Oh but that is an interpretiin issue due to poor wording. And I wish IFAB fix this. Because that is the source of the different teaching regarding if you can go back after you have signalled advantage.
Note the the use of the word "anticipated". This could/should be interpreted that untill you have actually decided (prerequisite for signaling) you are playing advantage, you are not technically playing advantage.
The first part requires the benefit to be there, no anticipation, and that is when you allow play to continue. So the first and second part can't co-exist.
Once upon a time, the Laws expected referees to immediately decide on advantage--and if it turned out not to be there, too bad, too late to do anything about it. The Laws were changed to allow the referee to be sure the advantage was really there. That change led to two schools of thought: (1) wait to see if the advantage is fully there, or (2) signal advantage promptly, and go back if it isn't there. Neither school of thought is wrong under the Laws, though (1) has become predominant and I gather is what is taught in most places. (Many practitioners of (1) believe it is the only right way; practitioners of (2) tend to believe both are acceptable.)
Both have different merits. I think (2) is particularly useful where there is a hard foul, a possible good advantage, and the R wants to immediately communicate that he saw it and recognized the foul, to avoid further escalation while permitting the advantage. The fact that (2) is permissible
also means that if the R signals advantage and quickly realizes that what he thought was there wasn't, he can go back.
The danger of (2) is confusing an advantage that wasn't there with a poor play or poor decision by the attacker. If the advantage the R thought was there was a pass to a teammate, but that teammate was OS, there was no actual advantage. But if the advantage the R saw (for a non-PK) was an unimpeded one on one with the keeper, and instead the attacker makes a poor decisions to pass to an OS teammate, the advantage was there and the player blew it. Similarly, if the advantage the R thinks will ensue is a ball bouncing to an attacker, but the nature of the ball bounce turns out to only give a very difficult play, the advantage wasn't there; but if the attacker does get a good ball (better than the FK) and mishits it, then the player fluffed the advantage and there is nothing to go back for. It's those plays in the PA that are particularly tough, because the question really is was the chance
better than a PK, which not many are. (And I think is what led to the "no advantage except a goal" philosophy.)