What’s Inside Skinner’s Box?

By Shamus Posted Tuesday Oct 29, 2019

Filed under: Column 87 comments

So let’s say you’re a stone-age Homo sapiens and you’re having a really bad day. You’re separated from your tribe, you’re hungry, you’re thirsty, and you’re exhausted. Nobody’s invented McDonald’s yet, so you have to forage for your food. You don’t know what vitamin C is, but you do know that for days you’ve had a powerful craving for something tart.

Then you come across a plant you’ve never seen before. Fruit is growing on it. You don’t know if it’s safe to eat, but you don’t have much of a choice. You take a bite and Woooooooooooooah. It’s got water, it’s got calories, and it’s got vitamin C. That wave of euphoria you experienced when you ate the fruit was caused by the release of dopamine in your brain

Link (YouTube)

So Borderlands 3 is finally out. It’s gotten a mixed reception so far. Like I said in my last video, it doesn’t have the same charm that Borderlands 2 did, and a lot of that is because of the way the reward system works. It might be a great game, but it’s a lousy skinner box. I think most gamers know what a skinner box is at this point, but I rarely hear people talk about why Skinner Boxes work the way the do.

The concept of a Skinner Box comes from behaviorist B. F. Skinner. The skinner box is a real device – properly called an operant conditioning chamber – and it actually covered a lot of different stimuli, both positive and negative. Doc skinner made animals go hungry and he applied electric shocks to others. He was a giant in his field, but he was definitely no Dr. Dolittle. The actual skinner boxes were probably not a fun place to be, but in gaming we’ve borrowed the term to refer to a gameplay mechanic that delivers irregular rewards. The original skinner box was a scientific apparatus, but now we use the term to describe gameplay mechanics. What’s really going on is that we’re using an exploit in the learning systems of our brain to create pleasure. Here’s how it works:

Learning Has Diminishing Returns

Ohh yeah, that feels good. Well, for ME anyway.
Ohh yeah, that feels good. Well, for ME anyway.

Getting back to that stone-age Homo sapiens, that blast of dopamine isn’t just so our primitive human can feel better. Yes, dopamine feels good, but it’s also linked to the formation of memories. Let’s say you find another bush a few days later. As you savor your prize, you start to think, “Hey, the last time I found one of these, I was in a valley, in a sunny patch, near water, with some sand around. Just like this time. Maybe in the future I should seek out these sorts of places to see if I can find more of this fruit.”

The dopamine doesn’t just feel good, it also tickles the parts of your brain that look for patterns and build muscle memory. Basically, this is the brain’s way of saying, “Whatever you did to make that happen, try to do it again because that outcome was really good.”

Under normal circumstances, this works as intended. It doesn’t matter if you’re performing music, shooting hoops, telling jokes, solving a crossword, or learning to punch lawless thugs unconscious. Every success gives you a little kick of dopamine, allowing you to gradually fine-tune and hopefully perfect your performance.

And that’s the trick here. Whatever you’re learning, it needs to be something with a non-obvious solution. If the solution to a problem is straightforward, then your mind isn’t going to be attracted to it. Everyone is wired a little different, but most of us don’t get a thrill from pressing a win button.

We only get rewarded when we succeed at something we haven’t mastered yet. Or at least, that’s where you get the really big rewards. You might enjoy a nice hit of dopamine the first time you clear a challenge room in Batman without taking a single hit. You might cheer. Do a fist-pump. If there’s a friend around you might try to get a high-five out of them. Maybe you’ll instagram it, or call your mom. Whatever. I’m not here to judge how you celebrate. I’m just saying that it won’t be nearly as special when you do that for the tenth time. And when you do it for the one hundredth time, you won’t even change the expression on your face. The triumph has become routine, and you need to tackle a new challenge if you want another big dose of dopamine. Your brain is telling you that if you want more rewards, you need to push yourself to overcome something new.

But what if there was something where you could regularly enjoy success without ever mastering it? What if there was a task that you could never improve at matter how long you practiced, even though you could still succeed once in a while? A task like that could theoretically reward you forever.

The good news is, such an activity does indeed exist. The bad news is… it’s a slot machine.

The Slot Machine

Do you feel lucky? Yes? No? Well, it doesn't matter, because your feelings don't impact the behavior of the machine.
Do you feel lucky? Yes? No? Well, it doesn't matter, because your feelings don't impact the behavior of the machine.

Those spinning wheels and flashing lights in gambling devices aren’t just there to make the game more “fun”. They’re required for the game to work at all. If you had a slot machine where you dropped in your money and it instantly gave the result, then it would lose the ability to trigger the release of dopamine in most people. Your brain does that because it’s trying to get you to find a pattern, and the tactile, visual, and audio cues all exist to create a bunch of information for your brain to analyze. Sometimes you win, usually you lose, but when you do win your brain will still give you that sweet, sweet dopamine, hoping that you’ll crack the code. If there is no code to crack, then the reward will never stop.

The interesting thing is that all of this seems to happen on a subconscious level. You can tell a gambler that a slot machine is pure randomness and that the lights and sounds don’t mean anything. They can know and understand this idea, but that doesn’t stop them from getting more dopamine when they win.

What Doc Skinner discovered is that we – both animals and humans – will fixate on a system that gives irregular or unpredictable rewards. If you stick an animal in a box and give them a button to dispense food, then once they push the button a couple of times they’ll have it figured out and then they’ll only push it when they’re hungry. But if the reward is irregular – if it’s random whether or not they get food – then the animal will press the button a lot, even when they’re not hungry. The irregular behavior makes your brain fixate on something until you can solve the pattern. A simple food button wouldn’t work on a human, but if you add enough spinning wheels and flashing lights, the complexity might be enough to make them curious.

Some people say things like, “I don’t gamble because I understand math.” It’s true that the math says that betting against the house is a bad idea, but this isn’t really why people gamble. Plenty of people with gambling problems can explain the odds and the toll that gambling takes on their life. It’s less that they’re confused about the mechanics and more that they’re looking to justify doing something they already know is a bad idea. Our brains are really really good at rationalizing and coming up with reasons to reward ourselves right now.

