the important aspect of not getting too caught up in gratifying yourself outright and not totally unforeseeable risks by completely basing your actions on the best strategy of your opponent.

The most dominant is in the most favorable position. You don't gain any favor just because you ignore someone. People can derive whatever they want from internal and external narratives. If someone strongly favors you, they might just assume that you're busy, or that they made you mad, or are in a bad mood. If they don't favor you at all they will concoct some other narrative, like that you're rude or self-involved.

This is the second time you are subjecting my explanations to reductionism that misses the essence of my point. I have mentioned the requisites for the particular kind of DP I am trying to talk about.

Re: Ignorance and DP
FYI, even if you don't gain favour by ignoring someone, you do gain an ego boost, a morbid pleasure in justifying your superiority by 'looking down' on someone. The mere preassumption that you might be better is a fun enough feeling to have.

 

[EndogenousRebel]

Perhaps one of us is mischaracterizing the others argument, idk I'm not insinuating anything. I'm more than anything trying to understand your points and trying to make understand my points.

The essence of an argument is the sum of it's parts. You can thank me for identifying weaknesses in it or not, but don't say I'm being overly reductive and a throw away reference to a reason as all it says is that you yourself have satisfied it by YOUR standards.

I thought I've demonstrated that I'm at least knowledgeable in actual real world applications of Game Theory. Dominant strategy is not static. I know this. But if the position of one agent is more favorable then they would always win. This is not how history, sports, daily life plays out. Strategies always have weaknesses and always need to evolve against the oppositions evolving strategies. It's becoming apparent that you are trying to make a point like that, so I don't how you haven't identified that in my text. I was just stating a key principle (in war at least): defense is not how you win.

Just like you can, I can choose to ignore points in your responses. Just like you can choose to ignore someone who wants to engage in conversation with you. The "strongest" premise to ignoring it is that the ignorer has concluded that engagement with said point/person is not worth their time to do X. If this conclusion is wrong then the ego boost is short-lived, only is so much as the ignorer can escape that fact.

Ignoring is most advantageous if you have a good reason. If you straight up tell someone it's because it feels good or you wanted to maintain dominance you might get a laugh, but if the person is actually around the same status, and are equipped with the mental to make you pay for it they will.

Just to add to the original intention of the discussion
Altruistic actions can get you "credit" with all sorts of people, and excuse future selfish actions. Not like anyone is keeping track of a point system, but it's just a funny thing. Like, why don't we just never excuse selfish action?

 

[sushi]

its better if you watch the game of thrones and apply what they are doing,

game theory is too abstract in many ways.

 

[crippli]

You can not use "etc". For me that doesnt work.

 

[birdsnestfern]

* deleted

 

[Hadoblado]

Reading back on the OP, I find the framing interesting.

The stakes are nested within the context of the relationship, which will cease to exist if you choose to ignore them. This kind of logic only makes sense if you're as afraid of being dominated in a relationship as you are motivated to have chat in the first place.

I'm not overly familiar with game theory, but it seems like it also ignores probability? If I have a good friend, the probability that they ignore me is low. Therefore, the expected gain for aiming for the +1/+1 is likely much higher than other options. And this can be seen in the way people behave, we do after all, chat. There should be any number of cues that give you extra information, which will change all the math to an unrecognisable state. People who come across as approachable are more likely to be approached, for a reason.

 

[dr froyd]

based on the payoff matrix in OP there's 2 nash equilibria: both always chat or both always ignore. So from an equilibrium perspective none is better than the other.

regarding dominant strategies, there's no strictly dominant strategy in this game. However "always ignore" is a weakly dominant strategy since it yields at least as good payoff as chat regardless of the opponents strategy. Since there's no strictly dominant strategies the equilibrium solution is found by comparing best-response strategies for each player.

in reality though this game would be a sequential game where the players repeat their steps many times - in which case the solution is a bit more complicated

 

[dr froyd]

i think it was Nassim Taleb that mentioned a real-life game-theoretical concept for how to make people treat you better: when people do something you don't like you become angry, but if you always deterministically become angry at the limits of what you accept, this is an exploitable strategy since then they know exactly how far they can push you. But if you randomly become angry at less clear "offenses", your boundary becomes unclear and less exploitable.

this is a case of what is common in more complicated "games" with hidden information; the optimal strategy is randomized where you pick actions based on certain probabilities.

 

[Black Rose]

I'm not overly familiar with game theory, but it seems like it also ignores probability?

Bayesian game - Wikipedia

en.wikipedia.org

In game theory, a Bayesian game is a game that models the outcome of player interactions using aspects of Bayesian probability. Bayesian games are notable because they allowed, for the first time in game theory, for the specification of the solutions to games with incomplete information.

 

[Hadoblado]

Thanks. I deserve that.

I don't mean it ignores probability completely, rather, it ignores the prior in the bayesian sense.

The probability that someone ignores me if I approach them is not 50/50, it's contextual and if I approach them I likely expect better than 50/50 odds, especially when you're gaming on margins as narrow as the dominance one feels when ignoring approaches.