1) how does it increase centralization ? Last time i checked (will vary on exact service model provided, but still) cloudmining doesnt attribute new mining, rather transfer ownership of existing hashrates. Depending on exact terms of service, theoretically users can aid DEcentralization by selecting alternative pools
Centralization of the hardware , just like storing all your coins at mtgox , leads to security concerns... owners stealing hardware, theives targeting mining centers , fires, network outages, ect...
2) yes. and no. far too many, far too volatile factors will shape the final return. Whilst obviously everyone wants to make money, the ROI is near impossible to accurately predict. I'd say cloudmining shouldnt be seen as a sure-fire way to make money,rather it should be seen as a way to try out mining without taking on the risk of spending > $2k on mining hardware then have to have all the associated headaches (cooling, storage, noise, maintenance delays etc). You can switch on / switch off your cloudmining at any point, and see daily BTC returns (even if miserly) and at worst, youre risking 10-15 bucks to learn first hand. On the other hand if you make a paltry 5mBtc per day for a year and Bitcoin raises by 30-40% you could see a significant ROI.
While a calculator cannot determine the expected future difficulty it gives a rough idea if you map out a range of possibilities and buying BTC directly is far less risky despite what you say.
1) assume company MINERS'R'US are contributing 500TH/s to the whole network. MINERS'R'US sells bob 50TH/s MINERSRUS new output MINERSRUS - 450THs, Bob - 50THs. No effect on total output. The risk of hardware failure is always there, although if you're mining remotely, the DC operator takes the responsibility for support, maintenance and upkeep.
2)i wasnt just referring to difficulty, but also to elec price, behavior of other miners and the behaviour of the FX market. I agree, mining is risky regardless of how you do it, I was just saying that it is far less risky if you're risking 15 bucks vs 2-3k, and you dont need to be a hardware specialist either.
Im not disagreeing that cloudmining is a risky venture, just simply that there is and should be a market for peeps who want a low-cost entry to mining