“I’ll just have one drink and then I swear I’m going straight home,” says the alcoholic.

“I’ll double my workout tomorrow,” says the person who wants to cheat on their diet.

“I think my luck is about to change,” says the gambler who’s been losing all night.

“But it’s on sale!” says the compulsive shopper who’s already maxed out two credit cards.

Deep down, these people know they’re wrong. The person is rationalizing a short-term reward at the expense of long-term setbacks because they’re really craving the reward. That’s a pretty common problem for human beings to have.

Maybe all of this sounds insidious, but a skinner box is a system that can be used for good or evil. It can destroy people, but it’s also the mechanism that entertains us, helps us learn, and motivates us to pursue our long-term goals.

Diablo and Borderlands are both games that harness this built-in reward system to make a game more fun. Note that this works best when it’s added to an already-satisfying game. Ideally you’re playing a game where the gameplay is fun, the story is interesting, and the visuals are good. If you take an already-good game and add a skinner box on top of it, you wind up with the kind of game that can keep you engaged for months or weeks, and that you can come back to years later. If you don’t have that stuff – if the gameplay is shallow, the art is ugly, and the story is lazy nonsense – then you wind up with something that’s just a skinner box. A lot of mobile titles wind up here. It can still hook a few people that have a strong proclivity for skinner box traps, but the majority of people won’t give the game a second look.

Welcome to the Casino

I have seen the future of the games industry. It's a raytraced 4k 60FPS always-online Casino built around social media integration and pay-to-win microtransactions.
I have seen the future of the games industry. It's a raytraced 4k 60FPS always-online Casino built around social media integration and pay-to-win microtransactions.

Gamers are frustrated with current-gen loot boxes and other microtransaction-based reward systems because instead of working with the mechanics of the game, it works against it. Defenders of loot boxes claim that nobody is forcing you to buy them. While that’s true, it doesn’t repair the damage that loot boxes inflict on the design. I want a game where I can steadily earn things through gameplay, and sometimes I’ll get a little bonus reward of a rare drop or whatever. A loot box driven game stonewalls that steady progress in order to push me to buy spins on their slot machine. In order to sell more loot boxes, the steady rewards are placed so far apart that the game feels empty if you’re not spending money.

I don’t know what happens next for loot boxes. I know a lot of us hate them, but I also know that a very small number of people pour a ton of money into them. Consider the person who spent $62,000 on Runescape. If an 18 year old MMO can make that much from one person, imagine how much current-gen games are making from loot boxes. If a thousand people boycott a game because of loot boxes, it only takes one person spending $62,000 to make up for all of those lost sales. The AAA area of this hobby is pivoting away from making games and towards making video game branded casinos, and the result is not pretty.


From The Archives:

87 thoughts on “What’s Inside Skinner’s Box?

  1. PPX14 says:

    The AAA area of this hobby is pivoting away from making games and towards making video game branded casinos, and the result is not pretty.

    This is so sad :( And they own the popular licensed franchises like Star Wars. At least AA is still around. And supposedly there’s still appetite from people like EA for things such as Mirror’s Edge.

    1. Chris says:

      At least AA is still around.

      I don’t feel that way. I find most of it either falls into AAA studios or indie.The middle sized stuff didnt seem to survive 2008 and the newer graphics that require more people.
      Kickstarter seemed to allow medium sized teams to have a shot at making a game, but that seems to have dried up as well. 22cans had to retreat into phonegames, for example.

      1. PPX14 says:

        I know, I for some reason decided to make my comment slightly more optimistic thinking maybe I underestimate the size of the AA offering – Hellblade, Overload, Echo… Edit: I also got Outcast Second Contact, which is fascinating that such a game got a remaster.

        To me it seems that AA is basically what a lot of or even most of the mainstream (e.g. LucasArts) games used to be. When I first heard the term I thought it redundant – that’s just games isn’t it? I guess AAA is the new thing, like you say, 2008.

        1. Mattias42 says:

          Honestly, I’d argued we’ve recently had a bit of a resurgence of the AA level games. Like DONTNOD’s Vampyr & Life Is Strange, Frogware’s Sinking City & Sherlock games, Hare-Brained Schemes Battletech, Deck 13’s Surge, and Spyder’s Greedfall, just for some semi-recent examples off the top of my head.

          Heck, Medievil of all games got a remaster 20 years (!) after initial release. Never thought I’d see that day come again.

          It’s definitely a tightrope it’s getting rare to see walked, though. Nowhere the bread & butter level it was back in the PS1 & PS2 era. I’ll grant that much.

          1. tmtvl says:

            Middle market is reasonably big. Depends of course where you draw the boundaries between indie and MM.

      2. John says:

        22Cans is Peter Molyneux’s outfit. While they, like many developers, have undoubtedly struggled with changes in the market place, I suspect that their primary problem is that they’re Peter Molyneux’s outfit.

      3. Bwah? I play an incredibly narrow selection of games and I can still list several AA studios off the top of my head:

        InXile (recently acquired by Microsoft)
        Obsidian (recently acquired by Microsoft)
        Larian Studios
        THQ Nordic (recently acquired Piranha Bytes)
        Crate Entertainment (makers of Grim Dawn)
        Deep Silver (makers of Outward)

        They make some weird experimental stuff but they put out a lot of good games.

        1. Asdasd says:

          If you accept acquisitions as a form of industry consolidation, those AA studios who are acquired effectively stop counting as independent entities in much the same way as if they’d gone out of business, being subsumed into their (usually AAA) purchaser. That said, I do agree with you that is still a fair bit of AA talent about, just not as much as in previous some of the expansion-consolidation cycles we’ve seen the industry go through (such as the early/mid 2000s, where the PC and PS2 in particular were awash with AA games.)

