It is important to understand that in saying that the moral status of an act is determined (at least in part) by its results, this is meant to include all af its results. It is not only the immediate, or short term, results that matter: long term results, side effects, indirect consequences — all these matter as well, and they count just as much as short term or immediate consequences. If, for example, I must choose between an act with a small immediate positive effect — but no other later effects — and an alternative act act that will have no immediate effect, but will eventually produce a lot of good, it is the second act I should perform. Similarly, if an act will have both good results and bad results, then these must all be taken into account. The question is: how good or bad will the results be overall, on balance, taking into account all of the results; and how does this compare to the overall results of the other acts available to the agent?

Now it was probably obvious all along that talk of "good outcomes" or "good results" was meant to include all of the results and all of the consequences. But for all that, it must be noted that these terms are indeed being used here in a somewhat nonstandard fashion. For normally when we talk of the "results" or "outcomes" or "consequences" of an act, we mean to be restricting our attention to those events taking place after the act that are caused by the act. It is only if the act literally produces some later event that we call that event one of the act’s results or consequences. For example, if the later event would have happened anyway, we don’t normally consider it part of the consequences of the act. But for our purposes in normative ethics, we need to give these terms a "wider," more inclusive sense than is normal. In thinking about the goodness of outcomes, what matters is not the goodness of what I literally produce myself, but rather the goodness of everything that ultimately happens, the goodness of the "upshot."