Show HN: I made a site that lets you bet against Bitcoin Hash Rate Volatility | Hacker News Reader

Show HN: I made a site that lets you bet against Bitcoin Hash Rate Volatility

futureblock.com

128 pointslocksley83 comments

Is this legal in America?

Edit: Why the downvotes? I'm curious if they got an opinion from a lawyer on the legality of their site.

IANAL, but my guess is it's in the same gray area that betting on sports with your local bookie or playing online poker is. More specifically, the site itself is probably illegal in the eyes of most agencies but the players participating probably aren't acting illegally.

That said, given the fact that there were a total of 14 bettors in the first two rounds, this sounds like the smallest of small fry when it comes to the world of online gambling.

I wasn't aware there was any grey-area for domestic bookies... just because it's not sports doesn't mean it's not bookmaking.

Online poker is grey-area because the poker outfit itself is generally outside the scope of US law. If it were more legal, you can be sure their servers would be hosted domestically in the US where it's cheaper and faster.

Your local bookie, however, is in the same scope as a guy running a local card room -he's taking a cut off of people's gambling habit. Whether it's sports or not, it's still going to be bookmaking, and still trouble. Extra trouble if it's over state lines.

The grey area with online poker is because it isn't explicitly made illegal in the wire act (online sports betting is explicitly illegal), but the UIGEA legislates against games of chance which there are arguments that poker is not such a game but it hasn't been tried in court. Many sites used that argument to offer online poker to US residents and had their domains and bank accounts seized, Pokerstars settled for over 700mm and bought Full Tilt Poker in the process. The US govt doesnt care where you are based if you are violating US law and attracting US customers.

The business is more than likely violating gambling regulations by soliciting US customers. IANAL, but Satoshi Dice did block US players due to legal concerns https://bitcointalk.org/index.php?topic=135521.0.

I was more so curious if a lawyer might try to categorize this site under currency trading or some other type of derivatives trading rather than straight gambling.

(IANAL)

Can you explain how they blocked US players? I skimmed that page and only saw "not possible since you play via sending to a bitcoin address; there's no IP to block".

Presumably they blocked US players who don't know how to use a proxy?

Can you explain further? Playing is literally done only by sending to a Bitcoin address. You win or lose and get paid back or not. There's no website involved in the actual game.

By "proxy" I didn't necessarily mean "web proxy". If at any point an IP address is visible it is feasible to exclude transactions that involve US IP addresses. This is, of course, trivial to circumvent - which was my point.

Bitcoin is a decentralized system, so there is no reasonable place to put these IP restrictions.

If there is a single Bitcoin mining node (out of thousands[0]) that will accept your transaction, you can play.

[0] http://blockchain.info/connected-nodes

You're not speaking to the important question, which is whether there is any way for the recipient to tell the originating IP of a bitcoin transaction. I am gathering the answer is "no". If "yes", then there is an obvious place to put the IP restrictions: in the satoshi dice service itself. Obviously, even in this case, it could be defeated with a proxy.

The point is that the recipient of the money doesn't interact at all to receive it; the rest of the network just "decides" (based on the transaction) that such address now has X more coins. The machine with the address/wallet is not consulted.

That is not the point. The information could be available, and the recipient could therefore act on it. If the information is not available then that is the point - but that could in principle be the case even if there was interaction.

But it's not like the recipient could invalidate the transaction even if it had that information.

No, but it could either return it or pocket it, rather than letting them play.

> You're not speaking to the important question, which is whether there is any way for the recipient to tell the originating IP of a bitcoin transaction. I am gathering the answer is "no".

Correct. There is absolutely no way whatsoever to determine the originating IP address of a bitcoin transaction, short of global network surveillance.

I believe all they did was block US users from seeing the website - there is no way to block users from transacting with Satoshidice.

Sorry for the confusion. They blocked US players from their website where they could get information on what was currently being bet on and where to send their Bitcoin.

IIRC Satoshi Dice worked by sending Bitcoin to a specific address for a bet and the address changed each time which they displayed on their website. By blocking US players from the website they could legally claim to be in compliance with the regulation (IANAL).

Hmm the site seems to load the same for me US IP or not. You do have to click "I agree" that you're legal, though.

