1) As I and others have said before, it is absurd that you attempt to draw conclusions about the long-term effects of an entire election method from one example involving a three-way tie.
2) The reason this example does what it does is because, in the case of an EXACT tie, the tiebreaker considers the margin of defeat. So yes, you can get margins-condorcet-esqe incentives to bury your favorite in the case of an EXACT tie. In a real, public election, this is utterly irrelevant since the probability of an exact tie is vanishingly close to zero.
-Adam
---- Election-methods mailing list - see http://electorama.com/em for list info
