BMM09: Tournament Brackets (Essential Eight Revealed, previously Play-In Game Thread)

The field is split up into four divisions:

Creed Division
Guerlain Division
Serge Lutens Division
Comme des Garçons Division

These were the top 4 vote-getters in jenson's niche poll a while back, they're just for fun and don't have any bearing on the tournament.

So how did we come to this field of 65? 48 were from the review system, 15 were selected by Basenotes members, and 2 are competing today!

Key:
Pos = Number of positive reviews.
Neg = Number of negative reviews.
Neut = Number of neutral reviews.
Total = Total number of reviews.

Group 1: Top 16 By Negative-Weighted Pct. ((Pos-Neg)/Total) - These are technically the strongest challengers of the tournament, as they had many positives and very few negatives.
Group 2: Top 16 By Simple Pct. (Pos/Total) - These have a large fan-base, but may also have their share of detractors too.
Group 3: Top 16 By Most Ratings (Pos + Neut + Neg) - Those biggest of names, these enter by right of sheer name-recognition.
Group 4:
5 Forum Selections
2 Selections By Bluesoul
2 Selections By Grant
2 Selections By mikeperez23
2 Selections By Dimitri
2 Selections By JaimeB
1 Play-In Game Winner

The Play-In Game competitors are the Lowest by Simple Pct. vs. Lowest by Negative-Weighted Pct.

If a fragrance appeared on multiple lists the highest rank it achieved was used.

Group 4, the user-contributed section, had all ranks randomized using the random.org random number sequence generator except for the play-in winner which will go in as a 16 seed.

The brackets were then spiraled into each other so that each set of groups was in each bracket section exactly four times, like this:

1234
4123
3412
2341
1234
4123
3412
2341
1234
4123
3412
2341
1234
4123
3412
2341

This helps smooth things out and ensure that the most rated fragrances didn't all have to face each other in the same bracket, as well as the user-submitted selections. The threshold for a fragrance to make a review-based appearance is 30 total reviews (user selections had no minimum). If you think that's too high, sorry; provide me with some more data next year and I'll try it again with a threshold of 15 or 20.

Voting is done with the use of Multi-Group Polls, I've tried this way of doing things before and it's pretty intuitive. You'll see there are multiple poll choices, split into groups of two. You may select only one from each group. And you may as well only pick one because picking both is the same as not voting at all. If you don't feel you know enough about two contenders to pick between them, you don't have to vote in that group. Again, the criteria you're voting on is "Which of these two fragrances would you most like to see advance?" This is deliberately vague so people can make choices based on reasons they're comfortable with. Maybe it's a holy grail of yours, maybe you love the monster sillage of it, maybe you just think the bottle looks cool. Whatever reason you want, it's a game. Whichever fragrance in each group has the most votes wins. In the event of a tie, the game will be replayed in the next thread.

Round 1 will start in early March, the total event will be 63 games over 26 days in 23 threads. Multiple games will be to each thread.

Round 1 = 4 games per poll (8 threads/8 days)
Round 2 = 4 games per poll (4 threads/4 days)
Round 3 (Sweet Sixteen) = 2 games per poll (4 threads/4 days)
Round 4 (Elite Eight) = 1 game per poll (4 threads/4 days)
Round 5 (Final Four) = 1 game per poll (2 threads/4 days)
Round 6 (Championship) = 1 game, 1 poll (1 thread/2 days)



That's just about everything, so let the games...begin!