I don't actually think the hand counters are going to adjust the voting percentages, so let's go on the assumption the percentage of the vote per candidate remains exactly the same after the recount.
Any hand recount is going to increase the total number of votes counted, regardless of who those votes are for. Well, this can't be bad, can it? After all, we want everyone's vote to count, right? Given the above statement of voting percentages remaining exactly the same, Republicans have no valid gripe, right? Wrong.
Each precinct is going to have some votes not counted, be it filled in incorrectly, not quite punched out, bubble not quite filled in, you name it. Of course everyone's vote should be counted, if we can come up with a better method I'm all for it (personally I like the "outdated" booths).
The problem is if you selectively recount votes in areas predominantly one party or the other what you are doing is weighting those areas over the others, because a larger percentage of votes do get their votes counted. The end effect is equivalent to an increase in total voters in party-specific areas.
Yes, yes, I'm cold and uncaring and should want everyone's vote to count, but the fact is anything short of a total Florida recount will not give an accurate representation of that state's votes. No, I don't want a total recount, but selective recounts could well turn the election, whether they take advantage of "the potential for mischief" or not.