Thursday, March 20, 2014

What Not to Mine? (aka How to Spot a Shady Cryptocurrency)

For those that are curious, here's the sort of thing I write about in my newsletter. There are charts showing current profitability and other recommendations as well, but I also provide some thoughts on the market as a whole. Rather than only emailing this to my subscribers, however, I'm going to post this on the blog. If you like this sort of thinking and analysis, please subscribe -- I'm still only asking the equivalent of $5 for two days, $10 for a week, and $30 for a month. Details are in this post. And with that said, let's talk about what not to mine right now...yes, it could be a big list, but I'll just pick on a couple options.

It shouldn't be too surprising that there are a lot of "scam" coins going around. The latest two I want to discuss are PANDA and 10-5 (with 10-5 having some of the same core developers as PANDA). With PANDA, "wolong" has apparently disappeared after trading in his 2.5% premine of PANDA at higher prices -- after promising to stop that sort of manipulation -- and the devs "learned their lesson" and are making the coin they really wanted to make this round. PANDA smelled fishy from the start, but I still mined some (and they're now basically worthless). Fine.

Here's the problem with TenFive/10-5: like PANDA, this is a premined coin. It's only 1% premine this time, but there's more to it than that. The block rewards of 10-5 are small, so the 105,000 premined coins would actually take a bit of time to mine, according to the original specifications -- around five days before the end of the initial month after launch. But now, the specifications changed after the coin launched! The original specs called for the following reward structure:
The "Flat" Reward Scheme designed to reward pi (really!?) for the majority duration:
  • First 1050 blocks: Payout of 1.57079632679 per block
  • Next 3150 blocks: Payout of 3.14159265359 per block
  • Next 6300 blocks: Payout of 9.8596 per block
  • Next 9450 blocks: Payout of 19.7192 per block
  • Next 18900 blocks: Payout of 25 per block
  • Next 37800 blocks: Payout of 50 per block
  • Next 75600 blocks: Payout of 25 per block
  • Next 151200 blocks: Payout of 19.7192 per block
  • Remaining blocks: Payout of 9.8596
Now, there are reasons to change from that reward structure -- for instance, at that rate of mining, pi isn't really the reward much at all, and the coin would be mined out in about three years (give or take). But there's a less benevolent reason to change the rewards to the following:
The "Flat" Reward Scheme designed to reward pi for the majority duration:
  • First 1050 blocks: Payout of 1.57079632679 per block
  • Next 3150 blocks: Payout of 3.14159265359 per block
  • Next 4000 blocks: Payout of 9.42477796077 per block
  • Next 5150 blocks: Payout of 3.14159265359 per block
  • Final value of 3.14159265359 per block for the remaining blocks
First, they estimate 2150 as the year when the coins would all be mined right now. Well, I can do math better than that! A block of Pi coins every 1.5 minutes means people will mine ~3015 coins per day, 84446 coins every four weeks, and 1101568 coins per year. At that rate, it would take 9.53 years to mine out TenFive. So if they want to mine until 2150, they need to up the supply to around 150,000,000 coins (oh, but that's not "10-5" so that would be bad... maybe 105,000,000 then?) The bigger issue however is this clause regarding the premine:
Stipulation: 4-week mandatory share lockdown. After lockdown period only a % based on quarterly performance reports will be paid out.
Have you figured things out yet? By changing the rewards structure to pay out fewer coins, in the first month before they dump their 105K coins on any willing suckers, they have reduced the supply from roughly 400K coins plus the premine to just 84K mined coins plus the premine. Fewer coins in circulation for the first 28 days may promote higher exchange prices, and then BOOM! "Hey, who just dumped 105,000 coins on the exchanges!?" Also worth noting is that we're now on blocks in the 6000+ range, and the reward is still at pi, so we didn't even get the 4000 blocks worth ~3X pi.

