Finer Points: How To Make Loot Boxes and Gachapon Not Suck

Making Loot Boxes Suck Less by Sir TapTap

Loot Boxes and Gachapon are all the rage these days; both the literal and figurative meanings of ‘rage’ in fact. While I frankly despise the practice and think they should absolutely never be in any game in almost any form, there’s a hell of a lot of ways they could be better and as a designer, it bugs me that even free Gachapons are usually done terribly.

How can we make such an inherently abusive and exploitative mechanic better? Here’s all the considerations.

Update 2018-02-19: Added “Speed It Up”, a section on how even a “good” gacha can waste your time.

Finer Points is a series of Game Design/Game UX articles by Sir TapTap focusing on very specific gameplay elements and the major impact on a player’s experience they can have.

It’s Gotta Be Free

Here’s the thing. Monetization influences game design. This is why Mobile games have a bad rap; most of them are F2P, and that heavily influences game design.

You simply cannot objectively and ethically design a gambling system when exploiting the player will earn you more money. Even if you did, you would be designing a game you couldn’t profit from. The only way to ethically include a luck-based system for earning items is if it has absolutely no tie-in to your monetization model. This is simply non-negotiable.

“But wait, I’m not one of those bad designers—” you sputter before I gently place my index finger on your lips with a gently “shush”. Granblue Fantasy’s designers no doubt thought they were “one of the good ones” as well.

The issue with bias is you’re not always aware you’re doing it.

No matter how good of a person you (think you) are, the thought will always be in the back of your head. “People pay money for this. If the odds are low, they’ll have to roll more times. That would get me more money.” You’ll look at your odds and that thought will be in the back of your head. You’ll consider the monetization aspect of every feature you put in.

You simply cannot design a paid Loot Box or Gachapon ethically. Period. No.

Why Is It Random?

So if you’re not bleeding your players drier than the Sahara Desert, why on earth would you include gambling mechanics? To my mind, you probably shouldn’t!

Most implementations of gambling in games, even before microtransactions, are frankly terrible. The slot machines in Pokemon? Grindy, unfair garbage. The slot machine in Sonic 2? A great way to slow players down for no reason in a game about speed. And every random element in a Zelda game ranges from a speedrunner’s worst nightmare (DampePls)  to a casual player’s annoyance.

But if you want to make a gachapon that doesn’t suck, you have to think:

Why am I doing this?
What is the player getting out of this?
Why is randomness important?
Is it actually more fun this way, or are you doing it to keep the player playing longer, to keep them paying money, or something else that’s all about you, not the player?

Odds are, frankly, there is no good reason for your system to be purely random. The reason F2P games have randomness is purely to extort more money from Whales, and if you’re not fleecing people you’re probably just annoying them or trying to needlessly draw out the progression curve (ugh) of your game.

So let’s say you do have a valid reason to include gambling mechanics to your game (as you might have noticed, I don’t personally think that’s often the case). There’s still a lot of design considerations we’ll have to take into account to make such a system enjoyable instead of infuriating.

Cull The Repeats

The simplest, most drastic improvement that can be made for any Gachapon system is to simply remove duplicates entirely. Nintendo’s ARMS apparently does this (I’ll admit I haven’t played it), and it results in a system of progression where you don’t know what you’ll get, but you do know you’ll get everything, and within a set amount of attempts.

This is, in my opinion, the only valid and truly “fun” aspect of gachapon style randomness. You preserve the excitement of not knowing what you’re going to get but you still prevent the “uuuggghhhh” moment of getting another “common draw”.

In my obviously hardline opinion, this is pretty much the only way to ethically implement such a system as by its very nature it basically removes almost any potential for abuse of the player. If you don’t trust yourself to make a non-exploitative model (and I don’t trust you), this is definitely the model to fall back on.

The only real arguments against this method are that it reduces the amount of time (or money) players spend inside your Gachapon/Loot Box system. If that’s a major problem for your game design, consider the fact that you have made a very, very bad game and/or have very, very bad priorities.

Don’t Downgrade Pull Rarity

Note that even with repeat culling you can cause a lot of frustration by what I can only describe as lying about reward rarity. Many games have systems where “Rare” pulls exist with better odds to get higher tier items, but still let you pull “Common” or otherwise inferior items from a more expensive pull.

As a recent example, Xenoblade Chronicles 2 apparently culls rare rewards but not common rewards and rare pulls can still give common drops. This resulted in a friend of mine not getting KOS MOS, despite actively working towards her for the whole game. He played over 100 hours, had completed every possible other thing to do in the game. As of writing this, he still doesn’t have her.. It’s also a big part of why I have not purchased the game myself.

