Will Developers Ante Up for a Gambling API? | Hacker News Reader

Will Developers Ante Up for a Gambling API?

blog.programmableweb.com

57 pointscoloneltcb32 comments

> According to Betable’s Griffin, social games average $1 of monthly revenue per user. Existing gambling games, which Griffin says are mostly boring, average $300 of monthly revenue per user.

That seems way, way off. $300 average would require either all of your players are hardcore and put in a lot of money, or you have a decent number of players at the high end skewing it. Assuming they're getting something out of it (at least some of the time), this number seems to be quite... deceptive. I'd love to see the data behind it.

It's the latter. A small number of big spenders ("whales" in the industry lingo) can easily skew average number. Median is a better metric than average here.

Of course, the social games metric includes a large number of players who are paying nothing. I suspect it wouldn't be quite so low if you discounted those players, which would make it a more fair comparison (as gambling would in the majority of cases require you to actual put some money up).

That's a totally fair point. We think of customers on a revenue spectrum. In social games, that spectrum begins at zero and goes up. For real money games, that spectrum begins at the minimum bet allowed - usually one cent on penny slots. Instead of trying to explain away ARPU by dividing customers into paying and non-paying, we want to draw attention to the inherent monetization problem faced by social game developers. Real money gaming far outperforms virtual play no matter how you slice it.

To directly address your question regarding paying users versus non-paying users, traditional social gaming ARPPU is around $30. The real-money multiplier would still be a massive 10x delta.

>Median is a better metric than average here.

Not in this case! Remember that the goal here is to maximize revenue (see footnote) so it is a bad move to ignore the existence of the whales -- because whales represent a big fraction of our potential revenue.

Footnote: since most of the costs for this kind of business are "fixed" costs, e.g., the cost of paying the developers to built the site, the way to maximize profit is to maximize revenue.

I was replying Daeken's comment on why $300 seems to be a high number. This number is not representative of average spent by a majority of users but that's what most people assume when they hear "average $300 of monthly revenue per user".

For you to maximize your revenues, by all means go after the whales as that's where the money is. I worked in online gambling industry for a number of years and have seen this happen quite a few times. I.e. an online casino with about 12000 of active users had 85% of its revenues generated by a dozen (as in less than 12) high-rollers.

OK. Looks like you no longer hold that "median is a better metric than average here," which is what I was taking exception to.

>I worked in online gambling industry for a number of years . . . an online casino with about 12000 of active users had 85% of its revenues generated by a dozen (as in less than 12) high-rollers.

Thanks for the interesting info.

This is very interesting for use with a mock gambling game - one where users cannot cash out. Users can buy in-app chips to keep playing. Is this the entire purpose of it? I'd think actual gambling online, even with no house edge, has proven to be a "den of thieves" (Full Tilt, Ultimate Bet) how would this actually solve the problem of trust?

One of our primary goals as a company is to be entirely white hat and trustworthy. Our investors, partners and the lawmaking body that granted us our gambling licenses all fully support our product and the service it provides. We worked hard to earn the trust of these institutions, and we will work hard to earn consumers' trust as well :)

I'm curious as to how this model will work if users can't easily deposit money online. I can only imagine they will inevitably face the same problems as online poker where credit card companies and banks will block any transaction related to gambling sites or 3rd party eWallet proxies.

In the early online poker years, you could easily deposit directly from your bank or credit card. Then banks started blocking these transactions. The online casinos came up with the eWallet proxies such as Neteller that served as an intermediary between the banks and the poker sites. Those too were blocked. Then gift cards became the main funding source, and eventually those were blocked. Eventually the process ended up with direct wire transfers or payment-by-phone. As each deposit method was blocked, it became harder for the casual players to get money online and they stopped playing.

You are correct, credit card companies in the US will block transactions to gambling sites. However, I must make it clear that Betable does NOT facilitate illegal online gambling. We only let players from legal gambling jurisdictions play games for real money. All other players can still play, but for virtual currency.

For a couple of years I've been cooking a gambling API and while there is a huuuge market for that, execution is what separates wheat from chaff. Here are two key things I believe will make it a winner:

1. Make it easy for developers to integrate in their apps.

2. Make it easy for users to deposit/withdraw/bet money.

So, if I am playing online poker on my iPad, I may call, raise or go all-in on very single hand. Can your API do that? What means all-in in that case? All I have in the bank?