Well some months ago it was a separate landing page, but I guess they changed it. They still don't display the information unless you agree that you are in a legal jurisdiction to use the service.

> Is this legal in America?

Not in the US and probably any other county unless he has a license.

The issue with this is that it appears the site is being hosted in Wilmington, DE. Since this is not a game of skill, it is almost certainly illegal to operate within that state and the states of all of those that bet from the US. Since it is illegal under state laws, the federal UIGEA (unlawful internet gaming enforcement act) also applies. It could move to a futures market model, but as it stands right now, a number of federal or state jurisdictions could press charges. I'd suggest, at least, that you move it offshore. That won't make it any more legal to take bets from US residents, but it will at least slow them down trying to track you and make it much more difficult to obtain evidence etc. If you don't want legal hassles, take it offline.

I'm somewhat confused with the calculation on paying out. Betting 1 BTC pays out:

    Bet above: 2.2284 BTC
    Bet below: 1.0903 BTC

So... I always win?

Edit - I'm an idiot, and forgot I'd have to bet twice.

That's the payout if you bet above and it beats the prediction, or you bet below and it doesn't meet the prediction.

I completely failed at maths just now. I forgot I'd be betting twice. Time to get off the internet, I think.

Very cool! I'm curious what tech/lib/api/sdk you're using to automate sending payments (ie. in the case you stated of someone betting over 1btc)?

Because of the nature of this site, we are very sensitive to any security related questions. Sorry!

Without revealing anything specific, we have a number of trigger conditions and any bet size over or under the limit is a trigger condition. That marks the bet as invalid and manual admin action is then required to refund the bet.

Security through obscurity is no security at all.

I don't think you understand that quote. Only security through obscurity is obviously no good, but that doesn't mean that security through obscurity is entirely worthless. It plays an important role in a layered defense for information assurance.

One suggestion: The site needs to be much clearer about the betting time period and the time period that is being bet on. I can't see anywhere on the front page that explains this. I had to go to the detailed betting breakdown to work out how the time periods even work.

If you can't tell exactly what you're betting on on the front page, you shouldn't expect people to bet on it!

I accidentally discovered the time period by hovering over the open/close/payout bar.

I think you're right about this. Any suggestions for UI/UX elements?

"After 9.10% House Fees Deducted"

Isn't that a little high?

It depends on when you place the bet.

> House Fees: To encourage betting early, house fees begin at 0% of your bet and increases linearly to 20% when the betting round closes.

(http://www.futureblock.com/how-it-works)

Well.... not really.

Parimutuel betting, not sure - but there is always a house fee - that's how the house makes money.

A traditional bookie would have at least 5 to 10 % built into his odds as well (like if you were betting on even odds, liek a coin toss, it would be bet $110 to win $100 for either side (so the bookie has a 5% hold if he can keep the betting balanced, no matter who wins)

etc etc.

In this case, there is 0% fee if you bet when the round opens.

I think the bootstrapping is worth more than getting an initial basefee, since it'll likely bring more players, making it worth it towards the end.

Is anyone else thinking with cloud computing and the right amount of seed money, you could bump the numbers before the round ends, raking in your winnings + whatever the cloud mines in the hour you run it for?

You'd need ASIC Miners, but even then, you'd have a minuscule impact.

More importantly, if you have that much Gigahashing power, it's not worth manipulating this. You're better off sticking to mining Bitcoins.

Not necessarily. If the market is large enough and you control a significant portion of the hashing power (e.g. pool operators), you could bet below and then halt all hashing for the checking period. You would potentially lose a block, but would offset that by gaining more than 25 BTC in bets.

It's a daily bet on a weighted 12 hour average. So you'd probably need to lose much more than a block.

Given the size of the market and the amount of hashing power required to do that you'd almost certainly lose more money from not mining during that time than you would make on the trade.

Interesting , it may be illegal but it can also be cheated by a sort of insider trading. Say i own a number of ASIC miners , I turn them off , and then bet for a number higher than the one expected by the public at large . At the very last moment i turn all of my ASIC miners on. I make money with > 95 % probability.

Unless it gets huge that will never be worth it. If you have enough hashing power to affect the difficulty that much, you'd make a lot more by just mining

One important question is: could one create a short BTC contract at a reasonable price? so that when I hold BTC I'm not subject to the downsides? For every short someone holding BTC would have to find long exposure which a BTC holder wants to give up. I'm not sure its possible, I don't think so.

