Kinda. I trade, but I also have a "going down with the ship" position. That's my, let's say, "contribution to the network". I will happily sell the trading position depending on price. The other part is kept offline, not just for safety reasons, but also with the intention of this being a significant financial stake in this experiment even if it eventually turns out it'll fail. So, yeah, guess I'm trying to rationalize here how I'm both motivated by both greed/selfishness and idealism. Not sure if that makes any sense to the outside observer...

Nice stats. Thanks for compiling and posting, TheKoziTwo.

I noticed however that by the 'seniority' metric, Boolberry outranks even Monero (relative, not in absolute numbers). Just another reason for me to hold both :)

Eh, sorry to barge in on this, but... not really sorry.

What the hell, NotLambchop? We had a a few pretty good discussions, I have to say. You're bearish, quite a bit more than the average poster here. I appreciate that (I hate the filter bubble this place can sometimes be).

I also get the annoyance at the anti-trader attitude that some people ... no, cut that -  a vast number of people put up in here. That said, nothing of that applies to what empowering wrote. I enjoyed his posts on the last few pages - the (kinda rare in here) perspective of a long term bull/Bitcoin optimist that is not also part of the TO DA MOON NAO! crowd.

What I'm trying to say... don't know what crawled up your arse today, but you're a bear (by position, I mean) - the price action of the past week(s) should make you a lot more relaxed than you're acting in this discussion

Agreed with your way to interpret the scale logarithmically (as opposed to linear), but I think we are less than half way there even on a log scale. Probably less than half of the first half, so I say:

More than 1, less than 25 on a (log) scale from 1 to 100.

Hey

Great to see the latest updates.

Two requests:

1) (very minor) it seems that color choice (classic, thematic) isn't remembered after reload

2) (more interesting) any chance you can make graphs "vertically fill the space"? As in, if I close one of the graph windows, the remaining ones fill out the vertical space. Would be useful to see price or bid/ask graph in more (vertical) detail.

Example picture:

With one minor addition (underlined), I believe this is an excellent summary.

Plus, risk control. A trader without at least some basic understanding of risk control is just a gambler.

Fair enough. But keep in mind the distinction I tried to make: objectively, it is probably not a good store of value (currently). But economic preference is subjective to a degree - if a sufficient number of people elects something to be a store of value (despite there being better alternatives), then 'store of value' is a usage case.

That said, the resolve of those who use it as such is obviously tested during the ongoing bear market / correction, so we will see how many are really store of value users, and how many will turn out to be speculative investors that see their investment go underwater for a longer duration than they had ever thought possible, and sell in panic.

An incredibly complex question, not easy to answer at all, imo.

First, you already said this much, but I want to make it clear: the currently "fair" price is the price paid on the markets. Which includes speculative value. Which is totally fine. That's how markets work.

That said, you're asking about what value would be "fair" removing any projections by investors into the future. Fair enough question.

I claim we can break down non-speculative value into two components: usage for transactions, and usage as store of value. Both are difficult to estimate precisely, but at least the first one can perhaps be somewhat approximated.

We know Silk Road had about a million of revenue per month. There are/were other illegal usage cases, but I guess SR was the biggest one. As far as legal transactions go: There are those who claim the SR/illegal usage was the vast majority of usage, but I don't agree with this. Not anymore. But I also don't think legal usage is vastly higher. Overstock reports about 1 million in revenue after 2 months. Like I said, this is a lot of guesswork, but adding all of the data I heard of together, and even multiplying it by factor 2 or so for good measure, I don't see it higher than maybe 1 billion USD per year at peak. Assuming a high velocity of money of 10, spread over 10 million coins, that implies a transaction usage valuation of 10 USD. Assuming a more reasonable slower velocity of 5, 20 USD per coin. Being very generous and assuming actual floating coins are only half of those mined and not lost, 40 USD per coin.

Still far below current market, huh? So, is the entire difference the speculative value you want to remove from the equation?

