The Podcast: A Turning Point For Loot Boxes?

Haydn and Brendan discuss the potential consequences of recommendations that the UK government introduces regulation

It's a smaller team this week, but no less meatier a discussion. Haydn and Brendan debate the potential fallout from the DCMS committee's recommendations that the UK government regulate loot boxes under gambling laws.

While these remain recommendations at the moment and are not enforceable laws, it's nonetheless a concerning direction for the ongoing debate around how the games industry monetises its audience. 

We discuss the various responses to the report, the implications such laws might have for the industry, and how this latest news -- when combined with headlines of crunch, abuse, mass layoffs and more -- is not painting the best picture for the games industry.

You can listen to our latest episode below, subscribe to our RSS feed, or download the file directly here. It is also available via Spotify, iTunes, Google Play, Stitcher, Overcast, Player FM, TuneIn and other widely-used podcast platforms.

More stories

Indiana Jones and the Raiders of the Blockbuster IP

Latest episode available to download now, in which we discuss the resurgence of licensed games

By GamesIndustry Staff

What we learned from PlayStation Vita | Podcast

Our weekly news discussion show is back, now with a new monthly section based on 10 Years Ago This Month

By James Batchelor

Latest comments (4)

Ian Griffiths Product Owner, HutchA year ago
Hayden, you say that the industry has a fair chance to prove that there were no problems but I do t think that's true at all. After all if there are no problems, how do you prove it? How do you prove a negative?

Of all the empirical evidence available there is none that shows that its gaming or even purchasing that forms the key part of any problems in or around gaming. The best most recent evidence on gaming addiction suggests that it's not an actual clinical condition.

There was nothing the industry could have said that would have pleased these regulators beyond parroting what seemed to be already drawn conclusions - that gaming is a problem and developers should spy on consumers and tell them when to stop.

From what I saw not only did the panel not seem to believe that developers and publishers thought it was the individual or parent's responsibility to control their own lives; as the individual benefits from the good and pays the price for the bad and is therefore best able to make that judgement, but the panel seemed to believe that developers agreed with their ideology and purposefully designed games for players to play longer against the player's best interests solely for the gain of the developer or publisher. The only way I can understand the behaviour of the panel is through their ideological dogmatism; that it is systems and outside players, not individuals, who are responsible for the state of an individual's circumstances and behaviour.

Let me be unequivocal in my opinion, this inquiry was a sham whose very title showed it's pre-conclusions. I believe it was nothing more than a witch trial; if games admitted they were evil they would be legislated. If they denied they were evil it was just evidence of the evil people behind them and needed to be legislated.

Edited 3 times. Last edit by Ian Griffiths on 17th September 2019 11:24pm

0Sign inorRegisterto rate and reply
Ian Griffiths Product Owner, HutchA year ago
Guys, please... please find me the company that weight's it's loot boxes 'against' the player. This idea comes up so often but I've worked in f2p for 11 years, I've worked with loot boxes that whole time, I have a maths degree and an interest in probability and statistics and I've never come across it or anyone saying they're doing it.

Now I'll preface that I don't know that it's being done, I'm happy to be proven wrong. I'm also excluding 'Pity timers' because those normally get put in when a designer says, 'Isn't it unfair that you can open loads and never get anything good. I'm strictly speaking about doing it to maximise revenue from an individual or pattern of behaviour of a cohort in a game.

A lot of this just comes down to volume or data and the complexity of the game. It would be easier for a company like King with a lot of players to test a change than say, Blizzard. So first, you've got the conversion issue, less than 10% of people will spend in your game. Let's say they do spend once, less than that will make a repeat purchase, and that goes on for each n-th purchase.

Right so let's say we've got this data, we've got an issue, confidence. The more people do something the more confident we can be that will be the general pattern. We can use statistical methods to determine how confident we about some property or behaviour of a group. Maybe we want to look at our confidence intervals for some sequential change we've made in a game. We can also look at things like effect size, the smaller the effect the more people we need to be sure of our results.

