PB12 commentary: 3. Bracketing and difficulty of rounds

Samer Ismail <stipenn_at_stipenn.yahoo.invalid> <*stipenn_at_stipenn.yahoo.invalid*> · Sat, 01 Feb 2003 05:58:13 -0000

Dan Greenstein wrote:

<<The difficulty was nicely ratcheted up from the prelims to the 
playoffs and then the single elimination rounds.>>

I have to say, in all honesty, that I did not specifically *intend* 
for the rounds to get noticeably harder from the first set of brackets 
to the second. If they did, it was a function of what the teams wrote, 
more than anything else.

As far as the decision to use the re-bracketing system: there are 
pros and cons to each system. My main dissatisfaction with the 
four-bracket system is that a lot of weight is placed on relatively 
few matches. And while I admit that 18 of 60 isn't a whole lot more 
than 16 of 60, I still think that every team that played:

* Had the opportunity to play at least one team they wouldn't play at 
any other tournament (except national championship tournaments);

* Had the opportunity to play at least one team in the top 10% or so 
of all teams in the nation;

* Got to play matches against teams of roughly equal ability in the 
playoffs/consolation rounds.

On the downside, there were a number of teams that were effectively 
eliminated from the playoffs by the end of the day on Friday, and 
there were three teams that left after lunch on Saturday. I admit that 
such is a downside of eliminating over 2/3 of the teams after lunch, 
but I don't see a way around it.

I do need to apologize for the delay in posting the schedules for 
Saturday afternoon. There were two reasons for this, one of which was 
entirely my fault, and one which is an inherent problem with the 
system:

* My fault was in not realizing there was a really simple way to keep 
track of the teams when writing down the schedule; I believe that is 
why--as someone (Dan?) pointed out to me, I had duplicated a match on 
at least one of the original schedules. Compounding this was the fact 
that the random pairing of teams for the afternoon rounds 
(A1-B1-C2-D2-E3-F3, etc.) put three teams that needed byes in specific 
rounds in the same bracket. [Had it put four in the same bracket, I 
would have been forced to re-bracket, as that simply cannot be done.]

* What wasn't my fault is the fact that I could not produce the 
schedule until all the results had been entered and all the ties 
broken. The only way around this would have been if I had been using 
completely blind packets, which simply was not possible.

While I admit that things could have been better organized, I would 
say that, for all the added hurdles this format created, in the long 
run, it still worked far better than I could have hoped, and I do 
think it was an improvement over years past.

--STI