[arthulia]

I suspect most people would not consider buying a $5 loot box if there was a 95% chance of only getting duplicate items, or a 0.0002% chance of getting the item they're hoping for.

Developers will probably increase the drop rates for things so that people will continue to buy their crates.

I agree it's still gambling but this is a step.

[ThrowawayR2]

Since you mention odds, it's interesting to look at the stats of the companies that provide official information about their odds.

For Blizzard's Hearthstone, for example, their stated odds of getting their highest category of cards (legendary) is 1:100 and given that that 1% chance is spread across several different cards across multiple sets, getting a specific legendary card is only a fraction of that 1%.

For Nintendo's Fire Emblem Heroes, the chance of getting their highest category of characters (5 stars), is a relatively generous 6%. However, that's split across 2 batches of 3%, one of which is split across 3 specific characters (i.e. 1% specific card) and the other which has about 100 cards (0.03% to get a specific character). As each character is one who appeared in the long-running Fire Emblem series of games, there's significant incentive to try and get specific characters.

And these are games from companies that behave relatively ethically in this space. I can think of Asian games that are much worse.

[0] https://hearthstone.gamepedia.com/Card_pack_statistics

[1] https://feheroes.gamepedia.com/Summon_Heroes

[minimaxir]

It should be noted that Hearthstone's odds are not stated (in the US): they were only found out by user research, and as a result of Apple's new rule, would likely need to be explicitly disclosed.

In contrast, Fire Emblem Heroes has always been upfront about the odds, and you can see the odds whenever you want. (although in my personal experience the 5 star odds are less than 6%, hmrph)

[ThrowawayR2]

As the link notes, Blizzard complied with Chinese law regarding posting odds and posted a page on their Chinese Hearthstone site (http://hs.blizzard.cn/articles/20/9546) stating that, on average, 1 legendary card could be expected for every 20 packs (of 5 cards).

It is possible that Blizzard uses different odds by region but that seems unlikely.

[minimaxir]

Fair distinction; added edit.

It would not surprise me if odds were indeed different by region. It would be smart to do that, anyways, since player behavior may be different.

[gonehome]

People are pretty bad at intuitively understanding probability - it’d have to be stated in a way they’d understand.

“You’d probably have to buy this one thousand times to get what you’re looking for”

People still buy lottery tickets though because “there’s still a chance”.

Not sure how you fix that.

[godelski]

I can't seem to find the video, but I remember Sid Meier talking about how he had to manipulate the battle odds to what people thought was fair combat. For example, they thought that if they had a 2/3 chance of winning, they should win practically every time and thought it was unfair when they lost. Even though they won 2 out of every 3 matches, on average.

The average person doesn't understand expected outcome, that it means ON AVERAGE it will take this many times to obtain the thing. Lottery tickets are a bad example because the expected outcome is still really high, though most people still understand this.

[canadian_voter]

For example, they thought that if they had a 2/3 chance of winning, they should win practically every time and thought it was unfair when they lost

Like that time I lost my entire squad in X-Com because they all kept missing their 80% to 95% chances to hit. There was much rage that day.

[godelski]

I always hear this with X-Com, and have experienced it myself. Though it is really possible to miss 90%s several times in a row. It shouldn't happen often, but our monkey brains hate it and it feels unfair. It is a hard balancing problem.

[tzakrajs]

Why not have a system that is able to average out the rolls to happen in a way that good/bad rolls don't clump together and are spaced more evenly? Couldn't you generate the rolls and then sort them in favorable way or reroll whenever a roll is too far outside of some predicted thresholds?

[godelski]

There's a question of how you want to play with statistics. Do you want to play with fair dice or dice that give you an advantage? You might be saying to yourself that your storage tactic is still a uniform distribution, but you're forgetting sample size. You're also removing some uncertainty.

So, you can roll dice over and over, but you actually need a lot of samples for the stats to converge. That's why it is called the law of LARGE numbers. It is still possible to roll two 12s in a row on dice, but we wouldn't expect it to be common. By your clumping together, you need to not only store a lot of data, but now your "random" events are dependent rolls and not independent.

Really it is just a question of what you want to do and how you want fairness perceived. Do you want your game to act like dice? Or do you want a slight advantage? BTW, i2om3r linked the video I was talking about there Meier discusses peoples' perception of fairness. In the end, you have to determine what is best for your game. Maybe stacking the deck makes better gameplay, maybe it doesn't.

[eropple]

Sure. But that's not what X-COM was (or the new ones are, though they mess with the rolls a little at lower difficulties).

Sometimes you just roll snake eyes. Then you deal with it.

[hermitdev]

Pretty sure this was after Civ 1, when you'd lose a battleship to a spearman surprisingly often. Been a while since I've played Civ 2, don't think there was the concept of cumulative damage, yet, but there was by Civ 3. Definitely seems "fairer". Sure a spearman can possibly damage a battleship, but unlikely to destroy one.

