Pool hopping... ethical or not?

[organofcorti]

Yes, the story of the laborers in the vineyard was about how the workers who worked longer (all day long) were envious because those who worked only 1 hour got the same unit of pay.

But that extra pay was from the vineyard owner's generosity -- that was the point. "Are you envious, because I am generous?"

The extra pay did NOT come from the other laborers, or they would have been an injustice committed, and they would have had every reason to complain.

Yep. Change the story a little bit:

A vineyard owner needs some vines cleared out. He asks some guys if they'd like the job. They ask how much it pays. The owner says, "I'm willing to pay $30 to get it done." The two guys agree to do the job. A few hours later, the owner decides he wants the job done faster, so he adds a third guy to the team, and then insists that the two guys who originally agreed to do the work accept $10 each, even though they did more than 1/3 of the $30 job.

Now that the person getting paid the same but working less has his bonus coming right out of the other guys paychecks, it seems a bit different, doesn't it?

I've been explaining pool hopping on multiple threads, bragging of my efficiencies, telling miners at pool threads that they're losing out because I'm pool hopping. I've tried really hard (in some cases successfully) to get full time miners to get their pools to change.

I've helped pools change and for brand new prop pools pointed out the importance of not being on prop because it will cost full time miners. I've posted this on their public threads where all their miners can read it.

If someone reads a post like that from me and doesn't care, then he is giving tacit but knowing approval for pool hopping.

OMG!!! Are they implicitly agreeing to let you pool-hop? Well then, you have got youself an implied agreement!!
Oh wait, that's circular, isn't it? Crap... Opps...

sorry not getting the circularity.

1. I'm telling them I hop.
2. I'm telling them it costs them if I hop.
3. I suggest they tell the pool to be hopper proof.
4. They don't do anything about it.

If there are no rules against it and if full time miners don't mind what I do after explaining the situation to them, then how can what I do be unethical.

I hope I've explained it better this time, and less circularly.

edit: Thanks bb. I always did want to be absolute. And circular, i guess.

[1715331700]

1715331700

[1715331700]

1715331700

[1715331700]

1715331700

[1715331700]

1715331700

[CanaryInTheMine]

sorry not getting the circularity.

Was being a bit sarcastic.  
Making fun of this mystical implied agreement out there: one "against" pool hopping and you obtaining one "for" pool hopping.

[Desolator]

I read something about this but didn't get it.  Each individual miner or even calculation has the exact same probability of forming a block as anyone else's.  There is no progress either.  How would switching to a different pool help?  You can't just jump in and say oh well my pool just finished so I'm gonna get on this one cuz it's close to being done.  It doesn't work like that.  Where's the math to back this up as being advantageous?

[organofcorti]

I read something about this but didn't get it.  Each individual miner or even calculation has the exact same probability of forming a block as anyone else's.  There is no progress either.  How would switching to a different pool help?  You can't just jump in and say oh well my pool just finished so I'm gonna get on this one cuz it's close to being done.  It doesn't work like that.  Where's the math to back this up as being advantageous?

Here's one link. I'll leave it to you to find the rest.

https://forum.bitcoin.org/index.php?topic=3165.0

[JoelKatz]

What is the implied agreement here?  I bet you for every miner you will have another version.  If 100% of pool users don't agree with it, then there is no such agreement.  only what is present here, debates about it.  As opposed to a pool operator creating an agreement for all to abide by...

The implied agreement is, as it always is, that all miners will interact in good faith with each other and deal fairly with each other. This is understood to be an implied portion of any agreement.
http://en.wikipedia.org/wiki/Implied_covenant_of_good_faith_and_fair_dealing

Show me someone who understands what pool hopping is who insists that it consists of interacting fairly with the other miners and contributing in good faith to their cooperative effort and I'll show you a liar.

[JoelKatz]

I read something about this but didn't get it.  Each individual miner or even calculation has the exact same probability of forming a block as anyone else's.  There is no progress either.  How would switching to a different pool help?  You can't just jump in and say oh well my pool just finished so I'm gonna get on this one cuz it's close to being done.  It doesn't work like that.  Where's the math to back this up as being advantageous?