    2. Higher_Peanut says:

      If you get lucky you can get a “totally not a sequel” to a franchise like Bloodstained was to Castlevania or a smaller game for a niche/dead genre. If it’s a big franchise though I wouldn’t expect anything other than cash shops and chasing trends with re-skins (like Star Wars) until it becomes unprofitable.

      I bet there’s a ton of appetite for good games tied up in EA and other publishers/developers. Unfortunately none of them will be in a position to contest corporate’s accounting decisions. They’re raking in too much money to listen now.

    3. Galad says:

      Eh, not like we are done with our current backlogs. Hell, I can probably last an year only on Neverwinter Nights EE mod campaigns alone, nevermind the rest of my library that I have not finished

  2. Lasius says:

    It’s “Homo sapiens”.

    1. Higher_Peanut says:

      If we want to get really accurate it’s Homo sapiens.

      1. Lasius says:

        Absolutely correct, but I didn’t have time to figure out how to write cursive here.

        1. Higher_Peanut says:

          Fair enough, I only know the instructions live below all the posts since I needed to spoiler something one time.

      2. pseudonym says:

        Yes! For people who do not know this but like to know: official species names consist of a genus (Homo in this case) and a species name (sapiens). The genus must start with a capital letter. The entire name must be written in italics. If italics are not possible for whatever reason, the entire name must be underlined.

        These rules do not apply to common names such as daffodil and dandelion.

        Another rule is that the name in the first description of the species always takes priority. This means that we are alway stuck with the moniker “thinking man” even though Desmond Morris’s description “the naked ape” focuses on a more distinct feature of our biology.

  3. Ander says:

    At the end I was hoping you’d talk about how loot boxes work as Skinner boxes and how that’s different than an in-game loot system as in Borderlands 3. I say this and the answer may be intuitive to people (“it’s taking people’s money instead of time and skill” for example), but I think it would have made the loot box mention feel more cohesive.

  4. Abnaxis says:

    If I’m 100% honest with myself, I only really have no problem with Overwatch lootboxes (I still don’t understand why these are the go to example of lootboxes) because they’re giving me something for free that I’d never buy (skins) otherwise. Maybe it’s cold of me, but I have gotten so disengaged with the various monetization schemes companies put out that I don’t care what they are as long as I don’t have to participate to have fun.

    1. Chris says:

      Because they are a clear example of after market monetization for a game that previously would just have been a box price. Either the skins wouldn’t have been there or the would be rewarded for special things. Like how you had to unlock characters in time splitters. Same reason horse armor became the face of bad DLC.

      Previously the game box buying player was the target audience, which is good for me since im part of that audience. Now its the whales, which I’m not part of, but it still affects me.

      1. Higher_Peanut says:

        They also serve as in game advertising for the monetisation. “Look at all this fun stuff you got, wouldn’t it be great if you could get the one you wanted without RNG? You’re really missing out.”

      2. Philadelphus says:

        Because they are a clear example of after market monetization for a game that previously would just have been a box price.

        Or, they wouldn’t have existed at all? Look at Team Fortress 2 when it came out for a single purchase price—no skins, not even any alternate weapons. Sure, those things got added later on, but Valve thought that what they sold was a complete game (and plenty of buyers apparently agreed).

        This comes up a lot in discussions of Paradox Interactive’s current DLC policy, where people seem to assume that anything that’s developed and released as DLC after a game came out could’ve magically been part of the same game they bought four years ago without any increase in the unit price.

    2. Echo Tango says:

      That’s one of Shamus’ later points – the games would be less grindy and more fun, if they weren’t trying to force people to buy things. Even if you’re stonewalling the company, you’re affected.

      1. RandomInternetCommenter says:

        But… There’s no grind in Overwatch? You have access to all your abilities the moment you launch the game. Unless something has changed (I only played the game during their free weekends, and had a grand time).

        1. Lino says:

          There is a grind if you want a specific skin or emote, or whatever. Back in the day, all the skins were available for everybody right from the start.

          1. Abnaxis says:

            I mean, if you really want to go back, back in the day the sounds were just achievements and you didn’t ever have to download anything after you bought the damn thing. The only time anything was updated was when they released expansion packs.

            1. Echo Tango says:

              Back in the actual day, achievements were for goals that you set for yourself, without the Skinner Box manipulation. You know, because games were for fun and enjoyment, not just manipulating players into longer play-times.

              1. Lino says:

                Also, back in the day, 60 dollar-games weren’t released in a broken, unfinished state. The fact that patching was difficult made devs more careful about how their games looked at release (also, the lack of patches didn’t allow them to add scummy microtransactions and lootboxes after release).

                1. Syal says:

                  Also, back in the day, 60 dollar-games weren’t released in a broken, unfinished state.

                  If you want to prove your point, solve my maze!

                2. Ninety-Three says:

                  Also, back in the day, 60 dollar-games weren’t released in a broken, unfinished state.

                  I too am a fan of such polished classics as X-COM, Shandalar, Bloodlines, and the entire Obsidian catalog.

  5. Narkis says:

    I don’t know what happens next for loot boxes.

    Legislation happens. Politicians and governments are slowly waking up to the digital casinos big publishers are making. There are going to be laws against that, they’re going to be heavyhanded and with unintended consequences (TCGs caught in the crossfire, probably), and I really doubt lootboxes will survive the governmental Eye of Sauron turning its gaze on them.

    1. Karma The Alligator says:

      One can hope. I know there’s a push for it, but it’s not happening fast enough, I think.

    2. galacticplumber says:

      Card games can function perfectly fine without randomization. You just offer the customer set cards or sets of cards for set prices. Without buying lots of useless tat the customer probably even spends less money while still being viable.

      You can still have the fun of booster pack tourneys by simply stating the entrance fee is for participation without necessarily keeping the cards.

      This whole concept was always a scam. It’s just more obvious when shown by people with less subtlety in a more information connected age.

      1. Ninety-Three says:

        What happens in Magic: The Gathering is that larger game stores will buy and open thousands of booster packs themselves, enough that the law of large numbers kicks in and it isn’t really random any more. Then they sell the individual cards they opened to players at straightforward prices determined by supply (card rarity) and demand (card power level). This is how most of the competitive players get their cards. The end result is exactly the system you describe, only with middlemen like StarCityGames taking a cut.

