Wouldn't it make more sense to get into a hashflare.io lifetime scrypt mining contract? ROI = 250-300 days for the contract and the L3 (including maintenance fees, which you'd have to pay anyways due to 400W pper unit, not to mention downtime)...
Genesis just came out with a 2-yr contract for twice as expensive today. Wonder who's going to buy that. Probably the same suckers like me who went in on their ZEC mining contracts...
EDIT: I backed out of the AliExpress deal - bought some could mining contracts instead. @0.20-0.30kWh it's just not worth it in my area.
Lets review that Hashflare Cloud mining contract:
$8.20 x 250Mhs = $2050 up front for a lifetime 250Mh/s contract.
Maintenance costs are 0.01 x 250 = $2.50 per day
Assuming you can get maintain a return of 0.00003 BTC per Mhs over your first year, you're earning (at today's BTC price) $820 x 250 x 0.00003 = $6.15 per day - $2.50 maintenance = $3.65 per day profit.
$2050 $3.65 = 561 days ROI, and at that time you will have spent $3452 to have mined 250 x 0.00003 x 561 = 4.2075 BTC
Taking that money now, you could buy $3452/$820 = 4.2097 BTC
As I've only accounted for the 1st years BTC returns (year 2 might average 0.000025 BTC/Mhs in year 2 if you're lucky), I think that given the choice of Cloud versus Buying an L3 and mining at home you'd have been better off with the third option of buying and holding a diverse basket of BTC, LTC, Monero, ZCash, ETH, etc and trading.
There's a few variables that can change this picture a lot, the diff of scrypt coins (which affects your raw mining output), the mined coin to BTC exchange rate (generally rises and falls with fiat BTC price), the BTC price itself and the efficiency/luck of pools you mine. 
As a steadfast miner, the difference with buying miners is that a.) once ROI'd the only cost going forward is electricity ($0.075 Kw/h for me), and b.) they generally retain 25% of their purchase price after 3 years so the true ROI is only 75% of the purchase price.