Huzzah!  /me continues cautious optimism

Yeah, I'm in this boat too (not that anyone asked for my opinion).  "Cautiously optimistic" is me right now.  If you have lost confidence in Garr, or Cointerra, it's pretty hard to come up with a reason not to sell now while you can.  (you might not make a BTC profit, but if you bought at 0.25BTC/share when BTC was $100 and sold at 0.05 with BTC at $900 that's still 55% net-value-profit, less than you would have made holding raw BTC but not a loss!)  On the other hand, if you think there is even a reasonable chance that Garr and cointerra will deliver (and I think there is a reasonable chance) depending on how deep your position is, hold or buy are very reasonable responses right now IMO.

I currently hold 100 shares + 6 mining contracts for a total exposure of 220 shares post-cointerra-delivery, I don't mind admitting, and my average share price is probably 0.13BTC/share including the contracts.  That's just a guess though, I have to go review my histories to get real numbers.

EDIT:  even if the market depth was there, presume I could sell all my holdings for a net of 11BTC, or $9900 at today's exchange rates.  This is probably a very small profit for me given the BTC/share cost and $/BTC cost when I bought in.  Following Jeff Bezos' "regret minimization framework", if cog goes to shit and is worth zero, I lose 9k.  If cog rebounds to 0.25BTC/share, and BTC stays around $1k, I stand to gain $55,000.  For me, I would regret selling if it did well WAY more than not selling and losing my initial investment.  Regret minimization is the way to be a happy investor.

Why was giving money to cointerra a terrible mistake?  Isn't that not true unless they actually fail delivery?  Regardless of whether or not hashfast delivers in dec/jan, that network size increase is taken into account in my projections (hence low, middle, and high being over 50% apart from each other).  If cointerra delivers what they promised, COG shareholders will make money and COG shares will be worth more than they are now, it's pretty much 100% up to cointerra (and Garrett running COG well and not screwing up).  If any one company released enough hardware at once to impact the network by THAT much it would have other impacts to the price and confidence of bitcoin itself, so we can (hopefully) presume none would do that.

...ok. that's fair.  Lemme *slightly* modify that statement.  "I don't think cognitive investors that bought cognitive when garr was talking about our upcoming cointerra deal have any reason to trust cointerra less now than they did at that time."  The cointerra "equation" hasn't really changed - but the impression of COG itself probably has.  I merely mean to say it is the most likely explanation, rather that worries about cointerra screwing COG.

If you disagree, however, perhaps there is worrying news from cointerra I am not aware of?  It is pretty easy to miss things like that, people don't like to talk about bad news.

Finally, for those interested in the time between now and then, eagerly watching the new transparent deposit addresses, here are the expectations:

assuming ~870GH/s, diff 908mil, 2% pool fees, we get:

 1/(908.0*1000*1000*(2**32)/(870*1000*1000*1000)/60/60/24/7) blocks/week *25btc/block *0.98profit => 3.306 BTC/week expected. Since 50% goes to growth fund, that is 1.65BTC/week, or 0.000159 BTC/share (assuming 10420 shares).

At the previous diff, the calculation was 0.000204, and the actual div paid last week was 0.000242, so presumably we had good luck or my GH estimation is low.  I still don't know of a good way to calculate probability distributions to say how likely it is our diff is 20% higher or 50% lower than expected over 1 week, 1 month, etc. but I am looking into it.

EDIT: Oh, the other explanation for 0.000242 instead of 0.000204 is maybe garr didn't pay divs to unclaimed shares and let the claimed ones absorb the extra, if there were ~8800 shares the expectation would be 0.000241, or maybe it is a combination of ~9500 shares and also some good luck...

Just to re-analyze the diff numbers...

Previously, I predicted:
Difficulty at the end of January (projected) => 268mil * ( 1.41**8 ) => 4187mil

Now it is 908mil @ Dec 13th.  The last diff interval was 12.6 days, so lets call it 12 for fudge factor.  The last diff increase was 28.4%.