I'm becoming increasingly convinced that premines are a clear sign that a coin is going to go nowhere but down over time. I don't care whether the coins are for the "hard working developers" (that had to take someone else's code, change a few variables, and recompile -- oh my! Five hours for a good developer could accomplish that!), "marketing efforts" (e.g. get it listed on an exchange, get other suckers to jump on, etc.), or bounties for people to make pools and such; the end result is the same: a big dump of coins early in the life of the cryptocurrency that will plunge the price down to places where it may not recover.

Someone asked me recently if I had ever tried any of the cryptocurrency IPOs. The answer is "no", and the more I think about it, the more I'm convinced I never will. There's just too much chance for abuse, and I don't want a part of it. And just to be clear, there are developers who really put in some time and effort. I think the Darkcoin guys have done a lot of good work, for example, and even things like Heavycoin and Fuguecoin at least require them to do something other than clone a github repository. But if you're not adding something truly new and complex, it's really not that much work to get a cryptocurrency started.

Now if you'll pardon me, I'm off to create my own cryptocurrency. I'll only do a 0.5% premine for my efforts -- 1 billion coin total supply, so I'll just take 500K. The first month of blocks will be to help set the difficulty, with low block rewards. Then the rewards will start to ramp up -- just in time for my huge dump! I'll borrow ideas from several other popular coins, we can all mine it for a month, and then when things are starting to look up I'll cash in and retire...or go create my next cryptocurrency. Sound familiar?

But everyone is doing these premines, and the developers really do need to be paid, right? Well, no, not really they don't -- not like this. It's no wonder the alt-coin market is tanking, as hundreds of new clones have come out and very few of them have any real worth -- other than for early pump and dumps. We even have websites that will adjust scrypt for you and help you create your own coin -- for a fee, of course. Which coins will actually survive the test of time? Besides BTC and LTC, well, it's a bit difficult to say for certain! I have my bets, and some will certainly be wrong. So bet it. If you want to know more, just send me a subscription request.

Good luck to you all as we mine our hearts out.

3 comments:

  1. Well said. Now if only people would stop *paying* for premined coins.. All these investors with lots of BTC are scared of altcoin investing because they've been scammed so many times.

    ReplyDelete
  2. I wouldn't have a problem per say with developers getting a few quid for making a coin. No, it's a free market economy if they can sucker a few people into parting with some BTC for a coin that essentially worthless - kudos to them. It's an obvious pump and dump and if I had time to scan the codebase I'd give it a lash myself.

    What I *do* have issue with though is that each and every new coin erodes the market cap (for a lack of a better word) of the whole ecosystem. One coin in itself is relatively benign, but 5-10 coins a week and people are losing confidence in coins which should in theory have some worth such as Dogecoin, Vertcoin etc.

    Before Christmas I was on about 30 USD a day profit with my 4Mh/s, now with 7-8 Mh/s I'm on about €10 USD profit. Doesn't take a genius to see where this is going. You might say, "Ahh, but that's cause the difficulty is going up"....and in some essence that's true, but I think it's because people are just losing faith. First we had Cute advertising (dogecoin, kitty coin) then it was Coins linked to celebs (Coinye coin, etc - what ever happened to Norris coin BTW :-)

    Then it was coins with funny connections, "Dope", "HoboNickels" coin.....now the current flavour of the month is conis linked to countries. Polandcoin, SpainCoin, etc.

    What's wonders await us next :-)

    ReplyDelete
  3. I couldn't agree more with your entire post. The deeper I got into alt coins, the more I began to realize how little most of the "developers" actually knew about anything. Even ones that seemed legit were more easily seen as either inept or con men after my senses were honed to the BS a bit better.

    I completely agree about Darkcoin, by the way. That guy is actually doing some real work and trying to add features to a coin instead of following the typical altcoin playbook - create coin, get fake or irrelevant vendors, get copy and past dice game, wait for others to see the glory in your unique reward structure (because that is what gives a coin its value??).

    With the 250 MHz KNC scrypt miners showing up in a few months, I wouldn't buy anything but Litecoin or maybe Doge in the scrypt altcoin market today. Maybe gpu miners will move to Vertcoin or Darkcoin, but I don't know if anything will survive the coming slaughter.

    ReplyDelete