Also, it's Sid Meier.

[godelski]

I can't remember which Civ it was, but I remember he was talking about an early one. I'd love to find that video again.

(Also, fixed my post. Thanks)

[QAPereo]

This is a huuuuuge issue in the world of informed consent too. When you tell someone the odds regarding a procedure, do they really understand those odds?

[Negative1]

Sid (and others) did much the same thing that F2P games do now; make the odds 'flexible' using accumulators, counters and other tricks. For example, the odds displayed may be 66% chance of victory, but over time, that number changes based on a number of fuzzy factors (number of consecutive victories/losses, time since last victory/loss etc...). He even had a random coin flip in there cause 'randomness is what makes games fun'. So even though it said 66%, there's something like a 1 out of 10 chance that you'll just lose, to spice things up. I mean really, how fun would it be if tanks always beat archers. :-)

In the pursuit of making a fun and interesting experience, it's not as bad as it sounds since it leads to a rewarding experience (though getting 2 hours of sleep because Civ is so addictive could be construed as negative). When providing questionable reward to the player and in a money grab via psychological manipulation, well, yeah, seems pretty exploitative and not so honorable.

What's most interesting to me is the wildly different perspectives that gamers on different platforms will tolerate. Zynga/Facebook gamers seemed to never care. Mobile gamers follow suit but some are a bit sensitive to this manipulation. Console gamers are familiar with these tactics since the horse armor days but have a red line (Destiny 2 just crossed). PC gamers have zero tolerance and will go up in arms as soon as their precious games are violated with F2P mechanics (though Ubisoft is doing pretty good here since they're doing it somewhat respectfully).

[i2om3r]

For reference, here is the video: https://youtu.be/MtzCLd93SyU?t=19m28s

[godelski]

THANK YOU!!!!! I've looked for that video several times and have been kicking myself for not bookmarking it.

[alasdair_]

The same thing happens with weather forecasts, where "95% chance of sun" gets downgraded to "80% chance of sun" because people assume 95% == a sure thing.

[glenstein]

I think that in addition to the problem of boosting interest by concealing probabilities, you have the added problem that probabilities are difficult to intuitively grasp, even if they're straight up given to you.

The same way that a person can have a correct (at least within a few orders of magnitude) understanding of their numerical odds of winning the lottery, yet still go on purchasing tickets. Our intuitive grasp of the idea of one in a million, billion, trillion, etc is such that those odds all "feel" the same. But those differences might make all the difference in whether purchasing a lottery ticket is rational.

[brianwawok]

How many people buy state lotto tickets?

What is the EV on those? Some it's something like .5 or less. HORRIBLE odds. If you want to gamble, go play $1000 hands of blackjack or something, you can get something like a .97 EV.

[wl]

The point of state lotto tickets is that a win is life changing and the price is negligible.

[bryanlarsen]

The EV on Canadian lottery odds sometimes exceeds 1. They're required to pay out 50% of intake, but if the jackpot isn't won a bunch of times in a row (with payout under 10%), the payout for a single draw sometimes exceeds the intake for that draw.

It's not profitable to buy all the tickets, though: that changes the odds enough that it becomes unprofitable.

[brianwawok]

I don't think the EV often goes over 1 though, because of the multiple winner situation. I.e. even if the jackpot is 150 million with 1:100 million odds to win... there is a chance there is 2 or more winners who SPLIT the jackpot.

There have been cases of people buying out all the numbers.. and there are times when it is worth the risk.. but not really relevant to my store. Millions of people buy lotto tickets every day, with an EV of .5.

[TheDong]

The probability of loot boxes for most games has been reverse engineered and is available online.

Making it available in the app is no different; the user would have to be rational and look for it.

A user that would be stopped by this can already go online and find the probability.

[QAPereo]

It’s a terrible trap for a smart person, to assume that x,y, or z is “something which is known to each and every schoolboy,” you know? Everyone isn’t like you.

[TheDong]

The point is that the apps will display it where they display Open-Source license notices: off in some deep menu no one but a suitably dedicated hunter would find.

The same sort of person that would google it.

[QAPereo]

Right, but most people wouldn’t even consider that. Most people don’t even know what “open source” means! It’s hard, but most people live in a combined state of technical ignorance, and a kind of aggression toward learning new things in that field. The result is that the average person thinks it’s all techie noise, so what you consider incredibly basic concepts are really overlaid on a framework of existing knowledge, intelligence, and most of all, interest.

[TheDong]

I think we're talking past each other and agree that people won't look for the probability or, even if they see it, won't care.

I am not assuming people look it up. I'm just pointing out this isn't really changing anything because people who would look at it or care, already could look at it often... but most people won't, so the change of requiring apps to show it in the app matters basically not at all.