EDIT: I misread the date on my watch, LOL, it's the 10th, but shifting everything back 3 days doesn't impact anything significantly.

Dec 13 => 25 => Jan 6th => 18th => 30th

Using the last diff increase as a lower bound, 35% as a middle, and 50% as an upper bound, we get:

Difficulty at the end of January (low projected) => 908mil * ( 1.284**4 ) => 2468mil
Difficulty at the end of January (middle projected) => 908mil * ( 1.35**4 ) => 3015mil
Difficulty at the end of January (pessimistic projected) => 908mil * ( 1.50**4 ) => 4596mil

Keep in mind 50% per difficulty adjustment means the network increases over 225% per month, which is probably (i hope) not going to happen even with cointerra, knc, and VMC dumping their hardware in January.

I now consider the end-of-January diff being under 4 billion to be a "somewhat safe bet", and under 3 billion to be "possible but unlikely".  Time will tell how accurate my projections are.

keep in mind that if we receive our hardware at 4.2bil, I calculated, cog will be earning ~24.38BTC/week which means first week divs should be 0.001BTC/week per share (at 26420 shares, presuming that COG.F and COG.F2 convert).  This historically, mining shares have been worth a valuation such that their instantaneous annualized return is between 20 and 30%.

EDIT2: To clarify, "instantaneous annualized return" is a metric I birthed straight from my asshole.  But I think it is instructive, since actual annualized return estimates are complete bullshit when the "coins produced" decrease by 30% every 12 days unless the corp is constantly buying hardware.

EDIT: to demonstrate what I mean, AsicMiner is currently 0.31BTC/share and the last div was 0.00199, 0.00199*52/0.31 = 33% instantaneous annualized yield.

0.001*52/0.20 => 0.26BTC/share.
0.001*52/0.30 => 0.173BTC/share

EDIT: Current analysis for COG is 0.05BTC/share, last div is 0.000242, 0.000242*52/0.05 => 25.2% instantaneous annualized yield, but that doesn't take into account the hashing power that arrives in January.

It seems the only reasonable explanation for COG's current very-low share price is uncertainty and lack of trust/transparency.  I don't think people distrust cointerra.

Thanks pascal, please share your findings (and let us know if you need any help analyzing the results - hopefully we can figure out whatever we need from the public blockchain once we have an address to start at).

Garrett,

I really don't want to pull your obviously-limited attention away from future profitable endeavors (like setting up the new mining addresses, ensuring things are ready for our hardware in January, etc) but please don't underestimate the importance of investor confidence and transparency into these past events.  What happened has happened, and no matter WHY it happened, knowing is better for investors than the current uncertainty.

If deposits were made directly to BTCT and/or Havelock, they must have still been made to one or more BTC addresses.  Could you tell us those addresses?  Then, with no further action on your part, investors could ask burnside/havelock to verify, or try to trace the flow of BTC to see if they end up in btct/havelock.  It isn't perfect, but it would at least give us some evidence instead of being left to wonder if it is the worst...

Thanks for your work, Garrett!

I'm usually defending Garrett, I consider myself the 'voice of reason', and given the few talks I've had with him I feel like he's a pretty cool dude. That said...

1. The poor communication is not cool.  However, it's been said already, either Garrett will take it to heart, or ignore it to the determent of COG.  I hope the former.
2. If mistakes were made, a pool ripped us off or malfunctioned (a pool having luck that bad for that long is highly unlikely - technical problems or scam are more likely explanations) then fine - tell us that.  But without understanding the cause of the under-expectations returns, we have no way to know if it was a fluke, or something that impacts future returns as well...
3. Yelling at Garrett, Threatening Garrett, Threatening to sell COG, etc are all unproductive.
4. If available, data about our payout address / mining details would be super nice to help substantiate and understand what happened over the last 2 or so months
5. Moving forward, having a non-public payout address is simply not acceptable.  I can't think of a single drawback and it has every advantage to address the naysayers and add much needed transparency.