No, because there's the "holders". Those that genuinely believe that Bitcoin is a great store of value, and who hold coins accordingly. And at this point, your question kind of falls apart and it becomes clear it can't be answered meaningfully: Those holders (by and large) believe that Bitcoin is a great store of value *now* (which means the non-speculative price of a coin that you ask for must include this store of value contribution).

However, if some oracle would be able to tell, without any remaining doubt, that Bitcoin won't survive the next 10 years (and will be replaced by another crypto, maybe), I am sure a lot of the holders would no longer want to use it as a store of value. In that sense, their usage of Bitcoin as a store of value is after all based on some speculative assumption: that Bitcoin will be around in 10 years from now, and will continue to function as it does now. But since this is somewhat speculative, this contribution should be removed from the value of a non-speculative coin price.

Assuming you'd want to include this contribution however, then we're faced with the problem how to calculate what amount of wealth (in USD) is stored in Bitcoin. That however is heavily dependent on current market price: an early adopter / investor who got his hands on 100k coins never put 100k*500USD= 50 million USD of his savings into Bitcoin, so it's a bit of a stretch to think that those 100k coins represent 50 million store of value usage. On the other hand, if he's not selling, then effectively he *is* keeping that amount of dollar in Bitcoin as a store of value.

My point is: you are asking the impossible. There is however a practical way to answer your question, imo - the non-speculative price is, approximately, the price that emerges after a sufficiently large market capitulation. So, by the current knowledge: 300s to 400s, probably.

Thanks for the clarification.

And I got your "5% of candles" comment as well now, I think. As in: if less than 5% of the candles are outside the bands, the bands are moving too slow.

I'm a bit slow on the uptake sometimes

Huh? I don't get it... here's the daily SMA200 with bands at 1 standard deviation above and below plotted. Or, in other words, Bollinger Bands set to 200.

Mid August and now has daily candles close below the lower band. My point was, that's the first time this has happened since 2011.

(And I'm pretty sure of that, went back to look at the old mtgox data to confirm.)


So, I'm not sure what you meant: I am wrong with my claim ("first time since 2011"), or do you mean I'm using too slow settings?

Nevermind. Pretty sure you meant "Don't rely on BB200, it's too slow for BTC." Correct?

Just a quick remark re: operating costs.

What do you think the wage is these guys get paid?



living in here when they're on duty



from this article: http://www.thecoinsman.com/2014/08/bitcoin/inside-chinese-bitcoin-mine/


I get the feeling to accurately estimate the lowest cost to produce coins, we need to get prices found in the US/Europe out of our head, and instead look at the Chinese market instead.

Hm. Not too happy about the way my post was understood then. I probably should have phrased it more clearly...

Like I said in my calculation post, I don't want to claim this is the correct lowest production cost per coin.

What I intend it to be was a "model" calculation -  to show how, with a relatively conservative estimates on the discounts a large mining operator could get, mining could still be profitable assuming those rebates, even though mining for "amateurs" (who don't get those rebates) is probably not profitable at the moment.

As an example, say the large miners only can eek out a 10% discount for the hardware (vs. the 25% I assumed), and their overhead cost (facilities to run miners, cooling, etc) is higher than I think, in which case even the large miners don't mine profitable at the moment. That is the point CMMPro is making (correct me please if I'm wrong here, CMMPro).

My point is: I believe it is at the very least possible that large miners get substantial enough rebates on hardware, and are efficient enough in operating, that they can still turn a profit. Especially considering that large investments continue being made in mining farms.


To condense the point I tried to make even further...

Group 1: "Bitcoin mining is definitely NOT profitable at current prices!"

Group 2: "Bitcoin mining IS definitely profitable at current prices."

My position: "I can't say whether it's profitable for sure, because I don't know the discount factor of large farms, but large scale mining might be profitable, even at current prices, if those rebates are high enough."

I agreed with that (the reduced margins) in my reply to cow-herder. Just as an aside, a sufficiently large farm won't need a pool, so not necessarily needs to pay a pool fee.