If the pool pays out based on the number of shares accumulated prior to finding a block, once the pool has accumulated a certain number of shares towards finding a block without finding one, a pool hopper can be sure that any additional share he submits before the pool finds a block will yield him a below-average payout.

[Sukrim]

I read something about this but didn't get it.  Each individual miner or even calculation has the exact same probability of forming a block as anyone else's.  There is no progress either.  How would switching to a different pool help?  You can't just jump in and say oh well my pool just finished so I'm gonna get on this one cuz it's close to being done.  It doesn't work like that.  Where's the math to back this up as being advantageous?

If the pool pays out based on the number of shares accumulated prior to finding a block, once the pool has accumulated a certain number of shares towards finding a block without finding one, every miner in this pool can be sure that any additional share he submits before the pool finds a block will yield him a below-average payout.

/fixed

Some are stupid enough to stay in the pool though and then complain that others aren't.

[EskimoBob]

Do you feel like crap after pool hopping?
Do you feel an evil joy from pool hopping?
Do you feel cheated, my dear honest miner?
Do you feel like your pool got abused (or did you do some selfabusing in the pool... oops)?

No problem my dear friend, let me fix that for you!

Transfer some coin to: 1Kae8NWf1pKsadDu53E9GeRo6pN9c4hrgy and you will feel better. Instantly!
 
PS! If you do not feel better after the transfer, you did not transfer enough bitcoins. Transfer more and see what happens.

BTW, think happy thoughts while you send me your bitcoin! 

[organofcorti]

Do you feel like crap after pool hopping?
Do you feel an evil joy from pool hopping?
Do you feel cheated, my dear honest miner?
Do you feel like your pool got abused (or did you do some selfabusing in the pool... oops)?

No problem my dear friend, let me fix that for you!

Transfer some coin to: 1Kae8NWf1pKsadDu53E9GeRo6pN9c4hrgy and you will feel better. Instantly!
 
PS! If you do not feel better after the transfer, you did not transfer enough bitcoins. Transfer more and see what happens.

BTW, think happy thoughts while you send me your bitcoin! 

Post like this make me want to spend time finding a way to deduct funds from addresses. 

Why? because eskimo bob came up with a way to try to deduct funds from addresses first?

[Meni Rosenfeld]

Pooled mining is an agreement among miners to jointly work on finding blocks and splitting the rewards in proportion to the contribution of each, with the pool operator as middleman. Pool-hopping is an attempt to gain more than the fair share of a pool's rewards for the work done, breaching the agreement, and as such is unethical.

The claim that "people were warned and still mine for proportional pools, so they consent to it" is understandable, but problematic because:
1. Not everyone reads the forum. I wouldn't be surprised if most of the users of Deepbit have never looked at its forum thread. You are taking advantage of those users who don't know any better.
2. Pool-hopping is hard enough to understand as it is. I think there is virtually nobody who has commented about the statistical properties of mining and isn't guilty of having made an error at some point. Add to it the people who actively deny that it works or is harmful to honest miners (rarer these days) and the FUD against hopping-proof methods, and an average miner doesn't know if to believe Random Internet User A or Random Internet User B. It is clear that anyone would avoid proportional pools if he had a good understanding of pool-hopping and the alternatives, so staying with the pool should not be seen as consent to pool-hopping.

However.

It is clear that the real problem is not with the hoppers, but with the pool operators who allow this to happen and deceive their customers. Abstaining from hopping will not solve the problem, as there will always be those won't even be bothered by the question of whether this is ethical, and silently eat at the due rewards of miners.

The only way to expose the corrupt proportional pools for what they really are is to openly mount a massive pool-hopping attack, demonstrating a statistically significant drop in the payouts of honest miners. One would hope that this will create an outcry leaving operators with no other choice than to back down and adopt an unbroken scoring method (such as PPLNS). To the extent that openly pool-hopping helps bring about this brighter future, I would say that it is acceptable.

[MikesMechanix]

In my opinion, the near-perfect analogy to pool-hopping is a game of poker where some people know the odds and the others don't. The ones with the knowledge (who put it to use) have a clear advantage.