Destiny's Cryptarch with a trollface
Nice Exotic Engram you’ve got there, Guardian. Would be a shame if something…happened to it.

Downgrading the rarity of your pulls is a big problem. This was a common source of infuriation in the early Destiny days where Engrams could drop loot of a lower tier than the apparent value of the Engram. The devs thought this was “fun”. They were extremely wrong.

Destiny and it’s sequel are frankly a treasure trove of examples of what not to do when implementing Loot Boxes/Gachapon. #LootCave

Nothing Is Purely Cosmetic

So the common argument goes, as long as it’s cosmetic you can be as exploitative thoroughly monetized as you like, because any loser Whale that’s stupid generous enough to buy all this stuff clearly has some kind of problem or just wants to be exploited, right?

First, exploiting people who are willing to be exploited is still awful; gambling addition is a serious problem that is increasingly common in children and abusing an addict doesn’t make you a shrewd business man with no moral obligations, it makes you an asshole.

Second, these are video games. Nothing is purely cosmetic in video games. Everything in a video game is inherently arbitrary; you may as well argue beating the game itself is “cosmetic” as there’s no more inherent value in beating a game than there is in unlocking a gold hat or any other “cosmetic” item that’s really hard to get.

Further, video games are all about collecting things. We even have a genre of games about collecting things; the unsubtly titled “collectathon”. We have Achievement Systems on most major gaming platforms because people want to collect things that tell other people they collected all the things. We even have Steam Trading Cards so people can collect things outside the game entirely!

Collecting things is a core aspect of gameplay. If it can be collected, it’s a gameplay element. Any cosmetic that can be considered “collected” is a gameplay element, and if there is any sort of “100% completion” in the game then collection is arguably a major goal of the game. So no, it’s not “just” cosmetic.

Finally, cosmetics are more than cosmetic, and more than collectables, especially in any sort of online game. If you have a dumb hat from a 10 year old World of Warcraft event no one can get anymore, it’s not just a dumb hat, and some people would pay you hundreds of dollars for it in all likelihood. Why? Because other people don’t have it.

Even the most meaningless, stupid garbage in the world takes on an inherent value when you have it and someone else doesn’t, but wants it. And good lord people want cosmetics. People want cosmetics so bad that PUBG is adding once-in-80-years drop rates into their game in order to drive artificial scarcity. For items people can buy and sell with real money, I might add. (If that sounds like gambling, it’s because it is.)

Cosmetics are some of the most valuable items you can give a player in a game. The highest rewards in games are often cosmetic; defeating a Superboss in Final Fantasy will often give you a completely superfluous item. After all, giving a useful item after the player has exhausted any possible use for it is a pretty poor reward.

Keep in mind the value of cosmetics as you design your drop system. They are no more frivolous or inherently worthless than your game itself is.

Let The Player Affect The Odds

A way to make Gachapon even slightly fun is to allow the player to affect the odds. Generally this means paying a bit more or using rarer materials in order to force

Now, it’s still possible to implement an odds system that’s still exploitative. Senran Kagura’s Panty Lottery (no I’m not making that up) for example allows you to pay extra for a higher chance to get a better item. Valkyrie Drive: Bhikkhuni (I’m not making that name up either) has a somewhat more fun system where you play a little minigame in addition which raises the odds if you do it right.

The problem is in both Senran Kagura and Valkyrie Drive, the odds of getting a new item still drops so extremely low you’ll likely be playing the lottery alone for nearly an hour, even with infinite money, trying to scrape the bottom of the barrel for the last couple items.

Why? Well, both games let you pay real money for a coin that guarantees a “new” draw. And that’s why you can’t trust yourself to have a paid Gachapon system and design it ethically. I love the devs of Senran Kagura very much, but the problem with their lottery system is very clear and the reason it’s not “fixed” is even more clear.

Disclose The Gachapon Odds

Actually, telling me the odds would allow me to reasonably assess my current situation and react in a cogent manner.
– Han Solo

One of the fixing-Loot Box design methods you’re most likely familiar with is disclosing Loot Box odds. This is because Apple recently made it policy to do so in iOS apps, and China even earlier made it a law to disclose odds in certain situations.

Of course, people immediately attempted to exploit the exact wording of these laws for monetary gain (one of many reasons you’re not going to see any “Overwatch is Loot Boxes done right” tidbits in this article I can tell you that much).

