The Psychology of Pooled Mining

[bitcoindaddy]

I have to admit that when I first started mining (a few weeks ago), I couldn't understand why a lot of people were doing pooled mining instead of going solo. To me, the people using a pool were foolish for giving up a small percentage of their profit.

If I were being completely logical, I would solo mine since that would maximize my profit over time. But after a couple successful blocks, then a long dry spell, I tried pooled mining and LIKED IT. Here's why:

When mining solo, I would wake up in the morning and go down in the basement and check my rig to see if I had solved a block. Most of the time I was disappointed to find nothing and the disappointment just seemed to get worse each day. I would repeat the process when getting home from work - more disappointment.

Then I tried a mining pool (Deepbit). Even though I was giving up a small percentage of my profit, I had the following benefits:

• Everytime I checked my progress - I could see a little improvement each time. Instead of disappointment each time I checked my progress, I got a little positive reinforcement.
• When I am at work, I can check my progress and see not only that I am earning money, but also look at my hash rate and be comforted that everything is working properly back home and that none of my cards went belly-up.
• When I solo mined, I was in constant fear that I would not solve any blocks during the current difficulty level and that it would move to the next level. With pooled mining, I know that I am always making progress and taking advantage of the current difficultly level.

[1714918299]

1714918299

[caveden]

• When I solo mined, I was in constant fear that I would not solve any blocks during the current difficulty level and that it would move to the next level. With pooled mining, I know that I am always making progress and taking advantage of the current difficultly level.

This is not only psychological. In case you don't solve a block and the difficult increases, I guess we can say you would earn less with solo mining.
I'd say solo mining is interesting only for those who can be sure to generate more than one block at each difficulty period.

[SomeoneWeird]

I thought the same thing, and as I don't have awesome resources to use, I opted to pool, and i've been making more than I ever would have doing it solo.

[Zibbo]

• When I solo mined, I was in constant fear that I would not solve any blocks during the current difficulty level and that it would move to the next level. With pooled mining, I know that I am always making progress and taking advantage of the current difficultly level.

This is not only psychological. In case you don't solve a block and the difficult increases, I guess we can say you would earn less with solo mining.
I'd say solo mining is interesting only for those who can be sure to generate more than one block at each difficulty period.

No, times when you can't find a block before difficulty increases will be offset by the times that you find two blocks.

What you are basically saying, is that "if you get unlucky, you earn less with solo mining", but it's just as likely that you get more lucky than you would expect, and earn more. Pooled mining just reduces variance (which is valuable to many people), but solo mining always has the biggest expected value.

Hmm.. Now that I think of it, there might be some value in being part of a really big pool, because that reduces the possibility, that someone else finds a block at the same time, and manages to invalidate your block. How often do such block collisions occur?

[CydeWeys]

Hmm.. Now that I think of it, there might be some value in being part of a really big pool, because that reduces the possibility, that someone else finds a block at the same time, and manages to invalidate your block. How often do such block collisions occur?

How would that work?  If anything there's more likelihood of getting invalidated in pool mining, because you first you send the solved block to the pool and then the pool distributes it.  In solo mining, as soon as you find a solved block, you distribute it immediately, without that extra network trip.

[ribuck]

I'm solo mining and haven't found a block since 17 April, and I can understand the temptation to join a pool.

But in the long run, one will always make a little more by solo mining than by joining any pool that charges fees.

[SomeoneWeird]

I'm solo mining and haven't found a block since 17 April, and I can understand the temptation to join a pool.

But in the long run, one will always make a little more by solo mining than by joining any pool that charges fees.

For some of us it would take years to find a block, so it's not really an option.

[Grinder]

But in the long run, one will always make a little more by solo mining than by joining any pool that charges fees.

That's not necessarily true. The difficulty is rising so fast now that if you have bad luck in your solo mining early on, you will probably never catch up.

[ribuck]

The difficulty is rising so fast now that if you have bad luck in your solo mining early on, you will probably never catch up.

If you have good luck in your solo mining early on, the pools will probably never catch up.

The way hashing works, there's never any "catching up". All there is, is the probability that the next hash might be a winning one.

With solo mining, you occasionally get a block of 50 coins. Or a little more, occasionally, with fees. With pooled mining and a 2% fee, your expectation would be to get 49 coins, on average, in that same timespan.

That applies no matter what the timespan is; no matter what difficulty-changes there are; and even whether you are CPU mining.

[nster]

you get paid for invalids on deepbit

[Zibbo]

Hmm.. Now that I think of it, there might be some value in being part of a really big pool, because that reduces the possibility, that someone else finds a block at the same time, and manages to invalidate your block. How often do such block collisions occur?

How would that work?  If anything there's more likelihood of getting invalidated in pool mining, because you first you send the solved block to the pool and then the pool distributes it.  In solo mining, as soon as you find a solved block, you distribute it immediately, without that extra network trip.

I think that the time it takes to transfer a block from miner to pool server is trivial compared to the time it takes for a block to propagate throughout the network.

My original reasoning was that if your pool has 800GH/s rate and the whole network is 2000GH/s, the threat of conflicting blocks comes from only 60% (1200GH/s) of the network, whereas when you are solo mining, you are facing 100% of the rest of the network.

And also assuming miners start building on block chains in first come first serve basis (is this correct?), in conflict situations pools will always build on their own block, making it very likely that their chain will win.

However, the difference these make might be small enough to not matter, depending on how common these collisions are.

[Grinder]

The way hashing works, there's never any "catching up". All there is, is the probability that the next hash might be a winning one.

We're talking about the profit, not the hashing. Average payout minus 3% both when the difficulty was 25 000 and when it is 250 000 easily beats bad luck when it was 25 000 and a lucky streak when it is 250 000.

[ribuck]

We're talking about the profit, not the hashing. Average payout minus 3% both when the difficulty was 25 000 and when it is 250 000 easily beats bad luck when it was 25 000 and a lucky streak when it is 250 000.

True. But it's also true that Average payout minus 3% both when the difficulty was 25 000 and when it is 250 000 loses significantly compared to good luck when it was 25 000 and an unlucky streak when it is 250 000.

If you want to maximize your expected profits, mine solo. If you value predictability and dislike variance, mine pooled and forfeit the pool fees.