The Enemy Of My Enemy Is On My Payroll

I was perusing an article about increasing nationalism in Russia and had an epiphany on reading the closing paragraph. Creating a political party to split the vote is hardly rocket science, and yet at the same time its so obvious I wonder if people have really ever tried to implement the idea. You'd have a hard time pulling this off for a major position (say, the Presidency), but for a lesser position (State Senate?, school board?) it seems like it would be a no-brainer. Its almost algorithmic in its simplicity:

1. Identify a significant split in the constituency represented by your opponent.
2. Identify your oponent's position on the issue.
3. Identify a candidate with similar positions on most issues, but an opposite position on the wedge issue.
4. Get that person on the ballot.

Obviously that's still a fair amount of work, but I would think that it wouldn't be unsurmountable for a seasoned professional. Which kind of makes you wonder if its happening now and we just don't know it.