Occasional stints of poor communication aside, you've done a fine job running COG, Garret, I look forward to you addressing these issues.

Thanks!

(Sidenote:  If the expected return is X, and the actual return is Y, given the hashrate, difficulty, etc., there is a 50% chance X>Y and a 50% chance X<Y.  Furthermore, one could calculate the probability Y < 0.95X, Y < 0.90X, Y < 0.85X, etc etc.  I don't have the maths handy, but my intuition is that the probability Y < 0.5X over a period of 6 weeks is vanishingly small unless a malfunction or scam took place. For example, a pool with high latency will reject more shares and/or have more blocks rejected, hurting payouts by a predictable amount.  Maybe someone should run the math).

Does it just outbid you by 1 satoshi or something?

This is why the acceptable number of bots to have running are 0 or 2+.  Someone should run another one =)

I stand by my prior maths/assumptions.  The value of COG is primarily in the BTC that it can produce, and crazy fluctuations in the value of BTC doesn't really change that.  (you but X shares with Y BTC and earn Z btc dividends, regardless of what BTC is worth).  The main thing that has changed is X BTC buys more hardware than it used to since generally bitcoin miners are built using supply chains and workers paid in USD.  This means the BTC our upcoming 28TH miners earn can buy that much more hardware in the future if the prices stay high.

The current dips are due to people profit-taking, their getting out (possibly at a profit despite the lower price, due to BTC value increase) is your cheap entry to future profits.  I see no reason not to predict COG shares being worth between 0.25 to 0.5 BTC per share when our hardware arrives (and possibly even more, depending on demand, difficulty trends, and other market conditions)

at long last, we can wank about the future value of COG again =)

Most of my past maths haven't really changed, but the rate of network increase has slightly leveled off over the last week or two.  I definitely don't think cog will be 2-4BTC like AsicMiner was at any time (unless we actually make the 1PH/s thing happen before anyone else, which is possible and AWESOME but I wouldn't bet on it).

That said, I think a really reasonable valuation could be 1BTC per share if we get our 28TH/s before the end of January and hear some really solid details on the next order as well (confidence we will be in the 200+ TH/s range, for example).  This is also contingent on cointerra not flooding the market TOO too much (just giving it all to us =). Obviously, the harder it is for the "average joe" to get CT equipment, the more valuable Garr's carefully constructed bulk pricing deals are =)

EDIT:  and in this context, when I say "leveled off" I mean "went from insanely exponential to just a little exponential".

This is awesome - I presume if you already have a havelock acct that has the same email address it will "just work"?  And if the email addresses are different youcan just transfer the shares from the 2nd account to your "main" one?

Seeing as how investr has proposed selling at 0.25BTC, and I have proposed buying at 0.15BTC, you can be reasonably sure it will be somewhere between those two, at least initially.  Where it goes from there depends completely on demand =)

Yeah, I was going to say we'd need Garr's buy-in before anything official.  But then again, from his perspective, if things are going to be in an exchange in a few days or something, he isn't going to want to bother.  Also, I'd rather he spend his time on the transition (presuming there are things he can do to make progress) rather than manually transferring shares.  I wish we had an idea if the delay is more likely to be days, weeks, or months.

I'll buy 10@0.15.  PM me.

well...I mean, you'll have to fight me for the cheap ones when everyone dumps. =)  I think 0.35 is a pretty fair price given the circumstances (no other way to buy right now) but I think they will go lower once they show up on an exchange and I think they will be worth more than that right after we receive our hardware in Dec./January.

You'll have to fight me for 'em =D

Thanks for the update, Garr.  The delays are unfortunate, but I am glad divs are starting back up.  That should instill confidence that COG is still up and running.

Set the address to an address you control, then garr can later ask you to prove you own the address by signing a message, and he can pay dividends to the address in the meantime.  Don't use an address you don't control (like a desposit address from coinbase, etc).