Allow me to move $100 from my account to any game I feel in the mood to play. Let me deplete that bag and go for more or just quit in the middle and get the rest back. Can your API do that?

Btw, call it iChips and put it on the app store (there is already one app with that name, but fight for the name) and use colorful casino chips everywhere.

Again, execution is king...

These are solved problems both in meatspace and online gambling. Your buy-in is separate from your wallet:

> I may call, raise or go all-in on very single hand. Can your API do that? What means all-in in that case? All I have in the bank?

When you "sit down" at the table, you specify how much you're buying in with, between $TABLE_MIN and the lesser of ($TABLE_MAX || $YOUR_WALLET). Your all-in bet is a bet of whatever you've got at the table (not in your wallet). That's a fundamental rule of poker games.

> Allow me to move $100 from my account to any game I feel in the mood to play. Let me deplete that bag and go for more or just quit in the middle and get the rest back. Can your API do that?

Again, that's a fundamental gaming paradigm that online gaming clients have done well forever.

Making a gambling API is IMO a developer's dream because all the rules are very clearly delineated already, and have been mostly agreed upon by decades of IRL usage and regulation. In other words, the spec's been written, tried, iterated on, and verified. As far as specs go, gaming rules are a thing of beauty.

These are the two things that will make a winning API:

1) Trust and integrity (including auditing).

2) Easy transactions.

The big blocker here is client integrity. What if I wrote a slot machine program that took 1% of winning rolls and displayed a loser to the player and took the winnings for myself?

Hey Domenic_S

So you know, that example of poor client integrity is something that we 100% prohibit. We audit every game before it goes live for real money on our platform, and since we control the gambling transactions, we make sure players get correctly paid out.

Also, we are audited by the UK Gambling Commission annually, so our gambling licenses are testament to our own integrity as a service (the UKGC does not take the license-holder commitment lightly!).

Interesting stuff and the ARPU generated from gambling revs will ofcourse be a massive draw to devs looking to use the Betable API. We're working on a social sports prediction game with a football(soccer) focus at launch but have consciously avoided building in gambling mechanics. We will run with an affiliate model where we can achieve around 15-30% of a users lifetime value from the provider.

Our aim is to focus on the gameplay and have the cash betting as an extra. With such tricky global regulations this frees us up to pick and choose partners that meet local laws and expand into regulated markets with new models.Of course Betable would prove a great expansion of our product in future.

Interested to hear HN's thoughts (affiliate vs betable) and would love you all to play and inform our beta later in the year: www.betstars.co.uk

Interesting solution to the regulation "problem"

I think the big question here is not whether developers ante up for a gambling API. Certainly some will, given the prospect of making more money. The real question to me is whether it becomes 'required' in the sense that in order to compete, they need to use it (or a competitor's equivalent). The social game industry seems to be all about fast-following on features, especially monetization

The main problem (as I see it) is you're limited by the types of games offered by the API. It would be cool if you could provide your own logic for a game, but of course that's a difficult problem for Betable to solve since they need to know that the house has an edge.

Multiplayer games would also be fun but then you have to guard against collusion and other exploits.

Hey dminor, you'll be happy to know that one of our company goals to to break real-money games free from the traditional gambling gameplay models (slots, roulette, etc). We strongly encourage and support developers that build innovative new real-money game mechanics into their game

Casinos regularly dedicate 80% or more of their floor space to slot machines because those games are the most lucrative and easiest for people to play. The Betable API probably offers a "good enough" solution for building these simple games that make up the majority of what casual gamers will want to play.

That's a good question, but I don't think it's the right question.

It all depends on how secure it is. You can have a very unsecure API that no one will touch or an incredibly secure API that no one will use because it's too difficult.

Creating an API is not the problem. Finding the right balance of security and convenience is the problem.

I'm curious how this would work. If I were to use this API to create, let's say, Bingo for money. How would I use the API without going to jail for running a gambling operation? I mean, I'm in the US so I guess I could always find an indian casino to back but that isn't likely without them taking a huge cut.

AsylumWarden, with us you don't need a gambling license and you run no risk of "going to jail" in the US. Betable handles and hosts all of the gambling transactions, so you don't touch any gambling. Therefore, you don't need a license and can be anywhere in the world.

