The strictly rational game-theory answer is that it would depend on the "expectation value". An "expectation value" is a weighted figure of how much an option is worth to you after factoring in how likely it is to pay off.

So for example, let's say I have two options: option A is redeem FTL right now. Option B is to wait and have a chance of winning $100 in games, this prize to be assigned randomly to someone who posted in the thread. Let's say that I estimate 50 people will post in the thread. Let's also say that I'm 100% confident that nobody will redeem FTL before the timer runs out, but I figure half of us will camp the the thread and try to redeem it the second after the timer runs out, so I'm estimating that I have a 4% chance (one in 25) of getting the FTL code if I wait.

Expectation value A is: $10 (cost of FTL) x 1.00 (100% odds of success) = $10.00
Expectation value B is: ($100 (prize up for grabs) x 0.02 (2% odds of winning it)) + ($10 (cost of FTL) x 0.04 (4% chance of nabbing it if I wait)) = $2 + $0.40 = $2.40

So in this case, the rational choice is to simply redeem the FTL code right now. And if you remove the assumption that other people won't redeem the code early and end the giveaway, the odds of option B paying out get even lower.

Of course, people don't make these decisions on purely rational grounds - there are all sorts of other factors at play, such as: whether their identity will be known (and they will thus be known to have "ruined it for everyone"); how well they know/like the other participants (and thus whether they have an interest in increasing the "wealth" of the group collectively, even at the risk of not sharing in it personally); the fact that most people are really bad at estimating expectation values, especially where one option involves a delayed payout; etc.

In the example you ran, I'd say that it was a given that people were going to redeem the codes right away. They had a 100% chance to immediately win a game they (presumably) want, versus a slim chance of possibly winning a game worth less than the code you posted and almost certainly getting scooped on the original game. The only reason they had to wait was altruism, which is unlikely to be a strong factor except in a small closed group who know each other. I'm betting you could have slowed things down by requiring people to publicly post and ask for one of the three games, and even more so if you had required them to say something like "Austrobogulator, I want to deny games to the community for my own benefit. Please give me [game X]." I still doubt it would have made it the full 24 hours, though.

Interesting read.

I did actually think that there was a chance that people would want to 'increase the wealth of the group'. Also, the idea I had for the giveaway was to post the codes in plain text right off the bat, although, I did add ?s, because having them redeemed by bots would be pretty shitty. Mainly, I wanted to subvert the way giveaways usually work - so, I didn't really want to go down the route of having people ask me if they may have the code. That's definitely something to consider for the future though.

I suppose I definitely am either naïve or optimistic (or both) given that I did actually think that there was a chance that this would last a day - especially on the Gog forums. Regardless, it was interesting to see the result, especially since only one redeemer carefully weighed up the outcomes of the situation.

Before you get too cynical, I would point out that we don't actually have any good numbers on how many people had an opportunity to redeem the code but chose not to. It's possible that if we had the full picture, we might see the GOG forum-goers behaving very altruistically in general, with a small percentage of outliers being enough to end the experiment.

... And how should they have gone about it, eh?

The strictly rational answer is that for the group it's a zero-sum game either way. Because YOU are part of the group.

Well, not exactly; I already have all of those games, so the codes are of no direct value to me personally...

The extra 5.99 games which you meant to give away if the codes stay unredeemed. (Although even unredeemed codes that you sit on are still "community wealth" - true, redeeming them and playing the games would create positive value in the form of "fun", but they might as well end up in a backlog.)

First of all, thanks for your generosity, Austro!

Second, as others have said before me, with the codes placed in plain sight, I see no possibility that they had lasted even a tiny fraction of the time required. You'd need everyone who read the OP deciding to wait, against only one, two, or three persons who didn't. The odds were definitely against it, I'm afraid.

Nevertheless, it was an interesting idea for a giveaway, and an interesting thread to read, too. ;)

If this giveaway has any saving grace, it's the discussion about game theory and whatnot :P

And besides, you've made the day of two different people.:D

I almost forgot: congrats, lugum and blotunga! ;)

You might enjoy this TED talk on behavioural economics and some of the common errors humans make.

Yep, watched that a while back. Very interesting :D