# Do you take one box or two boxes? **URL:** \ **Category:** In My Humble Opinion\ **Created:** [March 11, 2026, 5:59am UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360 "2026-03-11T05:59:26Z")\ **Posts on this page:** 19\ **Page:** 22
**Author:** ![Stratocaster](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/stratocaster/32/16393_2.png) [@Stratocaster](https://boards.straightdope.com/u/Stratocaster)\ **Post date:** [March 20, 2026, 7:55pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/421 "2026-03-20T19:55:42Z")
We’ll have to disagree. 1,000 heads in a row with a purportedly fair coin defies belief. That sample is plenty big. I don’t care who the subjects are. Once the rules are explained there’s no plausible way _everybody_ takes two boxes. Nope. ---
**Author:** ![griffin1977](https://avatars.discourse-cdn.com/v4/letter/g/977dab/32.png) [@griffin1977](https://boards.straightdope.com/u/griffin1977)\ **Post date:** [March 20, 2026, 8:08pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/422 "2026-03-20T20:08:10Z")
> [@Stratocaster](#): > > 1,000 heads in a row with a purportedly fair coin defies belief Yeah but it’s not a random choice. It’s an actual human being in a room with 1000 dollars. There are any number of ways that what we are told is true, but they made sure their prediction would be correct, without outright cheating. The most obvious way is they always, or almost always (maybe you have a couple of ringers) predict two box, and just phrase the problem in a way that will mean anyone in their right mind will take the 1000. I mean, was everyone else told about how accurate their prediction was? Without that bit of information _everyone_ will take two boxes and at no point does it say everyone was told that. Predicting 100% is completely plausible. 50% one boxers is not. ---
**Author:** ![Stratocaster](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/stratocaster/32/16393_2.png) [@Stratocaster](https://boards.straightdope.com/u/Stratocaster)\ **Post date:** [March 20, 2026, 8:13pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/423 "2026-03-20T20:13:39Z")
Feels like you’re torturing the hypothetical to produce your preferred outcome. ---
**Author:** ![griffin1977](https://avatars.discourse-cdn.com/v4/letter/g/977dab/32.png) [@griffin1977](https://boards.straightdope.com/u/griffin1977)\ **Post date:** [March 20, 2026, 8:23pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/424 "2026-03-20T20:23:26Z")
> [@Stratocaster](#): > > Feels like you’re torturing the hypothetical to produce your preferred outcome No I’m just accepting human nature for what it is. If you put a 1000 humans in room with a grand and tell them they can take the money, 1000 of them, or very close to it, are going to take it. There are a number of ways of making sure the “close to 1000” becomes a thousand that don’t involve magical all knowing computers able to predict the future. And other way that comes to mind. We aren’t told that only 1000 people were experiment on, only “it correctly predicted what 1,000 people would do when presented with the scenario you are about to face”. Maybe at 316 it got a one boxer (who left empty handed), so the 1000 referred to are actually number 317 to 1317. ---
**Author:** ![puzzlegal](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/puzzlegal/32/3827_2.png) [@puzzlegal](https://boards.straightdope.com/u/puzzlegal)\ **Post date:** [March 20, 2026, 8:28pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/425 "2026-03-20T20:28:43Z")
There is a difference. In your second example, i assume the predictor _is_ cheating, and that makes its choice contingent on mine. ---
**Author:** ![Stratocaster](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/stratocaster/32/16393_2.png) [@Stratocaster](https://boards.straightdope.com/u/Stratocaster)\ **Post date:** [March 20, 2026, 8:29pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/426 "2026-03-20T20:29:28Z")
Yeah, we’re getting nowhere again. And I will accept again that someone is wrong on the Internet. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [March 20, 2026, 8:46pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/427 "2026-03-20T20:46:03Z")
