A deal can still end up an overall net positive if the player opts out (as in "this team has been better off signing this player than not having signed the player"), yes, but the same deal would be better if the opt out wasn't there. Even in the example of Belle: yes, it worked out better for the Sox in retrospect, because Belle declined suddenly and unexpectedly, but Reinsdorf and many others were extremely upset at the time that it happened, because Belle was worth more than the remainder of his contract -- which is why it made sense for him to opt out and find a better deal. If the opt-out wasn't there, the Sox could have also avoided a declining Belle by trading him at that time, and the return would have been a net positive that reflected the surplus value of the deal.
This isn't to say that an opt-out should never be accepted in a deal, it's just that it is a player advantage and should come with a cost during negotiations.
All of that "paying for the surplus value in early years" still applies to a deal with an opt out. The team is still on the hook for that money, should the player age as expected or worse than expected. All of the "upside" is lost, because if the player ages better than expected, and thus the team is in line to get more of that surplus value than what they paid for, the deal gets torn up. Essentially all of the risk normally associated with a long-term, big money deal is still on the team, but without the potential for reward.
It's the fact that the decision is in the hands of the player, not the team, that makes the difference. If, at the time of the opt out, the balance of money/value for the remainder of the deal is off in either direction, the team is guaranteed to get the worse end of it.