Similarly, in both cases, the more people maximize their expected value by playing the odds optimally, the less there is to win. Ultimately, if everyone pool-hopped, the PPS pools would simply halt and the EV would drop to 0.

[Meni Rosenfeld]

PPS pools

Proportional. PPS is something else entirely.

In my opinion, the near-perfect analogy to pool-hopping is a game of poker where some people know the odds and the others don't. The ones with the knowledge (who put it to use) have a clear advantage.

Similarly, in both cases, the more people maximize their expected value by playing the odds optimally, the less there is to win. Ultimately, if everyone pool-hopped, the PPS pools would simply halt and the EV would drop to 0.

So it's not like Poker at all. You can still play poker if everyone knows the odds, you can't mine in prop if everyone hops.

[enmaku]

@Meni & Mikes: Agreed, I see openly pool hopping as a form of hacktivism. If we could get rid of bad ideas 100% of the time by simply talking about them I'd be all for it, but some folks won't listen to words and actions speak louder. If other miners at a pool feel any pain from my actions they should take their complaints to the pool operator who should have been using a fair system in the first place. If they don't feel any pain then they either aren't affected by my actions in a meaningful way or else have no comprehension whatsoever about how mining works and should perhaps pick another business.

Saying hopping is unfair to those who stick with one pool is like saying that arbitrage is unfair to those who stick with one exchange. If you don't like it, there are certainly ways to close the loophole and you'd be better served complaining to those with the power to do so than going after the arbitrageur. Mining is a business and you are all my competitors. If Target found a way to profit 20% more in a way that would possibly hurt Wal Mart's profits, would it be unethical to do so? I think you're all confusing personal ethics, social ethics and business ethics.

[bb]

Arbitrage trading actually helps consolidate markets.

[MikesMechanix]

PPS pools

Proportional. PPS is something else entirely.

Oops.

So it's not like Poker at all. You can still play poker if everyone knows the odds, you can't mine in prop if everyone hops.

The mining halting is really the function of the optimal hopping strategy. In poker playing the perfect strategy is close to impossible, but if everyone did, the situation would be close - everyone would win and lose the same on average.

[Desolator]

alright, I read the info and I think this is a myth.  The first strategy is to mine until 43.5% then hop off the pool and solo mine?

At the current difficulty, you have a 0.001239% (rounded) probability of finding a block within 10 minutes at 300 MH/s.  So basically you're wasting time that you could have spend contributing to the pool and getting a higher split.  Then if you beat the pool and find the block solo, you get 50 BTC but lose all the progress in the current pool.  And that's like betting on red and black at the roulette table except if the payments were less than your bet.  One is going to lose and you're betting your mining time on opposites.  If you thought you could beat the pool solo, you should solo the entire time.

The same goes for mining until 43.5% of what I assume is the estimated completion time of a block since there is no "progress" or "completion percentage."  You bet on two pools, only one can win, you're losing your mining time on whatever pool loses.  It doesn't seem logical.

[Meni Rosenfeld]

alright, I read the info and I think this is a myth.  The first strategy is to mine until 43.5% then hop off the pool and solo mine?

At the current difficulty, you have a 0.001239% (rounded) probability of finding a block within 10 minutes at 300 MH/s.  So basically you're wasting time that you could have spend contributing to the pool and getting a higher split.  Then if you beat the pool and find the block solo, you get 50 BTC but lose all the progress in the current pool.  And that's like betting on red and black at the roulette table except if the payments were less than your bet.  One is going to lose and you're betting your mining time on opposites.  If you thought you could beat the pool solo, you should solo the entire time.

The same goes for mining until 43.5% of what I assume is the estimated completion time of a block since there is no "progress" or "completion percentage."  You bet on two pools, only one can win, you're losing your mining time on whatever pool loses.  It doesn't seem logical.

You're wrong, but thanks for proving my point about the existing confusion.

Tell me this. Let's say the proportional pool has (2*difficulty) shares in the current round. If you mine for it now, then no matter what happens, you will get less than (Block reward)/(2*difficulty) per share you submit. So why do it instead of, say, a PPS pool which gives you (Block reward)/(1.1*difficulty) per share you submit?

[bcpokey]

alright, I read the info and I think this is a myth.  The first strategy is to mine until 43.5% then hop off the pool and solo mine?

