Lawrence Solum over at the Legal Theory Blog has this excellent discussion of the ex ante and ex post question as it comes up in analyzing best legal rules and depending on the orientation of the analyst as to whether we care most about outcomes (regardless of fault) or fault (regardless of outcomes). Ex Ante and Ex Post. (These would be extreme ends of the spectrum).
But the ex ante/ex post problem also extend to the analysis of any issue where you are not sure what the right distribution of resources and investments are. Ex post, when you know how a particular decision has come out, that decision tends to look inevitable and thus it is easy to criticize the choices made or not made as ones which ignore the "obvious" likely outcome. This is often termed hindsight bias (or Monday morning quarterbacking) and can make decisions seem easier than they were at the time they were made.
In contrast, a lot of decisions require us to make assessments ex ante - before we know how it is going to "turn out." And that is one of the problems raised in an earlier post on the value of scholarship. How can you calibrate the right amount to invest (even assuming you could come up with an accurate assessment of its cost, a proposition I am skeptical about), if you don't know how much of the "bad" stuff you have to support in order to get the "good" stuff. And that also omits the even more problematic question about assigning labels like "good" and "bad" to something which may defy easy calculation in market terms.
There is an analog to this problem in advertising circles. The old joke goes: "I know that 50 % of my advertising budget is wasted. I just don't know which half!" For advertisers the ROI problem (return on investment) is both ex ante and ex post since, although they can track dollar returns from investments in certain advertising campaigns, they can never be sure that they know why a given campaign seemed to drive sales so successfully. The annals of advertising lore are littered with examples of an advertiser trying to take a proven formula only to have it bomb. Hence the joke. But advertisers have one advantage over academics in any field - they are marketing a product which is sold in the market, not in the so-called "marketplace of ideas," one with a price in dollars (which of course may have little or not connection to its intrinsic worth in some other system of valuation) and advertisers can measure the sales number before and after a campaign. They may not be able to say for sure that it was the campaign that made the difference, but it is often a reasonable inference.
For the hapless evaluator of legal scholarship there are no such clear metrics for tracking the dollar value of scholarship and such metrics as exist (citation counts, downloads, US News peer reputation scores) are, as I observed previously, incomplete at best and woefully inadequate at worst. It may be that investments in scholarship are maldistributed, although I have yet to see an argument for this proposition that doesn't rely on generalizations about legal academics and law schools that I don't think apply as widely as some argue. Maybe the expenditures on scholarship should be reduced. But deciding on the "which half" part of that question is surely part of the problem presented by any such proposal.