Shorting will require more borrowing/lending in the BTC ecosystem. But there are other ways to mitigate downside risks, such as hedging (put options in particular). This is really our goal, to create a futures/options exchange.

As we get more volume, then we can do stuff like this and let the market dictate things like "Prediction Target" or "BTCUSD Strike Price".

I don't think that will work, because selling puts on Bitcoin is much too risky. Long puts and shorts are really the same thing (put has a delta and vega component). So while this is a most interesting space, I don't think this problem can be solved, at least not easily. But if the exchange system works better it's much easier to go out of BTC and into $,€,¥.

I'm sure you could buy Call options[1] to eliminate that downside risk, if someone is willing to price and sell such an option to take that risk for you.

[1] Call options[2] give you the right, but not obligation, to sell an asset at a given price, so if the strike price of the option is the price you don't want the BTC to go below, and the price of BTC does go below that value, you eliminate your downside risk because somebody has already agreeed with you to buy at that price - and you have to pay for that agreement up front.

[2] Options ~= Insurance, where ~= means 'pretty much equals'

You buy puts to hedge longs. Put selling puts on BTC would be pretty insane. As a whole there are no insurance sellers. Insurance will be too expensive. But perhaps there some kind of crazy solutions to offset volatility to the network or something similar.

The OP was talking about shorting so I mentioned Calls because of the buy back at the end.

There's not a lot of historical BTC price data and the volatility is all over the place, so using mathematical option value models would be pretty crazy, but they might give an idea of your baseline price before using a major premium risk factor multiplier as well. 'Greater fools' and all...I'm sure somebody would buy them if they were being sold.

Volatility comes from uncertainty. That is going to remain high. Which means if I hold Bitcoin there is the chance that I will wake up tomorrow losing 50% of the value. But insurance for that decline is to expensive. Perhaps one could create different kinds of insurance or transfer mechanisms which mitigates that risk.

That's what options are and there are pricing models that take into account high volatility accordingly. I call them 'insurance' because that's how I see them after learning and working with them. They are simply a transfer of risk from one party to another; the buyer of the option is selling their risk, the seller is taking it on for a period of time with certain conditions. The 'option' part is whether they are exercised, just like when you buy fire insurance (a risk you don't want to have) and your house doesn't burn down, the insurance seller is ahead and your risk as the buyer is reduced or eliminated. There is also the problem of determining how much risk the option seller can take on and handle before falling over when they have to pay up, but I'm sure there's also options available to mitigate that risk as well; it's turtles all the way down.

I think that is probably the safer way to go with this. That is to say, don't make it __actually__ betting, but rather make it be a short-brokering-service. Match people wanting to short bitcoin values with those who expect it to go higher, and it seems that this would no longer be betting.

@locksley: Couldn't the operator of blockchain.info game this by manipulating the output of their API?

Good point.

1. Blockchain.info is relatively trustworthy.

2. However, we have our own hash rate calculations by directly interfacing with the actual blockchain. This isn't in production yet. This is because hash rate is always going to be a guestimate. Depending on the time intervals that you pick, the guestimates vary. So we asked overselves, what helps us to maximise transparency. So we decided to go with blockchain.info's hash rate because everyone can check that.

Kudo's my friend, this is really cool idea, and it's also really well done. I think this would still be fun even if it were just for internet brownie points, but actually being able to bet btc is awesome - maybe I'll take the btc fortune (.3 btc) and let it ride :)

Cool idea, suddenly want to buy bitcoins just to participate :)

1,648,531 GH/s is the current target. I can bet below or above that. It maybe sound odd, because I am not really into the Bitcoins, but what (in the very uncertain case) if the target is hit (neither below or above)?

Difficulty is divisible to many digits:

https://blockexplorer.com/q/getdifficulty shows current difficulty at 189281249.28103

Actually, we're using Network Hash Rate rather than difficulty. Difficulty changes every ~2 weeks.

This is the API endpoint we're polling is https://blockchain.info/q/hashrate

Hash Rate is an inferred number though. See http://bitcoin.stackexchange.com/questions/11139/how-is-the-...

Aha! I should read more carefully.