At the current difficulty, you have a 0.001239% (rounded) probability of finding a block within 10 minutes at 300 MH/s.  So basically you're wasting time that you could have spend contributing to the pool and getting a higher split.  Then if you beat the pool and find the block solo, you get 50 BTC but lose all the progress in the current pool.  And that's like betting on red and black at the roulette table except if the payments were less than your bet.  One is going to lose and you're betting your mining time on opposites.  If you thought you could beat the pool solo, you should solo the entire time.

The same goes for mining until 43.5% of what I assume is the estimated completion time of a block since there is no "progress" or "completion percentage."  You bet on two pools, only one can win, you're losing your mining time on whatever pool loses.  It doesn't seem logical.

Have you read the math behind it and are disagreeing, or are you just saying that it doesn't "feel" right? If you have a problem with the math, point out the error in the proofs or formulate a proof that it doesn't work. If it's feel alone, well, you can either try it empirically or ignore it.

It "seems logical" to me, in fact I had considered such a thing way back when deepbit and slush were basically it for pools, when I noticed that on some rounds the amount of time I spent mining a block on deepbit would have given me more profit on PPS than prop, but on quick rounds prop was a much higher yield, and so if I could simply run a script that automatically dropped me over from one scheme to the next based on some calculation I could come out with more coin than normal on average. Obviously not possible directly due to delayed stats, but pool hopping I gather is essentially the extension of my idea to the modern system with multiple pools. You make a big return on short rounds, so if you hop around on average you will come out slightly ahead.

EDIT: Sniped by meni.

I still don't consider this unethical, for where do you draw the line? For example, BTCGuild is currently in a 24 hour slump, with a -40% expected return. Am I bound to forever mine on BTCGuild, because I have some hash there and there is some unwritten unspoken implied agreement? Or if I see that the pool is having a poor run, am I allowed to move over to say deepbit, and when BTCGUild starts to do better move back? Where are you drawing the line? 24hours? 12hours? 6? 2? 1? What is the distinction between being efficient and unethical? Do I have to wait until a block is completed on a pool to leave that pool then wait til a block is completed on the new pool to join? To me there is no agreement other than what has been written in the rules of proportional payout, and pool hopping is merely the extreme extension of following the luck.
I've seen no response to the question of coin hopping either in regards to namecoin, which is a near-perfect analogy of poolhopping which many anti-poolhoppers participated in.

I still don't do it (except namecoin hopping), but I see no problem with it.

[JoelKatz]

In my opinion, the near-perfect analogy to pool-hopping is a game of poker where some people know the odds and the others don't. The ones with the knowledge (who put it to use) have a clear advantage.

Not at all. People playing poker all know that they are competing with each other, that their interests are averse to each other, and that each one must protect themselves form being taken advantage of by the other. The expectation of good faith and fair dealing certainly doesn't include ignorance about the odds of the game. That is nothing like a mining pool where many miners believe their interests are aligned.

Similarly, in both cases, the more people maximize their expected value by playing the odds optimally, the less there is to win. Ultimately, if everyone pool-hopped, the PPS pools would simply halt and the EV would drop to 0.

But there is nothing like that in poker. The games works better when everyone knows the odds.

[Desolator]

since it's really super late at night and I had to go to work earlier before finishing most of those documents/explanations/proofs, is the basic theory that the difference in pool profitabilities mind-block is based on the drop or gain in that particular pool during that process of finding a block?  Cuz that would sort of make sense.

EDIT: well I got curious and read more of it and kinda get it now but the theory doesn't seem to hold water even now.  Any given moment in time results in an equal chance of anyone finding a block weighted by H/s.  So your payout proportionate to everyone else's is lower if you all started at the same time or started slightly later.  So if you all start at once and you've been going for a long time and you decide it's just not going well for that pool in that round, you jump to another pool which hasn't been processing for as long and have an equal chance of getting the block but for a higher reward.

...but every single pool starts processing a new block at the exact same time since everyone's working on the same chain and everyone knows when blocks are completed.  And if there "is no progress" then a smaller pool wouldn't be "farther" from creating a block and there would be no point in pool hopping.