And yet the future outcome will be the same in all cases. They look exactly the same Case one: it is cheating and there is causation of player choice leading to a sleight of hand placing or removing the money. Case two: both player choice and the prediction are caused by some aspects of the player’s nature that it was able to accurately assess, somehow, with high reliability. In each of those two universes the outcome is that the person who takes the opaque box walks away with a million dollars and the one who takes two walks away with a thousand. _I can’t by observation tell those two universes apart._ Neither can you. You however would opt to take the opaque box only in the first universe, that you are confident that it cheats. And take both in the one that you are less sure that it cheats. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [March 20, 2026, 8:48pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/428 "2026-03-20T20:48:32Z")
> [@Stratocaster](#): > > Yeah, we’re getting nowhere again. I disagree! I really am finding the explanations intriguing. The point isn’t to convince though. That’s not gonna happen. ---
**Author:** ![puzzlegal](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/puzzlegal/32/3827_2.png) [@puzzlegal](https://boards.straightdope.com/u/puzzlegal)\ **Post date:** [March 20, 2026, 8:55pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/429 "2026-03-20T20:55:36Z")
In the first, i also know know that half the people made each choice, and got the thing, and critically, i somehow know they aren’t confederates of the predictor. I am also more likely to take the single opaque box in the game show version. I can believe someone is giving away $10k over and over again. That could plausibly be an entertainment budget for a popular show. ---
**Author:** ![Babale](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/babale/32/15666_2.png) [@Babale](https://boards.straightdope.com/u/Babale)\ **Post date:** [March 20, 2026, 11:58pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/430 "2026-03-20T23:58:35Z")
> [@griffin1977](#): > > I mean, was everyone else told about how accurate their prediction was? Without that bit of information _everyone_ will take two boxes and at no point does it say everyone was told that. The hypothetical very clearly says that the previously predicted people were in the same situation as you, which means that they did have the problem explained to them, yes. > [@Stratocaster](#): > > Feels like you’re torturing the hypothetical to produce your preferred outcome. Indeed. ---
**Author:** ![griffin1977](https://avatars.discourse-cdn.com/v4/letter/g/977dab/32.png) [@griffin1977](https://boards.straightdope.com/u/griffin1977)\ **Post date:** [March 21, 2026, 12:27am UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/431 "2026-03-21T00:27:55Z")
> [@Babale](#): > > The hypothetical very clearly says that the previously predicted people were in the same situation as you, which means that they did have the problem explained to them, yes. It says “it correctly predicted what 1,000 people would do when presented with **the scenario** you are about to face”. You can read that as “the scenario” including the part about being told what the (999, 998, etc) people before them had done, but could easily not. That would be a very obvious way to almost guarantee 100% accurate prediction without any cheating or God like computers. ---
**Author:** ![griffin1977](https://avatars.discourse-cdn.com/v4/letter/g/977dab/32.png) [@griffin1977](https://boards.straightdope.com/u/griffin1977)\ **Post date:** [March 21, 2026, 1:00am UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/432 "2026-03-21T01:00:25Z")
> [@DSeid](#): > > hand placing or removing The key point is one of these is far more likely than the other. They could totally remove the million when you choose two boxes, that I’d get (though it would be much easier just to not put it in to begin with). No one is cheating to give me a million dollars, versus just saying “wow! You are the first person to beat the all powerful super computer. Congratulations! Here is your certificate saying that, which is all you leave with.” ---
**Author:** ![scabpicker](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/scabpicker/32/8268_2.png) [@scabpicker](https://boards.straightdope.com/u/scabpicker)\ **Post date:** [March 21, 2026, 5:49pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/433 "2026-03-21T17:49:12Z")
Heh, yeah. I still don’t trust anything that says it has a perfect prediction rate. This one sounds slightly more honest since it says it’s only 99.9% accurate, but it’s still ridiculous. So still a two-boxer at hear, but I still might just grab one. ---