That's not true in the U.S. If he uses your site's gambling API, he is engaged in the business of gambling and must have a license. Since the U.S. generally does not allow online gambling, it is effectively illegal regardless of where the servers are physically stored.

FultTiltPoker and Bodog were both hosted from offshore (and generally run from offshore), but that did not help their U.S. owners in the slightest when the FBI came knocking.

Moreover, by using a gambling API on your own site, you place all of your bank accounts at risk of being frozen by your financial institution (if they are U.S. based) as U.S. financial institutions are not permitted to transfer funds to gambling sites. You could use an offshore bank, but this would increase compliance costs and guarantee an annual audit by the IRS.

EDIT: This assumes games are played for cash. Games played for virtual currency are not considered gambling unless they are easily convertible back into cash.

> That's not true in the U.S. If he uses your site's gambling API, he is engaged in the business of gambling and must have a license. Since the U.S. generally does not allow online gambling, it is effectively illegal regardless of where the servers are physically stored.

It's perfectly legal for an individual or company in the U.S. to develop and operate gambling software. What's not legal is offering services to players in the United States or elsewhere where gambling is illegal[1]. Betable offers the gambling services to players, not the developer partnering with Betable.

Furthermore, Betable does not allow real money play by players in jurisdictions where gambling is illegal. Game developers in the United States (or anywhere else) do not need to be licensed since the regulated bits of the infrastructure are all developed and operated by Betable. We strive to meet and exceed the regulatory standards that we're subject to and provide a legitimate and trustworthy ecosystem for players.

> FultTiltPoker and Bodog were both hosted from offshore (and generally run from offshore), but that did not help their U.S. owners in the slightest when the FBI came knocking.

Full Tilt Poker and Bodog are both examples of online gambling operations that flagrantly and publicly disregarded regulatory requirements and/or the law, offering their services illegally in the United States or, in the case of Full Tilt, (allegedly?) defrauding customers. Unsurprisingly, the FBI took offense to these violations. However, there's really no comparison between those operations and Betable. As I said before, Betable does not offer online gambling to players in the US, which is the main issue the FBI took offense to.

> Moreover, by using a gambling API on your own site, you place all of your bank accounts at risk of being frozen by your financial institution...

Betable manages all of the banking and accounting systems associated with gambling using its own accounts. Developer accounts are not used for gambling. Betable acts as the arbitrator of all gambling transactions.

1. In fact, with the recent re-interpretation of the Wire Act even that statement is now inaccurate at a Federal level, but that's another post.

Gambling winners are most likely money laundering* recipients. In the overwhelming majority of cases, gamblers lose.

* http://www.library.ca.gov/crb/97/03/Chapt11.html

This is sensationalist, but I'll give you a little more background on what we do (even though many of these practices are not nearly as prominent as the article says they were in the 1960s - 1990s). In order to operate as a legal gambling entity, we have to build and maintain a host of elaborate security, anti-money laundering and counter-terrorist protocols (one of the things that comes with the territory in the gambling space). We work with top partners in the industry, who have decades of experience dealing with these problems. While security and anti-fraud measures have improved significantly, this is a very high priority for us internally that we've put considerable time into.

Sensationalist?

It is a link to the California State Library.

Right, I understand that. I felt that it was sensationalist in the sense that it's unlikely to happen, but very "sensational" if it does. I should have focused on my response to the points in the article.

Do they want me to pay for a random number generator? I don't understand it?

The way I understand it, they handle money and regulations, i.e. collecting money from the player, giving the player actual money back when the player wins, in a way that is 100% legal considering gambling laws in all countries involved in each transaction. That's a lot of stuff for a regular developer to take care of.

The part where they also run the gambling logic & randomness is probably necessary because regulations on gambling often require certain guarantees and auditability that, again, a regular developer may not (want to) be aware of. It's much simpler to be compliant if the dev only has to establish f.ex. win rates for different actions, and the gambling provider runs the actual processes with methods and random generators that comply with applicable laws.

Jare, you nailed it. My only clarification on your second paragraph is that the RNG needs to be audited and certified in ordered to get your license granted/renewed. The idea here is that it prevents shady gambling companies from defrauding players.

You are not paying for the RNG, you are paying for the knowledge of when and where it is legal to use it.

Reply on news.ycombinator.com