      2. Echo Tango says:

        Not only could the games function without the randomized bullshit, but they’d arguably be better experience for the players.

        1. galacticplumber says:

          Hell yes. Let’s illustrate the difference by showing both practices alive and well in one game.

          In Yugioh Duel links the primary methods of attaining cards are buying EX structure decks outright with in game currency and money, as well as buying packs from various sets with gems or money.

          While you are forced to spend some actual money on EX structure decks to get a full set of three of some of their more valuable cards, it’s like eleven bucks a box and a day or two of gems worth of grinding every few months. Cheap simple, and done.

          In box sets an individual pack is worth several minutes of grinding, or a dollar assuming no sales are in effect. Some of the rarest and most powerful cards only have one copy in one hundred and eighty packs per box. You can reset the box to full at any time for free, but that also puts back in copies of all the cards you opened but didn’t want.

          I’m sure you don’t need help doing the map for the effort to get a full set of three of a good card. Competitive decks will commonly have like six to twelve of these rare cards.

  6. John says:

    I am the kind of person who says “I don’t gamble because I understand math.” I may have even used those exact words before. I know that the odds are against me and I hate hate hate losing money, so I don’t do it. So, yes, I am possibly an ass of some kind. But I have never believed that people who do gamble are stupid or that they do it because they don’t understand math. I have always understood that people gamble for the thrill of it and as long as people gamble responsibly I don’t see the harm.

    When I was a kid, my uncle kept a novelty miniature nickel slot machine on the bar in his den along with a small bowl of nickels. I would never, ever sink my own money into a slot machine, but I loved putting my uncle’s money into that machine and watching the tumblers spin. I got excited whenever the machine spit out a few nickels, even though I knew that I’d have to put them back in the bowl and wouldn’t get to keep them.

    I don’t play novelty nickel slots any more, but I do play games with random mechanics. Each roll of the dice in Dungeons and Dragons, each shot fired in XCOM is a gamble. I get a bit of that sweet, sweet dopamine every time I save versus death or take down a muton. I suppose I do gamble then. I just don’t gamble for money.

    1. Hector says:

      I don’t think all random mechanics are gambling or related to Skinner boxes. It sounds like you are actively strategizing. That’s a calculation and risk assessment game, which is very different from randomized rewards.

      1. John says:

        Fair enough. I wasn’t really thinking about Skinner boxes when I wrote that. XCOM is definitely gambling though. It’s true that there’s strategy involved, but I could say the same of games like blackjack. The difference between XCOM and casino games is that there’s no money on the line and the game favors the player rather than the house.

        1. baud says:

          One things that’s cool with XCOM is that you can stack the deck in your favor by playing well, knowing what weapons/tools/class to use (grenades to remove cover, placement, status effects, skills), which enemy to prioritize, research/build order, building placement and so on. For me the most satisfying is when I’ve stacked the deck enough in my favor results won’t depends on the roll of a dice.

      2. Sleeping Dragon says:

        I don’t know if it’s directly Skinner box related in the academic sense but it is a similar mechanic where your brain encourages risk taking by rewarding risky successes. Anything I could say on the topic will be a gross oversimplification (particularly as there are libraries of research in the field) so let’s just leave it at stating that the reward system of our brains is somewhat skewed towards risk taking*, which factors into the whole gambling thing.

        1. Sleeping Dragon says:

          Forgot to the delete the * when I removed a footnote that took a detour into risk mitigation and learning through empathy and (indirect) information transfer.

          1. Hector says:

            I think there’s a big difference between trying to understand the system in order to manipulate it in your favor and being “programmed” by randomization of rewards. To use the X-Com example above, the point isn’t to “take risks” but to optimize your chance of success by managing risk. It’s unlike Skinner box mechanics in tat the outcome isn’t truly random, and in fact you may be able to guarantee success in the long-term.

            1. Sleeping Dragon says:

              True, but even if you don’t YOLO sometimes you take that 20% shot (if, say, you have no other options) and if it lands it tickles that part of your brain.

              Edit: Mind you, I’m not saying there’s anything sinister about it. A big part of entertainment is to learn to satisfy those “animal” (again, gross oversimplification) impulses in a controlled way without exposing yourself to actual dangers. There is nothing particularly wrong with Skinner boxes either, problems start when they’re designed and deployed to exploit.

        2. John says:

          That’s interesting, given that the conclusion of a lot of economics experiments is that people are risk averse in the sense that they often will not take even actuarially fair gambles. A fifty percent chance of losing, say, $100 more than outweighs a fifty percent chance of winning $100 in their private risk-reward calculations.

          1. Kyle Haight says:

            At first blush that seems reasonable to me given the principle of diminishing marginal utility. If you look along the range from $0 to $200, each additional dollar has less utility. That means the utility gain in going from $0 to $100 is greater than the utility gain in going from $100 to $200. Turn that around and it means that an equal chance of gaining or losing $100 will statistically lead to a loss of utility, so the bet should be rejected.

            1. John says:

              That’s the standard explanation, certainly. Other experiments suggest that people aren’t always expected utility maximizers in the Von Neumann-Morgenstern sense, however. Sleeping Dragon is correct when he says that the way in which gambles are presented can sometimes affect whether or not people are willing to accept them.

          2. Sleeping Dragon says:

            Ah, but I wasn’t refering to being risk prone or risk averse in general, just that the reward system is skewed towards risk taking, we have plenty of other systems in the brain and they often run counter to one another. To address your specific example, I’m not going to argue against your claim that most people would be unwilling to take that gamble (though I imagine this strongly depends on things like their means or circumstances in which the gamble was presented), I am going to argue that those people who would would experience an intense reaction from the brain’s reward system if they won. More intense than the positive reinforcement of “I’m feeling smart because I didn’t take the risk” or even “I found a 100$ on the street”. Also note that thanks to empathy we can enjoy the thrill of risk taking by observing someone else doing it (literally the entire formula of Let’s Make a Deal and several other shows) which coupled with our ability to transfer information promotes a generally risk averse populace with a subset of risk takers (the term “adrenaline junkie” comes to mind) who “teach us valuable lessons”.

