Topic: Any miner care to say why they're not using the X.IDIFF futures? (Read 2830 times)

Got it. Thanks, it was the auto exercise for coverage requirements, I failed to consider.

All extant IDIFF shares together will never pay out more than sum of all reserve requirements that were set aside when they got created. Hence the 0.13-0.18 estimated maximal cover/payout. To answer your concrete questions:

> a) when it settles, if difficulty is 20.8 mn, I would receive .208 BTC?
In theory yes, but actually they will get autoexercised much sooner when the rising difficulty meets maximal cover.

> b) when it settles, if difficulty is 14.8 mn, I would receive .148 BTC?
Again yes, but only if the cover per share is more than 0.148 BTC.

Can we try it again? I think your are talking about a reserve requirement necessary when selling X.IDIFF.JUN?

I'm asking what happens when you purchase a contract. I thought the contract settled on the third wednesday of the month for which they were listed?

So I still have my two questions.

No, the cover is actually lower and they expire before difficulty reaching 20m, but MP refused to disclose the exact value. Since it is a mix of futures created before March and after March, cover is somewhere between 2.9*2.9*January difficulty and 2.9*March difficulty. I don't have the numbers at hand but iirc it should be between 13 - 18 million satoshi. If anyone tries to sell you MPEx futures, ask him to run STATJSON and look for IMMCover value (and that's only for rough orientation, since it's averaged out over all people who created the futures).

If you're on Bitfunder, CoinBr.iDiff-E is covered up to 35 million, with same conditions/expiration date otherwise.

Got a clear "no" when I suggested that to him a while back. He just doesn't want to risk having unfunded coins in the wild. His policy is to even fund them if they get lost on the way from him to the reseller (the late funding is purely to ease potential customs problems). It's harsh, but I can understand that. He might change that policy, but currently that's the way it is.

Probably resellers should pool and make a "reseller insurance fund" to cover this huge risk.

Sorry for all the offtopic talk, btw.

Back to the X.IDIFF.JUN future.

I'm I correct in my understanding that I would purchase it today, for say 0.158 BTC.

Then:

a) when it settles, if difficulty is 20.8 mn, I would receive .208 BTC?

b) when it settles, if difficulty is 14.8 mn, I would receive .148 BTC?

Yes, it should work that way, you also should consider variance.

Just got an idea, maybe casascius should sell coins without balance, and they should be "charged" by dealers only after successful delivery?

Thanks for your answer, rini17.


I'm asking this whole question because I would like to calculate the cost of selling a casascius coin and I need to put a number on that risk.

If postal service insurance was for "the value of the shipment at the time they admit the loss" there would be no problem: just buy that insurance. But they only pay "the value of the shipment at the time of sending it", so I have to somehow cover the risk of bitcoin price fluctuations from the point I send the shipment to the point they admit they lost it. Otherwise I'm gambling and for a cost calculation I cannot use gambling.

Ok, it seems the options only help to solve a problem I don't actually have (reduce need for storing coins) and just add complexity. Let's say I set aside 5% (assuming max 5% get lost) of the amount I would have to reimburse the customer with for each shipment in BTC.

This simply means I can use 5% in my cost calculation, no? For example if all other cost added up (coins itself, 1 BTC loading value, shipping from casascius, packaging, shipping to customer, marketing, customer correspondence, ...) was for example 1.428 BTC, I could just add 5% cost for "insurance" and price would be 1.428 + 5% = 1.5 BTC, right?

In other words: I just put ~5% of each order into an "insurance fund".

Well, thanks a lot for helping me think. It seems it's simpler than I initially thought.

Of course, there is still risk: I could be unlucky and have 20% of the shipments lost (may some post office employee figures out what I'm sending and manages to steal many of my shipments). It seems I can't insure against this cost-effectively.

But then again: I could get run over by a bus tomorrow.

So, you sold physical coins worth X BTC in a month Y. Now you know you need C = 0.03*X bitcoins set aside to be able to cover for lost shipments in any case. You can stop here as a simplest solution, just put them to a cold storage and wait. Let's try to improve upon this, to not need to store whole 3%.

MPOE options can be had with 2 months expiration at most (if you buy next month options right after last Friday of the month). But I believe you can know after 2 months already approximately which shipments were undelivered, so use the options for first 2 months and then set C aside. Yes, calls struck at current rate will help with it, but they (next month on MPEx) usually cost around 0.5 BTC. In two months the outcome can be for example:

Bitcoin rate keeps stable or goes down - option will exercise worthless, you need to add whole C again. Total expense in bitcoins will be C*1.5 .

Bitcoin price rises twice - call options will repay itself. Since you got only 0.5*C from the options, you need to add another 0.5*C to have full cover, total cost is the same as if you put whole C aside upfront.

Bitcoin price rises 4x - call option repays 0.75*C, you need to add another 0.25*C, total expense was 0.75*C. So only in such case you are saving anything as compared to the cost to put whole C aside upfront.

This discussion would be perhaps fitting in MPOE bonds thread, as it illustrates how expensive the insurance against BTC rate movement already is, and to show to people aho want MPOE bot to raise prices that it would cause options to be useless.

Thanks for bumping this thread, an entertaining read in retrospect, as we are sitting at 8.9Mil difficulty today, BFL has not shipped yet (at least much of import), and price is quite interestingly different from the original posts time.

Guess the old axiom is true, "time will tell".