**Author:** ![Fretful\_Porpentine](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/fretful_porpentine/32/2873_2.png) [@Fretful\_Porpentine](https://boards.straightdope.com/u/Fretful_Porpentine)\ **Post date:** [March 22, 2026, 7:23pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/434 "2026-03-22T19:23:49Z")
> [@griffin1977](#): > > Though that changes both of our cost/reward matrix. I’m now betting 10 dollars someone is legitimately giving away 5 million, by giving anyone who wants to leave 10 bucks 10000 dollars. Thats a different bet to betting 1000 dollars someone is legitimately giving away 500 million by giving anyone who wants it 1000000. > > The first bet i’m probably taking. Sure it’s probably a scam and I’m not getting 10000. But it’s 10 bucks who cares? Second bet probably not. 1000 bucks is not life changing but it’s still 1000. I’d take the guaranteed 1000 over maybe getting a million if this isn’t a scam. Yeah, I think that’s pretty much where I am. I feel like it would be _stupid_ to leave a guaranteed $1000 on the table. $10 I can live with, and I might be intrigued and curious enough to do exactly that, even though I still maintain that _if the money is already in the boxes and the amount can’t be changed_, taking two boxes always guarantees that you do at-least-as-well-and-maybe-better than taking just one box. ---
**Author:** ![IridiumIodineSulfur](https://avatars.discourse-cdn.com/v4/letter/i/b9e5f3/32.png) [@IridiumIodineSulfur](https://boards.straightdope.com/u/IridiumIodineSulfur)\ **Post date:** [May 7, 2026, 1:54am UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/435 "2026-05-07T01:54:47Z")
Can someone explain why you would pick both? I read about this paradox before running into this thread. I’ve never understood. ---
**Author:** ![colinfred](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/colinfred/32/21338_2.png) [@colinfred](https://boards.straightdope.com/u/colinfred)\ **Post date:** [May 7, 2026, 11:15am UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/436 "2026-05-07T11:15:14Z")
> [@IridiumIodineSulfur](#): > > Can someone explain why you would pick both? I read about this paradox before running into this thread. I’ve never understood Here’s a thread to read where mulitple people explain it: [Do you take one box or two boxes?](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/) ---
**Author:** ![griffin1977](https://avatars.discourse-cdn.com/v4/letter/g/977dab/32.png) [@griffin1977](https://boards.straightdope.com/u/griffin1977)\ **Post date:** [May 8, 2026, 8:53pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/437 "2026-05-08T20:53:29Z")
> [@IridiumIodineSulfur](#): > > Can someone explain why you would pick both? There’s not much to explain. There are two boxes, one may have 0 or 100000, one definitely has 1000, you can take just the first box, or both. There is no downside to taking the second box, it can’t effect whether the first box is empty or not. Why wouldn’t you take both? ---
**Author:** ![griffin1977](https://avatars.discourse-cdn.com/v4/letter/g/977dab/32.png) [@griffin1977](https://boards.straightdope.com/u/griffin1977)\ **Post date:** [May 8, 2026, 8:59pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/438 "2026-05-08T20:59:56Z")
> [@scabpicker](#): > > I still don’t trust anything that says it has a perfect prediction rate Again when the thing your predicting is “will I refuse a guaranteed 1000 dollars that’s right in front of me when there is no downside to taking it”, I don’t think there anything particularly unbelievable about it. ---
**Author:** ![IridiumIodineSulfur](https://avatars.discourse-cdn.com/v4/letter/i/b9e5f3/32.png) [@IridiumIodineSulfur](https://boards.straightdope.com/u/IridiumIodineSulfur)\ **Post date:** [July 31, 2026, 7:15pm UTC](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360/439 "2026-07-31T19:15:33Z")
I wouldn’t pick both because if I do, I only get the 1000. If the predictor is truly accurate, I get 1000 if I pick both and 100000 if I pick just one. It seems to me like picking between 1000 or 100000? [Previous page](https://boards.straightdope.com/t/do-you-take-one-box-or-two-boxes/1028360.md?page=21)