They're not really borrowing from anyone - they never get access to the borrowed coins - they're only allowed to invest them, and only for as long as they are able to pay back the loan - so there's never any chance that they will default on the loan, and so there's never a need to hold any collateral.

In effect the lending means the site is operating a fractional reserve operation, which many will dislike the thought of. But I think in this case it is sound to do so. There's no chance of a "run on the bank" causing problems, since all loans are automatically secured by restricting what the lender can do with their loaned coins.

Sudden large changes in the maximum profit per bet would probably result from this feature, and were another reason I didn't love the idea.

When you say it could "go bad fast", mostly that only applies to people who over-borrow, don't you think? It wouldn't go bad for the site itself.

One thing that was pointed out when I brought this up in the past is that investing at huge leverage is effectively a big gamble. No sane investor would do it, but guess who would... that's right - gamblers would! It's a way of gambling with high variance and a positive expectation. So maybe offering such a feature would turn the whale gamblers into whale investors, and we'd be left without any whale gamblers.

I still think it's an interesting idea though.

The site looks good, is there any way to try it without invest?

I've been watching the house profit for this site and here's my take:

I used to invest in JD, but with DiceBitco.in, as dooglus pointed out, the 1% bankroll wager is crazy and I'm not inclined to invest in this new site just for that reason alone.

In any case, reducing the bankroll wager down from its current 1% would attract more prospective investors which would actually increase the max bet far more in the long run.

... And if you could find a way to "Randomize Roll" before any turn while solving potential DDOS issues, I would definitely be a regular on your site.

Did I say that? Whenever the site owners have asked me about whether they should reduce the max profit, I've recommended they leave it at 1% for now. Plenty of investors are happy to go with full Kelly I think.


I'm not sure a lower risk factor would attract more investors. I'm pretty sure the bottleneck at the moment isn't "whoa, 1%?", it's more likely to be "who are these guys? I've never heard of them!"

You can randomize before your 1st roll, and then after every 3 rolls. At JD you could only randomize every 10 rolls.

It doesn't matter, because you get to pick the client seed after the site picks the server seed, so all the rolls are decided by YOU, not by the site.

But tbh i never change seeds , i just trust the fairness of pd. I can afford that coz i only play for fun with rly small amounts that i can afford to lose any time .

I'm assuming you're referring to PD3. From my understanding of PD3's hashing algorithm and personal verification, all future rolls are pre-determined after you've accepted the server seed and input your client seed. This holds true for DiceBitco.in too (I think).

The server seed is verifiably unchanged until you choose to change it before any given turn. Your seed is also unchanged unless requested otherwise. The only hashable element that change from bet to bet is the incremental nonce, which is sequential.

For each roll, all three of these elements are hashed together to produce statistically random results.

Here is PD3's provably fair page:
http://primedice.com/#/verify

Due to the laws of large numbers, it is almost inevitable that a losing streak will match your bankroll given a sufficiently large n of rolls.

How u r randomizing the seed, mt_rand() ? Please say it only if it is disclosable.

The "Login" button is very confusing if you are already logged in. It might be smart to change it to: "Switch account" or something like that.

Perhaps you should create two different dice games, and investors could choose which one (bankroll) to back? For example one dice game with a max bet of 0.5% of the bankroll, and the other 1%? Although it might defeat the purpose because the bankroll would be smaller each in both cases than just one.