'Want' is one of the most peculiar variables when it comes to the transactional dynamics of life. Most people think that 'to want' is to have a desire to own/do and use it as an indicator to work towards the object of desire. The problem with this line of thinking is that it treats 'want' as a leading indicator without catering to the changes in beliefs.
Although it is true that there is no good alternative to start with and you have to go with whatever you think you 'want'. The problem, however, arises when you keep going without stopping to reassess the validity of the 'want'.
Think about it, how do you know if it is a 'want', a 'want-to-want', or a 'have-to-want'. It is literally impossible to distinguish. And to add to the burn, most people try to use it as a leading indicator, only to be discouraged when it doesn't pan out the way they wished it would; while attributing the cause to elements like distractions, will-power dissolution, cognitive exhaustion, circumstantial(financial/emotional) distress, etc.
Although it is possible that these are the real causes, I also think there is more to it than meets the eye. For eg. Emotions, whether positive or negative, abstract away reasons. The belief that overwhelming desires should know no reasons—although true—might be playing some part here, possibly the part of creating a misconception that all overwhelming 'wants' are real 'wants' and one mustn't heed to reasons even if they arise. When in reality, you can only ensure the validity of most of your 'wants' post-hoc.
That said, I do think that there are variables that can be used to gauge the validity of the 'want'. To be precise, two variables: Constraints and Resilience. These variables can be used as leading indicators to predict if the 'want' has any possibility of yielding any result.
I won't go into the details here as I am still trying to understand it myself. But out of the two, I think resilience is far more difficult to model than constraints, for constraints no matter how weak can be managed by using a stronger enforcement protocol. In the case of resilience, there is a lot of investment that goes into it from developing it to managing the rapidity of recovery, to persisting the state post-recovery.
With constraints, the challenge is ensuring that you have a good enforcement protocol. Even a weak constraint can be adhered to if the enforcement is strong enough. Although there is a lot of things that can be done with the constraint design, it is the enforcement that usually makes or breaks the thing. Remember that I am not saying that constraints alone cannot predict the success rate, I am merely stating the importance of enforcement when using constraints as variables. FWIW, there are designs that allow constraints to operate independently of enforcement protocols, but they are far and few between. I like to call them 'autocatalytic constraints'. (What would be a good example here? Death? Computation? Consciousness?).
That said I think there is still a lot to understand here. I still do not understand how to create a strong enforcement protocol without relying on standard means or incorporating an external enforcing agent. I still do not understand how to design a good constraint that works without relying on explicit enforcement protocols. Or for that matter understand the dynamics of building and managing resilience. But I do feel that there is something here that is worth exploring. Your ability to recover will not be the same across domains, so that should tell you something about the possibility of reaching the finish line. Your inability to deviate owing to the constraints should tell you something about how far the constraints will be able to hold before you reach (or fail to reach) the finish line. It's just that I do not know at the moment how to make sense of it all without getting caught up in the never-ending drama of prescriptions vs vague generalizations.