That works mainly because the money comes with a heavily implied threat: don't vote the way we want and the money spigot stops, or even reroutes into the coffers of your opponent.

But if that all happens, including the opponent funding, and those opponents get routed, then the bluff's been called and the lobby's hand has been found wanting.

The article suggests something like 90% of their spend was intended to change results. Can you help me understand your comment? I don’t get it.

On those terms, they also wasted a lot of cash. 90% of it went to candidates who lost (or opposing candidates who won).

Why would a candidate compromise their platform like that if they aren't going to lose without those donations and won't even lose if the money is spent against them?

I don't understand how a blanket statement like this can apply. In a voting district where one party is heavily favored, such that that party's primary election winner is basically going to win the general election (e.g. New York City), then primary spending seems like the only place to influence the election.