I know this is off-topic here but maybe not so much because I'm asking for help on how to use a financial instrument for a real purpose .

Problem is this: I'm sending casascius coins all over the world. Based on past performance of the postal service, out of 120 coins I sent (in 66 shipments) 4 got lost (in 2 shipments). That's 3% of the shipments and 3.3% of the coins.

I insured these shipments, but the postal service will only pay damage up to the value of the coins on the date of the shipment.

For one shipment of 3 coins I sent January 10th (and am only now going to receive compensation for the damage) this amounts to a whooping €46. To reacquire the coins I would now have to pay at least €350.

Now to remove that risk I could buy call option on BTC, right?

They would have to expire at least 5 months in the future and have current bitcoin price as strike price, right?

Looking at mpex.co, which option would I have to buy?

thanks in advance for anyones help.

For people who think 20th March is too far off, we have listed CoinBr.iDiff-E future that expires at 20th February on BitFunder. Discussion with more examples.

If you want to buy original MPEx futures and options without paying 30BTC for new account, CoinBr supports that already. We can also create the futures/options manually if asked, for you to sell short.

That may be right. I don't know much about mining but I'd have imagined that at current prices fpga mining is going to be efficient all the way to 8 digit difficulty.

On the other hand, 3mn corresponds to about 25Th. Any, literally ANY estimate of future hash output that I've read adds hundreds of Th to that figure, over the course of the next month or two. Now, these two can't be simultaneously true, that a. hash increases by 100 Th or more and b. diff stays under 10mn. Only one or the other. If general wisdom in the miner community is that Th will go up a lot, then these are a bargain. If the wisdom is these are overpriced, then necessarily the common wisdom is that not as much as 20Th will be added in the next two months, which literally means no asic deliveries at all. At all, zero, none.

Which isn't even all that improbable, actually. Just a case of the real consensus, as established by the actual people with skin in the game, being completely turned around by the fake consensus as established by the fake people with nothing better to do than blather on a forum/blogs/whatever. I guess this point is instructive, at least to me.


If this is the case you can either put a bid in there for less than 6-7 million (seeing how currently the only bid is for 4.3 million and just a few hundred at that) there should be plenty of space for you to maybe turn a profit. Conversely, you could create some futures out of your own capital and put in asks in, for less than the 8.x million currently asked. In either case you stand to profit handsomely (to the tune of 20-25% in BTC over two months) if you follow either of these and your predictions are correct.


I don't know. It'd seem that the 30 BTC fee is a moot point, seeing how there's 25 Th worth of mining which, at a conservative 1 BTC per Gh (which is less than what asics cost) still means a solid 25,000 BTC worth of equipment. It'd seem that tens of BTCs is not too much of a barrier to keep one away from insuring tens of thousands of BTCs worth of equipment, but maybe.

I don't think there are currently many miners who can profitably mine at a difficulty of 8.4 million (current price of X.IDIFF.MAR is 0.84 BTC).

I'm not sure the proposed hedge makes sensse for miners who ordered ASICS from suppliers and don't know when or wether they will receive the hardware, because it involves the risk in the case BFL doesn't deliver until march:

Let's say I'm in this situation: I ordered from BFL early on and I should be in first batch of BFL delivery (this is not really the case, I ordered but will not be in first batch). It is likely that avalon will deliver the first batch (18 TH?) until the March futures are due. That would bump difficulty to only roundabout 6-7 million... a loss if I buy some now for 0.084 BTC. So it doesn't make sense for the case of BFL not delivering until March.

In case BFL does deliver and I'm receiving the hardware roughly at the same time the other customers do, the hedge might make some sense (don't know how many TH BFLs first batch would account for, but including the avalon batch it would probably bump difficulty much higher than 8 million, so the hedge would probably make money).

So in that first case (BFL doesn't deliver until march) the miner isn't really a miner hedging but merely a speculator on difficulty... and maybe even a bad one because he speculated on difficulty >8.4 million, which might not happen in this case.

And in that second case (BFL does deliver), the miner really is a miner, but he will likely be profitable even at a difficulty of 20 million. So the added benefit from having hedged is merely quantitative.

Is it possible that this (together with the fact that there's a 30 BTC fee to get into MPEX) would explain current miners-to-be reluctance to buy the diff futures?

Maybe it makes more sense for BFL customers to hedge against non-delivery using for example http://bitbet.us/

Here's a bump, who knows, maybe someone actually says something?

I live to serve.


Sure, options been around for over a year now. I guess you missed all the other stuff too?

I've actually been debating offering BTC/difficult options and futures recently. I wasn't aware of any other source for them until now, so thanks. I'll be researching these later.

awesome... although I'm not a miner, I will read. EDIT: a very good read, explained so even I can understand. Thanks.

Through the magic of evil etc, your wish comes true.

Maybe the cross-section of "miners" and "people who are comfortable using financial instruments" is not as big as you think.

If that's indeed the problem it would probably help to explain in even simpler terms how a miner could hedge and what the benefits / costs would be,... maybe using an example.

Currently all that's available are BTC options. There used to be futures for a multitude of commodities (cattle, metals, indexes, etc) but these were discontinued for lack of trade volume.

This is more about difficulty futures however.

I'm curious.. where can one buy/sell futures on BTC (serious question here)?

Sigg

It'd seem irresponsible not to hedge if one's running a farm, and it'd also seem trivial to calculate both how much to sell and for what price based on the known operational costs etc. So I'm curious, what's the friction point?