# zero-sum game

Victor Steinbok aardvark66 at GMAIL.COM
Sun Jun 3 02:19:31 UTC 2012

```The China Daily link (shortened here http://goo.gl/8AHyo ) only uses
"not a zero-sum game" in the title, so it's hard to judge for sure. But
I don't think I have any problems with this one--the implication is that
in a zero-sum game there has to be a winner and a loser and the US-China
relations don't need to be that way (that is, both sides can win). Given
both a reasonable and an unreasonable expectation (reasonable because
cooperation benefits both sides and unreasonable because, with two
sides, a positive balance for one is a negative balance for the other
(i.e., if one has a trade surplus, the other must have a trade deficit).
The India/China story ( http://goo.gl/8i8lv ) is similar, except instead
of bilateral trade, it's a question of US benefit from relations with
two competing nations (it's not zero-sum for US in any case; the issue
here is whether it's zero-sum between China and India).

The trash story (http://goo.gl/5z6sV ), however, makes no sense. They
seem to imply that "zero-sum" means "no one loses", which is not the
meaning at all. The GigaOm title (App discovery is a zero-sum game
http://goo.gl/BbD3J ) is also ridiculous, given what the original
meaning is.

Wiki is pretty straight-forward:

> In game theory and economic theory, a zero-sum game is a mathematical
> representation of a situation in which a participant's gain (or loss)
> of utility is exactly balanced by the losses (or gains) of the utility
> of the other participant(s). If the total gains of the participants
> are added up, and the total losses are subtracted, they will sum to zero.

With two or more players, that simply means that, for anyone to win,
someone has to lose. Politics is not a zero-sum game (unless you're a
2010-12 Republican). A poker tournament /is/ a zero-sum game.

The Skip Moen piece starts out not unreasonable. If you look at it from
the perspective that there is a fixed pile of of blessing/grace at any
given time, then some getting it means others will be left without.
Zero-sum means unchanged balance, not that balance is always zero.
However, then the explanation goes off the deep end. It's almost like a
pair of dueling zero-sum games. Not only is the grace balanced
"horizontally", it is also balanced "vertically"--that is, if you use up
your limited supply now, there will be none left for you in the
afterlife. Or whatever... So you have multiple balancing going on, so
it's a bit complicated to set it up as a zero-sum game as a whole.

I am not sure what to make of the TechCrunch piece. Karbasfrooshan
suggests that there is a fixed pool of customers, so if each customer
must buy A or B, but not both, then it's a zero-sum game. On the other
hand, this is a really stupid economic model, so the critics are right
too. Zero-sum does not mean that the total utility must stay zero--only
that the net change in utility is always zero.

The Sven Grimm piece on Africa ( http://goo.gl/Uabwu ) has the advantage
of not even being wrong. His use of zero-sum seems to be OK, but it's
actually not relevant to the situation that he assigns it to.

The final piece (time: http://goo.gl/8i8lv ) seems to be a bit of a
stretch, although I don't sense that the usage is particularly wrong. If
you have a fixed resource, the distribution of that resource is a
zero-sum game. But the winner/loser is not a third party (the person
doing things within a limited amount of time). The players are the
different activities--the time spent on one cannot be spent on another.

So, it seems, in most cases the usage is either correct or marginally
correct, but is then attributed to the wrong "game". There was one
glaring exception where the usage is clearly wrong and one where the
usage is correct but absurd.

Another example that is both right and wrong popped up today:

http://goo.gl/ItUE7
> The fact is, fund companies and other providers of 401(k)s are getting
> rich off these plans. And in this zero-sum game, future retirees are
> definitely the poorer for it.

We have a clearly identified winner and a corresponding loser. So the
contextual conditions appear to have been met. But if you read the whole
piece, there is absolutely no explanation where this phrase comes from.
In fact, 401k funds are most certainly not a zero-sum game, by any
stretch of imagination. Here, the author merely seems to imply that
whatever the clients lose, the companies gain, which is not entirely
true, even when it comes to fees.

Another Indian piece appears to have it completely wrong.

http://goo.gl/lEfZ9
RBI armtwists some dollars out in 'zero-sum game'
> The Reserve Bank of India on Thursday forced an injection of about
> \$2.5 billion into the currency market with a unequivocal fiat in its
> fight to rein in an increasingly flyaway rupee.

An external injection of a commodity is precisely what a zero-sum game
is not. Another way to define a zero-sum game is a "closed system".
Here, there is an obvious violation. Reading a bit further suggests that
the headline was pulled from a quoted comment (as the quotation marks in
the hed actually point to as well).

> "If this assumption is true, there may not be generation of
> significant dollar supplies in the system as companies will resort to
> selling cash dollars and buying forward dollars to meet import
> obligations. The RBI's firm actions help to release bearish momentum
> on the rupee to provide marginal correction and not trend reversal. On
> the other hand, it also generates higher demand from importers with
> forward dollar looking cheap to acquire. Hence, it is a zero-sum
> game," said Harding.