            Anyway, my point isn’t to arrive at some major conclusion here, behavioral studies are a hard field in part because all those systems are interconnected, it’s very east to muddle or misinterpret results and it always has that vibe of “self-diagnosis” because the researchers are also human.

          3. Phill says:

            People are moss loss-averse than risk-averse (and tend to be over-optimistic about probabilities – they think an 80% chance of success should go in their favour 99% of the time, and a 20% chance of success is roughly a 50-50 proposition, but that’s another story).

            Offer someone a choice between A) a guaranteed $94, or B) a 95% chance of $100 and 5% chance of $0. Choosing B would on average net a slightly higher average reward ($94 vs $95), but most people will choose A, the worse option *on average*, because the extra value of $100 vs $94 doesn’t outweigh the emotional downside of the chance of ending up with nothing.

            Conversely, offer someone a choice between C) a guaranteed *loss* of $94 and D) a 95% chance of losing $100 and 5% chance of losing nothing – an expectation value choice between C: $-94 and D: $-95, and most people will chose D, again the worse average option, because of the chance to avoid losing anything at all – they want to avoid the $94 loss.

            (You can change it to be C: guaranteed 6$ win vs D: 95% chance of $0 and 5% chance of $100 if you prefer, with the same result).

            The key is loss-aversison. People are prepared to take risks on gaining things, but are very risk-averse with the possibility of losing things.

    2. BlueHorus says:

      I am the kind of person who says “I don’t gamble because I understand math.” I may have even used those exact words before. I know that the odds are against me and I hate hate hate losing money, so I don’t do it.

      I’m kind of in the same position, though it’s not that I’m good at maths. (Though you don’t have to be good at maths to appreciate the terrible odds).
      I have gambled, and I have won, but winning didn’t feel good. Like something doesn’t work in my brain and I don’t get the dopamine. Losing is still annoying, though, so the net result was not good.

      Ironically, it was playing a computer game that helped teach me this; specifically Fallout: New Vegas. I cheated my way to a maxed-out Luck stat, went to the in-game casinos…and STILL managed to lose at Roulette every time.
      Helped reinforce just how stacked the odds are (in the House’s favour).

      1. Scampi says:

        Actually, the odds in Roulette are not THAT stacked in the house’s favor. Over the long term (I believe I remember this correctly), you lose 1/37 of your wages (~2,7%) on average, irrespective of your strategy, which is not too much, I guess. Still, there is not really any way you can trick your way to better odds, so it’s probably better not to play unless you rigged the table yourself.
        I read about those specific odds years ago in a gaming magazine (tabletop, not video) and since that day, I never forgot about it.

        1. John says:

          You can beat the house (or at least come close) if you’re a math professor with a system that the house hasn’t caught on to yet. For example, you can always try the following. Place a bet. If you win, walk away. If you lose, try again with double your original bet. If you win, walk away. If you lose, try again with double your previous bet. And so on. The trick is to walk away as soon as one of your bets pays off. The odds that you come out ahead obviously depend on the game you’re playing and the depth of your pockets but they’re relatively good, especially in a game that’s close to fair. There’s still a non-zero probability that you’ll run out of money of course.

          The problem with this and similar strategies is that casinos know about them and will kick you out if they catch you doing them. Video game casinos not so much. I used to do this sort of thing in the elven casino in Majesty. Never did go bankrupt.

          1. Warlockofoz says:

            Martingale strategy. A small steady gain balanced by a remote chance of massive losses. Feels good until that remote chance comes up, which eventually it will, and it doesn’t change that the odds favour the house.

            What can work in more complex betting games is tracking the game state closely (eg. Counting cards, analysing a shuffling mechanism, finding and exploiting other imperfections in randomisation) and making large bets while the state is temporarily in favour of the player.

          2. Scampi says:

            Actually, strategies like this would work if you were regularly successful on low wager levels and kept restarting from the bottom.
            What prevents them is the betting limitation that doesn’t allow bets above a certain limit. Many casinos apparently have them specifically to prevent this kind of betting strategy.
            Considering you lose the same amount of your bet on average per game, it will, again, not help you in the long run. After all, you don’t get to chose which games you will win.

            1. The Puzzler says:

              What prevents the strategy from working is that even if you’re allowed unlimited bets, sooner or later someone who tries this will have a really bad run of luck and lose everything. If it’s not restricted by the table limit, it’s restricted by the wallet of the gambler.

              It’s changing the game from an almost 50% chance of double-or-nothing to a game where you have a large chance of a small gain and a small chance of a large loss. The average loss per spin is still the same.

              How many times would you play a game where you have a 99.9% chance of winning $10 and a 0.1% chance of losing $11,000?

          3. Ninety-Three says:

            No no no, that’s all wrong. Your expected value doing that strategy is still negative (assuming the house has an edge to start), and casinos love gamblers who think they can beat the house. There is no honest system that beats the house, the closest you can come is counting cards so that you’re no longer making random bets.

            Consider a simple game in which you wager $X and then the casino flips a fair coin: if it comes up heads you get $2X, tails you get nothing. If we show up and place ten $10 bets, it’s clear that each bet has a 50% chance of winning us $20, 10 bets times 0.5 times $20 = $100 expected payout from bets that cost $100 to make and we have profited precisely $0 in expectation.