Haha, given the daily hash rate volatility can be +/- 300,000 GH/s, there is a 1/300,000 chance that'll happen.

But in all seriousness, if that does happen, we'll refund all the bets without taking any house fees.

Well, you know, if there is a chance that something goes wrong ...

Anyway, cool site, great idea and thx for the links above!

Don't bet more than 8 Bitcoins because you can never win in your detailed example?

There's a max limit of 1 BTC, if anyone accidentally bets more, the bet is refunded (without the house fee). There's also a min limit of 0.01 BTC.

Is this just purely a source of mutually-verifiable entropy for pure-entertainment bets, or is there something people can use these numbers for / something people can use this as a hedge against / &c. ?

Great question.

That's exactly what we want to eventually do. For an efficient market to run, there needs to be a way to manage risk.

However, if you mention "derivatives" to any normal BTC user, the mostly response of "huh?", or "aren't those the things that made the world collapse in 2008?".

Future Block, as it currently stands, is an entertainment site to bet on something novel. However, our hope is that:

1. It serves as an education tool to introduce people to derivatives, because derivatives are essentially about taking one position or the other.

2. To build liquidity for a derivatives exchange. Parimutuel applications are great for this because you don't have the 'chicken & egg' problem where you need 2 types of customers for it to work.

Hope that answers your question!

More or less, thanks!

Big enough player can guarantee a result. Luckily only 1 coin limit so far.

Big enough player won't have one coin limit...

Hmm, I'm not sure what you meant, but it does occur to me that, obviously, one has to trust the house to be honest here.

I'm not sure there's any way to be sure the house is really paying out the entire pot -- let alone to be sure the house isn't intentionally manipulation the actual market itself!

Good work. Pretty cool site.

This is cool: "Our long term goal is to build an open derivative platform that anyone can participate in. "

Not that I don't trust you guys. Is there a way that I can test the site?

Thanks!

Well the nature of Bitcoin is that all transaction are anonymous but transparent at the same time. For example, you could check out all the bets we've received and payouts we've made. e.g. http://blockchain.info/address/1DXa2tKkoKiR1p7nbt2xSCFdUxzGd...

I'd encourage you to place a small 0.01 to 0.02 BTC bet just to test out the site, and see that it'll update the pool/odds dynamically.

How does the parimutuel betting work? If I bet now, am I guaranteed the "Betting X Pays Out" amount if I am right, or does that change with more bets on either side?

From the 'How It Works' page [1] - " During the betting round, payout odds will change depending on how many BTCs are placed on either side at any given time. Payout odds are finalized once the betting round closes."

[1] http://www.futureblock.com/how-it-works

Parimutuel betting is where all the bets are placed in a pool and when the pool closes we then distribute the winnings amongst the bets. [1] In fact, this is how a lot of horse racing works

This means you are not guarranteed betting X pays out Y. And this does change as bets are made on either side. And so one may think, why not just bet the last minute? House fees are initially 0% at 6 AM (PDT) but roll up linearly to 20% by 6 PM (PDT). This is to encourage betting early and discourage bet sniping at the last minute. [2]

Having said that, we do plan to move towards a floating model, more similar to the futures market as we gain transaction volume.

Current prediction targets are set by us. And bettor bet on either sides. As we gain liquidity and transaction volumes are increased, we plan to then turn this into a real time exchange so that the transactions are settled real time and we essentially have a moving prediction target.

In fact, this how we got our name at future representing futures market and block representing bit coin. :p

[1]: http://en.wikipedia.org/wiki/Parimutuel_betting [2]: http://www.futureblock.com/how-it-works [3]: http://en.wikipedia.org/wiki/Futures_exchange

Futures markets typically have value outside of speculation. Do you see any such value in a hash rate futures market?

Illegally bet on a possibly illegal currency the same week Silk road gets raided... What could possibly go wrong?

Except Bitcoin has not yet been declared illegal anywhere in the world and two of the strongest economies on Earth (USA and Germany) have recognised Bitcoin as some type of currency/commodity at least.

Good way for mining groups to make money during downtime: bet then go for maintenance / upgrades.

Is "Locksley" meant to be a play on Dread Pirate Roberts (characters played by Cary Elwes)?

What country are you operating from?

one thing I have learned. proving your legal is very, very expensive .

Reply on news.ycombinator.com