Proof of Stake

According to Buterin blog post, formula for validation of proof of stake is:

SHA256(prevhash + address + timestamp) <= 2^256 * balance / diff

where:
balance= my balance on address
diff=network difficulty
address: my address that i use for stake
timestamp: timesatmp of the block
prevhash: hash of the previous block of the chain

Suppose I resolve stack for Nth block, I can brute force a field (address output with 0 coin transfer) in a transaction that I inserted in order to generate an hash block that validates the formula for the (N+1)th block. For example, if i have 1% of total coin amount and i resolve a block , i should try in mean100 attempts of the resolved block to validate the formula for the next block, in this way a staker can advantage and resolve also next proof of stakes blocks.
There should be a system that after resolving a block, randomize the pre image of sha256 for the next block to avoid this problem.
I would like to know what solutions have been used to solve this problem.

Thanks