You are still very very funny.
My concessions: Yes, sometimes round robins are not used, and when this occurs, my argument has less weight. But otherwise, it has much weight. And yes, sometimes TU's are gotten at the giveaway, so here it might reveal buzzer skill---a part of a team's exellence--or, it still reveals cognitive superiority---miniscule, to be sure, but there nonetheless. And, if you THINK not, you should propose a rule change---no bonuses allowed on buzzer races after he giveaway. Until you do, winning a TU earns a team the precious opportunity you seem to value most---the opportunity to rack up PPB. And rightly so.
But your tendency to dismiss wins and losses remains stunning. Your argument actually invalidates national tournaments and their final rankings at the end. Why? By definition, wins and losses against any team at nationals will involve a small sample against said team. I will give you our own examples at both PACE and the HSNCT in 2010, and I will use examples from teams we played only once. I will start with PACE : we lost to Dorman B, we defeated Walter Johnson, we lost to Alpharetta, we defeated Dunbar B, we lost to Lisle, we defeated Loyola. Now, turn to the HSNCT and more teams we had only one match with: we defeated Olmsted Falls, we lost to Solon; we defeated Chaska, we lost to Berkeley. Using your argument, one match does not allow any conclusions. Yet both PACE and the HSNCT ranked all of these teams at the end, and I believe they did so correctly and with both good cause and a good conscience. Why? Because in tournaments, TWO CONSTANTS are at work. FIRST, wins and losses against a field of teams ultimately refine the field into fairly reasonable gradations of quality, with the teams WINNING MOST FREQUENTLY against the field rising to the top, and those losing falling to the bottom. SECOND, the field plays against a set of questions DESIGNED to be relatively constant in its degree of difficulty, so that results in one round can be reasonably compared to results in others.
This pair of constants within a FIELD is what you seem to ignore. My interpretation is that a win is not simply a win when these constants are at work---a win is a win within a pattern of wins and losses against a constant field and a constant set of questions, so even if you play team only once, your overall wins & losses record is quite meaningful precisely because of the constants mentioned. And, of course, if a team consistently wins at a high rate at tournament after tournament, finishing at or near the top, clearly it might not be the best team, but surely it is among the best teams, as its consistent winning shows.
But here is where your theory of wins and losses becomes stunning. You suggest a win here or there proves little, since it is such a small sample. In another thread you called such wins "outliers". What you ignore is that teams rarely play each other more than once in a tournament, except perhaps in advantaged finals But using your logic, even an advantaged final has only two matches at most in addition to a possible preliminary match, thus, samples remain small,3 at most, so, logically, no tournament result necessarily means anything due to these small samples. Taken to its logical conclusion, your argument undermines all actual tournament results, inclluding national tournament, results, given the small samples they inevitably deal with. And if you suggest that national playoffs sometime have mulitple matches, you ignore this easy criticism: How do we know the finalists deserve to be there, since their earlier preliminary wins prove nothing---maybe the teams they defeated were truly superior, if only they had enough sample games to show this. SJHS was eliminated from the HSNCT by Berkeley, knocking us 10 spots lower in the HSNCT RANKINGS at the end. Should we not complain to the HSNCT saying that one win means little, yet the HSNCT made it worth 10 spots in the rankings!!! Would not everyone see us as trying to deny Berkeley its deserved finish and ranking? I think so, and it undermines Berkeley's fine result, does it not? We lost, Berkeley won, so Berkeley deserved to finish ahead of us.
So, in closing, it is my considered judgment that your theory ignores the constants I mention, constants designed to make tournaments meaningful even if almost all matches are against different opponents--in your vocabulary, "small samples". But it is not the isolated win or loss that matters, it is the PATTERN OF WIN and losses within, ultimately, a playoff field, against a common set of questions. And it is your common dismissal of such PATTERNS of wins, calling them isolated facts at most, which amazes your adoring and myriad fans here in the quizbowl world.
You really should reconsider your tendency to dismiss wins and losses in a controlled tournament and to substitute out of context stats like PPB in near isolation of all else. As the old saying goes, "stats are for losers". But if we were to focus on one stat alone, I would still go for the pattern of wins and losses over any other.