In what I'm sure will be a shock to all of you, I'm going to go ahead and disagree somewhat with my esteemed colleague.
theMoMA wrote:But I do think that hosts have a responsibility to the players to keep accurate stats and to fix errors when they surface. The attitude of simply posting and forgetting stats is an extension of the problematic attitude that some schools take, which basically forgets that teams spend time, effort, and money to play these tournaments and deserve better than half-hearted hosts.
So, I think this is probably the crux of where we part. I certainly do think that hosts have a responsibility to their teams and to the editors, and those responsibilities include a timely posting of accurate statistics. So far, so good. However, for the most part, hosts already do this. They collect stats and they post them and they're about 99% accurate. What we're talking about here is not some egregious failure to carry out basic hosting duties, but multiple people posting things like "I had one more power in this round" or "you switched my stats with someone else's." Is it regrettable when it happens? Surely. But it seems like on your account, every time someone posts this, the stats person now has to go back and change the SQBS file and reupload the results, and if 10 people make different requests at different points in time, it starts being a royal pain in the ass. At some point the marginal utility of additional precision in stats-keeping becomes so low that in my view, it's not even worth it to bring it up.
 As a footnote, one thing that's tacitly being assumed in this discussion is that whoever is making the request for the correction has the right
stats. I see no reason to assume this.
In general, I'd like to see host sites do a better job recording stats accurately, entering them accurately, and fixing them when there are mistakes. Scores and stats are a part of the game that we shouldn't have to question and there's really no good reason not to make the best effort to get them right. The idea that some mistakes are too small to fix seems to me to be one of the attitudes that leads to sloppiness in the first place.
There's a very good reason not to make the best effort because the best effort by definition would involve ferreting out and correcting every error, no matter how inconsequential. And that's time-consuming, so we settle for the next-best effort, which is to let individual people find the mistakes and then correct them as time allows. In my view, I would be perfectly happy with the next-next-best effort, in which we simply let these things drop. I do think some mistakes are too inconsequential to worry about; in any case, it's not like they result from some wilful malice on the part of the staff, but rather from simple mistake like writing a value in the wrong column. I reject the assertion that not correcting these errors is somehow derelict on behalf of the staffers.
As an aside, I see parallels between this discussion and my positions on grammar and packet formatting. I guess on the whole I'm a fan of getting these technical things exactly right even when almost right would probably suffice. The reason I think that's useful is because people actually get these things detrimentally wrong quite a bit, so I'd rather we just try to get them dead on than have endless debates about what's "right enough."
I see those parallels too, although of course in this as in that debate I am on the opposite side. Almost right is good enough, in my view; we have way more important things to worry about, because the failures that are actually detrimental to tournaments are different in kind and not just an accumulation of these little kinds of things. The people who run good tournaments but make a trivial copying mistake are not the same people who run bad tournaments in which copying mistakes are the least of your worries.