            Now let’s try John’s strategy: we have $63 in our pocket, start with a $1 bet, and double-or-nothing until we win or go bankrupt.
            A. There’s a 50% chance our first bet wins, we’re up $1.
            B. There’s a 50%*50%=25% chance our first bet loses and our second bet wins, we’re up $1 (lost $1 in first bet, gained $2 in second)
            C. There’s a 50%*50%*50%=12.5% chance our first two bets lose and our third bet wins, we’re up $1 (lost $1 in first, $2 in second, gained $4 in third)
            D. There’s a 50%*50%*50%*50%=6.25% chance our first three bets lose and our fourth wins, we’re up $1 (lost $1, then $2, then $4, then gained $8)
            E. There’s a 50%*50%*50%*50%*50% = 3.125% chance our first four bets lose and our fourth wins, we’re up $1 (lost $1, $2, $4, $8, then gained $16)
            F. There’s a 50%*50%*50%*50%*50%*50% = 1.5625% chance our first five bets lose and our fifth wins, we’re up $1 (lost $1, $2, $4, $8, $16, then gained $32)
            G. There’s a 50%*50%*50%*50%*50%*50% = 1.5625% chance our first six bets lose, in which case we’re out of money and walk away bankrupt, down $63.

            Adding together A through F, we have a 98.4375% chance to gain $1 and a 1.5625% chance to lose $63. Or to put it in fractions, we have a 63/64 chance to gain $1 and a 1/64 chance to lose $63. Expected profit: $0. Play a game where the house has an edge and your expected profit will be negative.

            1. John says:

              Yeah, that’s so. I was way too positive about the Martingale. Fortunately, I’ve never been foolish enough to put my money where mouth inadvertently went. To the extent that I have a defense, it’s that I did note the non-zero probability that a budget-constrained gambler would eventually go broke.

          4. Chad Miller says:

            The problem with this and similar strategies is that casinos know about them and will kick you out if they catch you doing them.

            That’s not the problem, and no they won’t. The Martingale is negative expectation if there is a limit on how much you can bet, either because the casino has a maximum bet or because you don’t have an infinite bankroll, which is to say, it’s negative expectation in any real-world application. The negative edge per bet is the same as with any other strategy, it’s just that you end up with the negative results clustered into the occasional very expensive unlucky run.

            If casinos didn’t have a maximum on what you can bet, the result is essentially that you win small amounts until you go broke all at once, which the odds say you will.


        2. Chad Miller says:

          Over the long term (I believe I remember this correctly), you lose 1/37 of your wages (~2,7%) on average, irrespective of your strategy

          It depends on the specific wheel and rules being used. Most US casinos use a double-zero wheel where the house edge is twice as much. https://wizardofvegas.com/guides/roulette-survey/

  7. Hector says:

    Skinner boxes quickly come to bore me. I’m not claiming to just be too smart for such trickery or anything, just that I find such things really irritating in a fundamental way. Some randomness can be part if the fun, sure, but raw Skinner mechanics get old fast.

    I’ve been concerned in recent years that studio after studio is chasing lootboxes to the detriment of the actual games, and end up wishing there was the non-whale edition. I.e., let me earn rewards in a satisfying manner rather. Then I’ll buy the game. Instead, I see franchise after franchise introduce more paid and random features that actually cut against the game itself. In some cases, this cripples a potentially good experience. See Fallout 76 or Ghost Recon.

    1. Gethsemani says:

      You are not alone. What’s often overlooked when discussing Loot Boxes/Skinner Box mechanics is that all humans exist on a sort of sliding scale of dopamine sensitivity, average dopamine levels and dopamine release intensity. Some people just don’t get the intense high that others experience from engaging with a Skinner box, while others get it so intensely that it trumps some base drives like socialization and sex. Depressed persons often engaging more with Skinner boxes, supposedly because they suffer from low dopamine levels and that short hit is the closest they get to normal with the lowest amount of effort. Persons in manic episodes or intense euphoria also tend to engage more with Skinner boxes, but for the reason that every dopamine hit keeps that buzz going (and in the case of people in manic episodes, the total assuredness that they’re just one click away from the grand prize).

      For people not suffering from mental illness, a Skinner box can go either way. Some of us are just less receptive to the buzz of dopamine while others are more. For most people the effect of the dopamine hit when engaging with a Skinner Box is heavily dependent on their mood at the time. If you’re feeling down or a bit low, a Skinner box can feel pretty inviting as it gives out quick dopamine hits with minimal effort, while someone who’s already pretty content (or does something else that offers high dopamine release, like free range climbing or firefighting) might just feel it is generally a waste of time.

      Despite the media presentation, Skinner boxes are not heavily addictive instant gratification for most people. We all get some rewards from them, but there’s a reason why only 0.01% or so of all people who play games with loot boxes are considered “whales”. For most others, it is not a big enough dopamine hit to justify the cost of continuous purchases, if any at all.

      1. Philadelphus says:

        Interesting explanation, thanks. I was starting to wonder if I was just wired abnormally* when slot machines and loot boxes just don’t seem to provide, well, any dopamine for me.

        *Could still be the case! Wouldn’t surprise me.

        1. Hector says:

          Well, there are people who get addicted to gambling, but not everyone does. So its probably more of a spectrum to begin with.

  8. Benden says:

    I enjoyed the video and the words! But I’m dying of suspense here because for weeks I’ve been trying to figure out what is making BL3 less good for you, Shamus, than its predecessor(s). I understand from your prior BL3 posts that it has something to do with loot vs progression (getting guns powerful or strategic enough to make the game feel good). Is the specific thing that is causing that problem the thing you’re talking about in this video?

    In this video I noticed a huge difference in the BL3 footage near the beginning of the video and near the end. The front footage looks like BL2 at its best to me. You’re taking out trivial shielded enemies in 2-3 shots, occasionally switching to a gun that seems to pump out 20 shots/second to handle some other mooks, and encountering occasional baddies that need a half-dozen power shots or more time under the bullet shower to go down, but they all feel fast and hectic and you’re very mobile. The footage near the end of the video feels like you’re encountering bullet sponges that happen also to be wearing fun dampeners.

    Is this difference the difference between engaging with some kind of pay-for-loot system, and not engaging with it? (I assume it has to do with the shift codes.)

    I’m trying to puzzle out the big picture—I understand that the loot drops don’t give you what you need, but I’m trying to understand why mechanically that is so; what is the difference between your experience and that of someone who says the guns are great?

