Re: Winning versus scoring

I try this again, could I get an explanation as
to why Princeton B was ranked ahead of University of
Maryland? This the biggest problem I had with the rankings.
Was it all because they scored an "incredible" 395
more points in 12 TUs than us? Or maybe because we got
a "lucky" break and only won our game against them
by 125 points, which is of course a "small margin"
that can be made up by 3 TUs (with one power) and 30
all three bonuses, of course a victory like that is
essentially meaningless. Finally, we only had a "narrow"
three game lead over Princeton B. I can understand with
all those minor factors accounted for, Princetion B
is definitely more "statistically dominant".

Thank you NAQT, for showing that winning close games
doesn't mean anything. Also for showing that winning
against quality teams doesn't mean anything either,
because it could be any factor other than the fact that
the winning team is better. It's all about how a team
running up the score on a weaker team that counts. This
system doesn't even take into account if a team plays a
weaker team in a slow room, they are disadvantaged
because they are more likely to score higher against
weaker teams than against stronger teams.
I don't
mind getting screwed over some "magical" numerical
formula, but please don't justify it with stuff like
"win-loss" records can't be compared across regions. I
definitely think it can, just weigh the good wins against
the bad wins maybe even the good losses against the
bad losses. Game by game results is in the stat
program and can be had, at least in the program we use. I
definitely think it's a better way, because beating Michigan
B by 20 points is vastly different than beating a
team by 400 points that had below 5 Points Per Tossup
(PPT). Just like losing by 20 to the same below 5 PPT is
different than getting smashed by Michigan B 400
points.

Good luck to all qualifiers, and congrats to them for
playing in a tournament where winning
matters.

Andy 
University of Maryland

This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:43 AM EST EST