Tonych suggests a double spend attack in which:
As long as the total amount of double-spends exceeds the fees, the attack succeeds.
However, as I stated in the more extensive discussion of double spend attacks above:
A transaction is considered confirmed when the sum of transaction fees of all transactions descending from the given transaction is more than twice the value of the original transaction.
This rules out the kind of double spend attack tonych considered. By the time the transaction is confirmed, at which point the attacker would like to publish their secret TDAG, it is already unprofitable for them to do this.
Providing individual transactions are small relative to the network, this will not be an infeasibly high barrier for confirmation.