If your league has neither IP limits (weekly or season) nor transaction limits, you could indeed do just that (probably at the cost of ERA and WHIP). It's not that uncommon a ploy, but many leagues consider it to be against the spirit of the game and add limits.
In H2H leagues you can set min. IP limit for each week.
The way to undercut the 'churning' strategy you describe is to set the pitching categories to ratios instead of counting. It would be as simple as changing 'Ks' to 'K/9'.
classic categories: Wins, Ks, ERA, WHIP, Saves
ratio categories: Wins, K/9, ERA, WHIP, Saves
When using the churning strategy you typically sacrifice ERA and WHIP in order to win Wins and Ks. If you change from Ks to K/9 then churning will most likely result in you sacrificing 3 categories in order to win 1.
"The game has a cleanness. If you do a good job, the numbers say so. You don't have to ask anyone or play politics. You don't have to wait for the reviews." - Sandy Koufax
We had a similar problem in our head to head leagues the last two years. This year we are placing a roster moves restriction once the season starts. Becasue we want to have unlimited roster moves before the season starts, our commish is just going to monitor the number of transactions that each team makes after the start of the season. Not really that hard to do.
Like a previous poster stated, you could change k's to some ratio involving k's like k/0 or k/bb. We tried this last year using both k's and k/bb, and didnt' work all that well. Another suggestion would be to add losses and walks as negative categories. Therefore, if a team want to flex pitchers everyday, the team would likely win k's and wins, but likely lose losses and walks, and also likely to lose era and whip. Still not ideal as it basically just makes some of the pitching a wash and therefore putting an emphasis on hitting, it is one way of reducing the incentive to flex pitchers everyday
I don't think this is as much an issue as you might think. The reason is that the players on your team will probably be better than the people available in the free agent market. If they're not, then you're probably not winning anyway.
So let's say you get to the 5th day in the week (Friday) and realize you're down by 2 wins, and all your starters won't be starting on the weekend. Do you drop 4 of your pitchers to pick up 4 pitchers that will be starting? I doubt it, since the other teams in the league would jump all over the people you dropped.
And if you're planning on rotating 2 pitching "slots" by droping a person each day, the quality of pitcher you'd be adding would be atrocious. Think people like Brett Tomko (5.28 ERA, 1.52 WHIP, 13 wins), Jason Johnson (4.18 ERA, 1.56 WHIP, 10 wins) and so on. Yes, you MIGHT pick up a win or 3-4 extra Ks, but you'll be losing in WHIP and ERA. Plus, if you don't get the win, you're going to be tanking in 4 of the 5 pitching categories.
Basically you're suggesting fielding a large pitching staff made up of players that nobody else wants on their team...