But disclosing odds properly can go a long way toward informing the player. For instance AbyssRium recently added their odds due to Apple’s policy, and it’s rather stunning exactly how low the odds get. Keep in mind the “Big Luck Shell” column on the right refers to a pull that costs $5 USD worth of in-game currency.

AbyssRiums Luck Shell Gachapon drop rates, which max out at 0.17 for the most expensive option
AbyssRium discloses odds, showing the incredibly low chance of actually getting what you want

As you can infer from the AbyssRium example, this feature alone does not prevent a Gachapon/Loot Box from being exploitative. Rather, exposing the odds shows the player exactly how exploitative your Gachapon is. This both encourages you to make it not exploitative and also informs the player as to whether the odds are worth their time.

Prior to this disclosure I had to make a significant effort to note how anti-player and exploitative the Luck Shell Gachapon was in my AbyssRium guide. It’s not really much less exploitative with the odds disclosed, so the warnings remain up.

Speed It Up

Too often, even an “ethical”, free Gachapon added to a game for the purpose of fun will in fact be very, very tedious for one unexpected reason: elaborate pull animations.

Often devs will try to make Gachapon exciting, after all the mechanic is designed by the (questionably) “fun” noise of a toy lottery machine going “Gacha!” as it turns around. In addition to a triumphant noise, this will often involve an elaborate animation meant to build some suspense and make the experience more psychologically impactful. The whole process is very similar to elaborate real money gambling machines (I wonder why).

But like all too-many repeated animations in gaming, these get old after a few attempts. What was cool the first try is just an agonizing slap in the face after the 100th time. If your Gachapon needs more than a couple dozen pulls, the player should be able to bulk-pull or skip the animations easily; ideally both.

To test how frustrating your Gachapon is animation-wise, create a test case where you have more than enough ingame money to buy your way to victory, then sit down and manually pull until you “complete” the set. If you hate the game now, odds are your player will too! The exact amount of pulls one should be able to buy in bulk will vary per game of course, but if the player already has the money, you’ll want to make sure they don’t have to waste an hour watching the pull animation just to get their rewards.

A Good Example: Smash Bros Melee

One of the earliest and best Gachapon systems that many players likely experienced was the Trophy machine in Smash Bros Melee.

Melee’s system was simple, by playing almost any mode in the game you would earn Smash Coins, which had no purpose but to put into a Gachapon machine to little digital “Trophies” that came with a cute little description. They were concentrated doses of Nostalgia that helped accentuate everything that Smash Bros was about, and they were very unobtrusive.

Smash Bros Melee Trophy Lottery screen
Smash Bros Melee Trophy Lottery screen

A literal Gachapon with the machine and capsules and everything, Melee’s machine might seem like the very thing I hate, but it’s a very good implementation. Melee’s Trophy machine:

  • Trophies were cosmetic, and not the “show off to your friends in WOW” style of cosmetic
  • Coins were earned at a fair right through almost any method of playing the game, meaning they scarcely felt like a grind
  • Coins had no purpose but the lottery, meaning nothing is meaningfully lost by pulling for the Gachapon
  • No money could be spent
  • Completion was extremely possible within a reasonable amount of play
    • Additionally, there was no reason to continue after completing the collection
  • Odds were displayed directly
  • More coins could be spent to increase the odds of a new trophy

This implementation only violates one of my recommendations: it doesn’t cull repeats, but it’s other aspects make up for it. Due to the pool being fairly shallow and the ability to raise odds, 100% completion becomes very possible even without repeat culling. And technically speaking, being able to affect the odds like this is just a more limited method of repeat culling.


The number one question you should ask when designing your Gachapon/Loot Box system is: “is this fair?” and the second: “is this fun?”.

Video games are one of the purest cases of User Experience Design; everything is designed specifically for the enjoyment of the user…until Free to Play mechanics slip into the mix. As a game designer you are by definition designing user experiences, and must keep them in mind first.

Your game probably shouldn’t have a Loot Box, but if you do, keep in mind this bulleted list of considerations:

  • Are we just doing this for money?
  • Is this actually fun for the player?
  • Does pulling most often feel like earning rewards, or being punished?
  • Is rarity clear? Can a player be surprised by a lower-tier item than expected on an expensive pull?
  • How can we present the odds in a fair manner?
  • Can someone with a gambling addiction play our game without problem?
  • How can we ensure the grind isn’t endless?
  • Would I enjoy a game with these mechanics myself?

Randomness in gaming doesn’t have to be bad. Most of the bad Loot Boxes and Gachapons out there are in fact deliberately bad (for the player anyway). But they can be better, and as a designer you should always strike for “better”.


Support my guides and articles on Patreon!