Over a long career you accumulate a fairly clear sense of who the generous people are. In my experience they share a characteristic that took me a while to notice: none of them appear to be keeping track. They make the introduction, read the draft, take the call from someone who can do nothing for them, and then they forget about it entirely. 

The people who do keep the ledger are easy to recognise too, and they tend to be the ones least worth knowing. It is not that they are unhelpful, because they often help a great deal. It is that the help comes with a faint transactional quality that everyone can feel, and that the accounting is visible in who they choose to help: the people who appear likely to be useful later. Which brings me to the practical argument, quite apart from the question of what kind of person one wants to be. Scorekeeping systematically restricts your generosity to those you have assessed as able to repay, and that assessment is a remarkably poor predictor of where anything comes back from. The PhD student becomes the person who hires you. The founder you helped for free, whose company failed, introduces you to the one that does not.

So the generous behaviour turns out to be the better strategy, which is a slightly awkward conclusion because the moment you adopt it as a strategy you have started keeping the ledger again. I think the resolution is simply to treat the help as the point rather than the investment, and to accept that most of it will never return anything, which is fine. So, as a reflection for today: who have you helped recently who had no plausible way to repay you?