    I’m trying to figure this out because I wouldn’t have ever tried BL1, 2, and whatever that other thing was if it wasn’t for your largely positive (even through the mists of cynicism and critique that are a beloved part of the character of this site) discussion of them here. I’m still trying to decide if I should be playing this game now alongside the zeitgeist, or if I should come to it years from now after all the tinkering is done, or if I should ignore it entirely. And understanding what makes the difference between the two gameplay footages in this video would help me a lot…but your video/transcript only draw some arrows, and do not quite name the problem, so I’m unsure if pay-to-succeed is the whole problem, some of the problem, the worst part of a more complex problem…?

  9. Kamica says:

    Shamus savagely calling people out =P.

  10. CloverMan-88 says:

    A man spent 150.000$ on a Transformers mobile game. That’s how much current games make from lootboxes.

    1. baud says:

      It’s how most pay2win games make their dosh: loads of freeloaders supported by a handful of high-spending whales. And it’s skewing their gameplay towards reeling in as many whales as possibles, with all the tactics you see in such games.

  11. Paul Spooner says:

    I notice you never answer the question. Is it just too terrifying? To admit, “It’s me, and I don’t want to come out.”

    1. Hector says:

      Boo, Paul. Booooo!

  12. Sven says:

    This was very interesting, thank you.

    In the early days of mobile gaming, long before micro-transactions and loot boxes, I was a big player of Bejeweled. I played it on PC and PocketPC, and initially, I would usually play the normal mode. I was trying very hard to figure out what the pattern was that decided when you lose. You lose when there are no more moves, but is there something you can do to prevent running out of moves? Can you plan your moves ahead so it becomes impossible to run out? Does the game reward you with more moves for certain things, such as using special gems? Conversely, does it punish you for mistakes?

    After quite a while, I came to the conclusion that there was no pattern. The new gems you got were entirely RNG, so whether or not you’d have any more moves was also entirely random. The only thing you could do that somewhat affected the outcome was saving up the best special gem (I forget it was called) since it could be swapped with anything, so it provided an “out” if you did run out of moves. But ultimately, the game decided when you would lose, not the player.

    When I realized that, I immediately stopped playing normal mode. From then on, I would only play lightning mode, since there the game stacked the deck so you would never run out of moves, and you score was determined by your speed, not the RNG gods. I guess it still had a little bit of randomness, since you were dependent on getting “add time” gems, but that never felt as unfair as the arbitrary “no more moves” of the normal mode, and at least felt more skill based.

    I haven’t played Bejeweled in many years, but what Shamus was saying about gamblers and slot machines made me think of this. Once I knew that there was no pattern to find, I stopped playing. For the same reason slot machines don’t hold any interest for me: I know it’s not skill based, so I get no reward from winning (other than the actual monetary reward, obviously).

    I guess that’s why I’m not a gambling addict. :)

    1. SAD1 says:

      The Version of Bejeweled on my phone currently has both a “Classic” Mode and “Zen” Mode. Classic is RNG, Zen is Never-Lose. There is an option in Classic to buy a re-shuffle when you lose, even though you could just play the Zen Mode instead.

      – Face-Palm!

    2. Geebs says:

      I don’t really think PopCap games ever had actual gameplay. Peggle is another example.

      1. Ninety-Three says:

        Hey, PopCap made Plants vs Zombies which is definitely an actual game.

  13. Asdasd says:

    “f a thousand people boycott a game because of loot boxes, it only takes one person spending $62,000 to make up for all of those lost sales.”

    This almost makes it sound like studios are turning their backs on steady, small-stakes income from the regular audience in order to gamble on hitting a jackpot with the whales. Could… could game development itself be a giant Skinner box for them?

    1. Karma The Alligator says:

      I don’t think it’s that much of a gamble for companies. They know a lot of players will spend some money (not necessarily the 62k, but enough to offset a few non-payers).

      1. The Puzzler says:

        I’ve worked on a couple of free-to-play games.

        You’re competing with a ridiculous number of rival games, including all the big, popular FTPs with much higher budgets, so you basically have a small chance of becoming popular and fashionable and hugely profitable, and a large chance of losing a fortune. Basically you compare the advertising spend to attract each customer with the average spending by each customer (minus Apple’s share). If the second number is bigger, you keep advertising more and more and rake in the cash; if the first number is bigger, you give up.

        But for some reason, investors threw money at us to keep trying. It’s almost as if they got a thrill from making these speculative investments. If only I could think of a metaphor for that…

  14. Taxi says:

    The problem I have with microtransactions in general: video games are escapism. I want to run off to the fantasy world and do stuff I can’t do irl.

    Think about it. You can pay a single fee of 10 or 30 or 60 $/£/€ and be a pilot or an army general or a star hockey player or a troll slayer for hours and hours.

    But microtransactions break that. Suddenly I need to use real world money, which pulls me right back into the crappy real life where I’m broke and poor and miserable. I can no longer just save up X amount – the fun I’ll have is directly connected to the amount od money I have irl – completely breaking the escapism concept!

    And lootboxes, that’s just… Evil.

    I’ve not played any modern AAA game in a while and this isn’t the sole reason, it certainly isn’t helping. I surely won’t pay for a game just to pay even more.

    1. Taxi says:

      At the same time tho, mind you that obstacles have been in videogames since conception.

      Arcade games were deliberately difficult; then home games were also either too difficult or too grindy in order to artificially prolong the game length.

      Because of course if the game was too short it wasn’t worth it – but at the same time if it was too grindy, that was bad too.

      In a way, microtransactions can solve the problem. Let’s say you can choose to buy a new game for 30 but it will have too much grinding; or a faster leveling version for the regular 60. That, in a way, would make sense even tho it sounds stupid if you read it. Still better than sell a barebone base game for 60, regular version for 100 and with version constantly pulling change out of your pocket.