So let's say we've got a loot box and we want to know if a 1% roll is better or worse than 1.2% roll for Legendary items. Well, given that roll not many people will get it, couple that with the above effects of making your game free and you may not be dealing with enough players to understand the effects that we'll. You then have compounding factors like potential changes to long term behaviour, and unless you want to run these tests for a very long time in what is a quickly changing environment, you're not going to get a good read. Now, you could do something more drastic; compare a 1% drop rate to a 10% drop rate, you'll get a clearer answer but you won't have as granular an understanding of the way small changes affect player behaviour. Also, as you probably don't want to wholesale roll out a potentially suboptimal change to half of the players you'll stick with a smaller subset, again requiring longer to get results.

As for knowing what a player wants, well, while we have some control over desirability and probably know what a player may generally 'want' it's he'd to know exactly what they are after. Like, in Animal Crossing how well could you predict what item a player is going to want next, you may have a fairly good idea but I doubt good enough to make a reasonable prediction.

So if someone is doing this level of drop rate manipulation I suspect it's through arrogance and they probably don't have enough evidence to be sure they're making the most optimal decision. But as with most development, there's probably not enough time and resources with the next Halloween, Christmas or Valentines Day update.

To sum up, I think there's too much assumption of bad faith in the people behind loot boxes and way too much speculation about effort going in to 'manipulate' players. I suspect 90% of the time there's just some economy balancing guy who hasn't thought much beyond, or should get it in about 1 of 10 box openings. Honestly, the number of experienced people I've interviewed for this stuff who can't figure out how many boxes you have to open to be 95% sure you'll get a desired item is shocking.
0Sign inorRegisterto rate and reply
Brendan Sinclair North American Editor, GamesIndustry.bizA year ago
Well, we don't know they aren't doing it. But we know they've patented techniques to do it (!). And we know they're using analytics and A-B testing to maximize revenues. And we know they aren't being forthcoming with legislators who want to understand how their business works (

So it's basically a question of credibility at this point. And the most generous assessment I can make is that the industry has badly misunderstood people's concerns and misplayed its hand over the past few years, such that parents, politicians, and the press aren't really extending them the benefit of the doubt here. (

I get that it's rough for people working on free-to-play or loot box games to see their work demonized for a bunch of stuff they might not be doing, but I'd ask them to consider how it looks from the outside, when all we have to go on is their word that they would never do anything wrong, but they also don't refuse to talk about what they are doing.
0Sign inorRegisterto rate and reply
Show all comments (4)
Eddie In Product Manager - Games, Mobile, Boss Fight EntertainmentA year ago
Ian, myself and heck even EA can scream at the top of our lungs that none of us engage in odds manipulation, but why trust us to be anything but serve our self-interest? In the age of social media and livestreaming, there's already plenty of data available for the public to prove whether their game is engaging in odds manipulation in loot boxes. For example, one could record a FIFA streamer's Ultimate Team pulls and perform statistical tests to verify whether anything other than what's happening on the rate card is occurring. It is precisely for this reason that any reputable company would never engage in practices like odds manipulation. The cost of getting caught with that sort of thing far exceeds the benefit of getting a bit of extra revenue.

You mentioned AB testing and analytics. I'll tell you what I do actually do. Nobody likes getting ads, particularly ones that aren't relevant. I, as a marketer, don't want to show you something you aren't likely to buy, that causes people to churn. Analytics can tell me about your game state and purchase patterns so that I can offer you something that you want, at a price point you're likely to purchase at. Most gaming companies aren't particularly sophisticated in this regard and we're no where near say Amazon when it comes to our practices. Heck, most gaming companies don't have very smart targeting at all and just spam every offer on everybody.

I've had conversations with folks who say that using data to make my operations smarter is inherently shady. If this is where we're at, then we have way larger problems than just microtransactions in games. If not, I hope we can largely agree that it's better to display offers for level 50 stuff to level 50 players and not show $99.99 offers to people who've never spent before.

Edited 1 times. Last edit by Eddie In on 20th September 2019 9:24pm

0Sign inorRegisterto rate and reply

Sign in to contribute

Need an account? Register now.