This explanation does not point to a zero-sum game, but rather to a
negative relationship--the increases in one correspond to losses in
another, but there is simply no balance to be maintained.

Yet another Indian piece on the condition of India's mutual funds seems
to be just wrong.

http://goo.gl/gcV76

A sports reporter, on the other hand, summed the situation up rather nicely.

http://goo.gl/VkdwQ
> For better and for worse -- mostly better -- that's Rondo. The points,
> the assists, the stats, they don't mean a lot to him. He views
> basketball as a zero-sum game: Somebody won, somebody lost and the
> rest of it is, as he says, "irrelevant."

VS-)

PS: New proverb mash: when you're dumber than a bag of hammers,
everything looks like a bunch of rusty nails.

On 6/2/2012 6:56 PM, Benjamin Barrett wrote:
> Thank you for that response. I'm glad to see there's a reason for my =
> puzzlement.
>
> The grammar "X is/are a zero-sum game" seems to be normal.
>
> I looked at a few examples on the web, and some seem to use it =
> correctly, while others I still cannot figure out. Here are a few that =
> seem odd:
>
> Here's on one Gigaom that seems to mean "nothing positive results from =
> engaging in the zero-sum game (using an app to find apps)":
> http://gigaom.com/apple/app-discovery-is-a-zero-sum-game/
>
> And here's one from the China Daily that seems to mean "promoting one's =
> own interests without regard to others' interests":
> =
> 18.htm
>
> Here are two that take "zero-sum" more literally:
>
> =46rom the TriplePundit, seemingly posing not replenishing wasted =
> materials as a zero-sum game because then there is no waste (zero equals =
> zero):
> http://www.triplepundit.com/2010/08/zero-waste-a-zero-sum-game/
>
> Suffering in this life is balanced by rewards in heaven, resulting in a =
> "zero-sum game" (though there is also mention that this same zero-sum =
> game involves the wicked benefitting while the good suffer):
> http://skipmoen.com/2012/05/20/zero-sum-game/
>
> When used more or less correctly, it seems that "zero-sum game" does =
> indeed mean one party gains while the other loses with the difference =
> being more or less equal:
>
> Here's an article that says that competition for products is a zero-sum =
> game, in that you buy Company A's or Company B's widget but not both, =
> but that content is not a zero-sum game:
> http://techcrunch.com/2012/03/31/is-technology-a-zero-sum-game/
>
> Here, the US says that it can grow relations with both India and China, =
> and therefore its relations between the two are not a zero-sum game =
> (pitting two relationships against each other in the zero-sum game):
> =
> http://timesofindia.indiatimes.com/india/Ties-with-India-China-not-zero-su=
> m-game-US/articleshow/11751450.cms
>
> Here's one that puzzles me from the New York Times where the zero-sum =
> game is an African country having good relations with China OR the US =
> but not both (the article argues that is not the case):
> =
> ar-with-china/avoiding-a-zero-sum-game-in-africa
>
> Here, "time is a zero-sum game" (opportunity cost in terms of time) =
> because you can do A or B within time X, but not both:
> http://www.brasstackthinking.com/2010/11/playing-the-zero-sum-game/
>
> Benjamin Barrett
> Seattle, WA
>
> On Jun 1, 2012, at 9:27 PM, Victor Steinbok wrote:
>
>> Like other originally technical terms, this is one is often misused.
>> There should be some tracking done on how people use the expression, =
> but
>> I doubt it's any indication of an actual shift.
>> =20
>>      VS-)
>> =20
>> On 6/1/2012 8:20 PM, Benjamin Barrett wrote:
>>> I've struggled with the expression "zero-sum game" for quite a while. =
> I often think it's my lack of understanding, but I think I've found a =
> clear example where the use varies from the standard definition, such as =
> "(of a game or situation) in which whatever is gained by one side is =
> lost by the other: altruism is not a zero-sum game" for "zero-sum" in =
> the Mac dictionary.
>>> =20
>>> "On Facebook, 'Likes' Become Ads" by Somini Sengupta in the New York =
> Times (on post-gazette.com)
>>> =
>>> =20
>>> -----
>>> "So Sponsored Stories creates a zero-sum game," Mr. Goldman wrote. "I =
> as a user probably don't get any value from the public presentation of =
> my implicit endorsement (if anything, it might hurt my position with my =
>>> -----
>>> =20
>>> Here, Facebook users who "like" a product are then featured as liking =
> that product to their Facebook "friends." Mr. Goldman's face is =
> appearing on his "friends'" feeds or ad bar, but I don't think he's =
> losing anything at all.
>>> =20
>>> Even in the case of altruism (the example in the Mac dictionary), the =
> benefactor is going out of their way to do something nice to the =
> recipient of the altruistic act, so there is some sort of loss, but can =
> Mr. Goldman's appearance as pseudo-ads somehow factor in a zero-sum =
> game?
>
> ------------------------------------------------------------
> The American Dialect Society - http://www.americandialect.org
>

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org

```