Anyway, in reality, I'm sure the large operators' calculations are a lot more complex (since they presumably replace outdated hardware on the go, and need to take into account how much of their coins they can sell without depressing price too much)

But my point remains that, assuming you will be able to sell over the next 200 days for a price of around $500 per coin, investing $459 doesn't seem such a bad deal to me.  This would be a ~9% profit, that's hardly "tiny". (EDIT: at $500 per coin, that is. Profit would have been a lot higher at previous prices. Point is that, even now, there'd be 9% profit)

Now we're talking. I don't think the margins are as good as they used to be for the big operators (that part is obvious, given lower price and higher difficulty), and are maybe even getting close to a net zero.

I have the suspicion (that I can't prove) that we are currently pretty close to the cost of a coin mined by a large farm. I don't think it is coincidence that the mid to high 400s are revisited so often and persistently. But it's nothing more than a suspicion, I admit. Maybe post hoc reasoning to explain why price keeps coming back to where we are now.

You're re-doing my calculation from above, but you let the hypothetical S3 run past its profitable range wrt energy.

My cutoff was ~200 days, in which case the total cost to produce 1 BTC with a 25% off S3 is $459 (this is the price per 1 BTC - so you will need more than 1 S3 running to get 1 full BTC after 199 days, to be clear).

Tell me again how this is unprofitable?

I don't know how you got your numbers. Let's base the calculation on Antminer's S3. Specs used for calculations as taken from their site, see here.

Let's assume difficulty increase of 15% per adjustment, and as energy cost: $0.05 per kWh.

If mining starts now, then after 198 days, the unit will stop mining at a profit based on ongoing energy costs (assuming $500 per BTC market price), and will have generated a total of 0.64 BTC during its (profitable) lifetime.


Now, given the raw numbers so far, we get:

(((0.58*480)+(198*0.43)))/0.64 ~= $568 production cost per 1 BTC.

Which is indeed higher than current market price, and suggests mining isn't profitable anymore.


However, keep in mind the point I've made several times by now: amateur mining is probably unprofitable. It's the professional, large scale mining that is interesting, under the assumption of reduced hardware and energy costs.

So, I'll make one more assumption: Let's say Antminer has a 25% profit margin on their hardware going for consumer prices, that a large miner (maybe Antminer running a farm themselves) is able to deduct.

You can reject this as pure conjecture, please feel free to do so. I however consider this to be a rather conservative estimate. In this case, we get:


(((0.58*0.75*480)+(198*0.43)))/0.64 ~= $459 per 1 BTC.


Now, to get back to the point I've been trying to make over and over again:

There are several unknown variables that apply to large scale mining, such as: cost of operation (manpower, rent, cooling, setup, organizational overhead), cost of acquisition of hardware, and finally: energy cost. For that reason, I don't make any claim that the above value (of $459) is "the correct" production cost per coin of a large mining operation.

My goal in the above calculation is simply to point out, that, given the relatively conservative assumption that a large mining farm can acquire S3s for 75% of the consumer price, and make use of electricity for $0.05 per kWh, the production cost (taking into account hardware cost and energy only) is below current market, i.o.w. mining would be profitable for such an operation.

Yes, I left the delay part out because I wasn't sure if we had the same idea about this mechanism. So, agreed on your point. However (and correct me if I'm wrong), eventually, difficulty will need to decline (or at least rate of growth will) if mining is unprofitable - together with assumption of course that miners are, on average, ecomomically rational.

What I mean is: up to now, it is possible that we don't see the effects of unprofitable mining in difficulty because of the delay in the feedback loop, but the longer there is no effect on the difficulty, the less likely it becomes that mining is unprofitable (unless we assume a growing delay, which I don't find plausible.)

Correct?

Probably true for the majority of amateur miners.

Almost certainly not true for the professional mining farms.

Yes

But you're calculating now (average)  profit of miners, while I talked about _cost_ per coin.

I never said price (market price, that is) can never fall below production cost, which is what you have in mind maybe. That said, I believe production cost is a price attractor, and we won't fall below it unless extremely strong momentum brings us there. If price would fall below production cost for a longer period, we would certainly see the effect of that in hashrate / difficulty, but we don't (yet).