if you didn't know that stoves can be hot, then it would seem irrational to avoid touching them so vehemently, compared to any given object. the conditioned behavior is not to touch something that is hot... generally i consider CBs to be rational in-and-of-themselves unless there is an exception, such as the stove not being turned on.
to apply here, the cost-benefit analysis for those who procrastinate favors doing something else before the task in question, as a rule of thumb. there can be exceptions here too, but until someone has a reason to do something sooner rather than later, the rational behind procrastination wins, and that is that immediate benefits are worth more than only potentially suffering consequences.
for example, task A is undesirable, activity B is desirable... let's say someone does activity B, reaps the positive emotional experience, then accomplishes task A within their deadline (albeit with less breathing room)... meaning they were able to benefit both in activity B and in avoiding the consequences of not doing task A. or they could partake in activity B, find that whoever assigned them task A no longer needs it done. or that they again do activity B, accomplish task A late, and are mildly berated but ultimately that consequence does not compete with the benefit of activity B. if we replay any of these scenarios with the person doing task A, they miss out on activity B for potentially no reason.
when this sort of thing happens over and over again, im sure you can understand why people begin to prefer procrastination. really, the concept of procrastination is illogical, it is saying that certain choices have innate value and that priorities can be determined on an objective basis.
edit: if procrastination is merely opportunism in practice, then the only argument for not procrastinating is when the value (subjective) of accomplishing the task in question is greater than the sum of the benefits of the opportunities one took advantage of that prevented its completion.