      1. Higher_Peanut says:

        Previously if you leveled too slow you solved it with cheats/mods. What we have now is games made extraordinarily grindy in order to sell progress to you on top of the lootboxes. Publishers are incentivised to stall out progress in the game the moment you get hooked.

        Paying for less grind is awful, if the game is so bad you will pay to skip it why was it worth paying for in the first place? Microtransactions won’t solve this problem, they’ll create it and sell what you used to have back to you.

      2. shoeboxjeddy says:

        Something like Mario Run or Pokemon Picross is cynical but not immoral. You can play for free for a bit, but we’re gonna make it unpleasant. Pay us, and it’ll be a good game. Unfortunately, that’s not the business model. It’s “Pay full price or even Collector’s Edition price, then keep paying us, FOREVER.”

    2. Asdasd says:

      I never saw Bioware the same way after they sent an NPC into my camp to try and hawk me some DLC.

      1. Karma The Alligator says:

        That’s one reason to wait until you can get a “definitive edition” or something that includes all the DLC, so this doesn’t happen.

  15. Kyle Haight says:

    Non-cosmetic microtransactions remind me of an old line from the tv show Blake’s 7, in which one character’s idea of diplomacy is characterized as “breaking your leg and then saying ‘here, lean on me’.” The game is a system designed to pull me in, get me hooked, make me unhappy and then get me to pay to make them stop making me unhappy for a while. That isn’t fun, at least not to me. So if I see a ‘game’ that looks like it follows that design, I avoid it like the plague.

  16. Alecw says:

    I think it’s a combination of the skinner effect plus a feeling of progression and control that actually improves your ability to get the win states.
    This is why Diablo was so good – you had skinner loot box in full effect, but levelling and gearing made you tangibly better at getting to the goodies.

    1. BlueHorus says:

      Yep. A Skinner Box alone is not enough – as a background to enjoyable gameplay, or some other progression, it works well. On its own…not so much.

      I’ve played Diablo II a bunch, but every time, I lose interest after level 30, where you reach the top of your character’s skill trees and beat the final boss. After that your only option is [the equivalent of New Game+].
      And as soon as I’m not gaining new abilites or going to new places, suddenly I’m bored, despite the fact the loot is better than ever.

      1. Scampi says:

        Well, that’s where knowledge of lategame items came into play for me. Once I found out there were some amazing items and secrets later that would grant me new powers and open up new places (Uber-Tristram etc), I wanted to see that stuff and in some cases, play some of the most boring character classes to lvl 80 just to get some amazing item that made it worth the investment in my mind.
        I do see your point, though. I feel the same way about more recent games in the genre. Diablo II and to a reduced degree, Titan Quest, were the last games in memory to tickle that fancy for me.
        I tried Grim Dawn, but whenever I reach a certain point, the game starts to bore me.

        1. The Wind King says:

          I always felt Grim Dawn’s problem was in Item Design / Balance, some of the unique items are so good they’re almost unskippable (a lvl 50 crossbow that increases “ALL” skills by +2 lvls springs to mind) and that once you reach a certain point in that game the number of enemies, particl effects, and sheer visual noise just becomes unbearable.

          Great for the first difficulty level, and for most of the second once you’re feeling POWERFUL, but it loses it’s charm in latter stretch of the second difficulty level

  17. General Karthos says:

    Speaking as an alcoholic myself, this is accurate. I have, in the past, told myself that I’ll have one drink and go to bed, only to find myself, hours later, drunk out of my mind, wondering how this happened AGAIN. In the last few years, since gaining sobriety (and fumbling it a few times as well), I’ve learned that, to quote a popular phrase “one drink is too many and a hundred is not enough”. Anyway, there are similar addictive behaviors that I might be worried about. If I go to a casino, I bring only cash. I leave all my credit and debit cards at home. If my friends want to stop off for a snack while we’re on the way to/from the casino, they’re going to have to pay for them, because I’m almost certainly going to lose my money at the casino, and it’s all a question of how much I want to lose. I’ve never smoked or done marijuana (despite the fact that it is recreationally legal now) because I worry that I wouldn’t be able to stop. The same applies to prescription painkillers. I tough it out through surgery on over-the-counter painkillers that aren’t addictive. (Thankfully, I’ve only had the one surgery thus far, to remove my wisdom teeth.)

    I hate, hate, hate loot boxes/random loot in games. Because I am tempted to do it again if it doesn’t reward me the way I want it to. My friend and roommate in college had it worse. When he knew loot was random he would reload and run the same fight again and again until he got the loot he wanted. Sometimes, when he had a series of free periods in a row, I would go to class, and two hours later, get back from class and he’d be fighting the same enemy for the umpteenth time. Now, I’ll grant that in some of the games he played, he also got SPECTACULAR music, or a challenging fight, but most of time, he got nothing of the sort. The games he did this with included games like Dragon Age: Origins and it would have included Mass Effect if the fighting mechanism hadn’t turned him off. (Neither he nor I are experts at shooters. I generally turned the difficulty to minimum because I was in it for the story anyway. On turn-based combat RPGs, I always turn the difficulty up, because I’m fairly good at those. He’s even better due to his perfectionism.) He never finished Dragon Age, despite modeling his character after Sephiroth and thus being attractive AF to Morrigan….

    Anyway, I hate random loot, and in his lucid state, he does too.

    But we both buy Magic: The Gathering booster packs when we get together….

Thanks for joining the discussion. Be nice, don't post angry, and enjoy yourself. This is supposed to be fun. Your email address will not be published. Required fields are marked*

You can enclose spoilers in <strike> tags like so:
<strike>Darth Vader is Luke's father!</strike>

You can make things italics like this:
Can you imagine having Darth Vader as your <i>father</i>?

You can make things bold like this:
I'm <b>very</b> glad Darth Vader isn't my father.

You can make links like this:
I'm reading about <a href="http://en.wikipedia.org/wiki/Darth_Vader">Darth Vader</a> on Wikipedia!

You can quote someone like this:
Darth Vader said <blockquote>Luke, I am your father.</blockquote>

Leave a Reply

Your email address will not be published.