# poll: how do you seed your tourneys and why

Discussion in 'Archived Threads 2001-2004' started by george kaplan, Jun 27, 2002.

1. ### george kaplan Executive Producer

Joined:
Mar 14, 2001
Messages:
13,063
2
Trophy Points:
0
One thing that always spawns controversy (not necessarily a bad thing) in the tourneys is the match-ups. I thought it might be interesting to debate some of the ways that these come about, i.e., the seeding of the tournaments.

I know some are completely random. I think that's a pretty good method, although it does allow for the possibility of some first round match-ups of the best films that really should be meeting much later in the tourney.

For myself, I use a complex method. I won't bore you with the details, but I develop one seeding based on my own evaluations of the films, and a second seeding based on my predictions of how HTF voters will view the films (I'm actually a pretty good handicapper). These are averaged, and then a probability model is used to throw in some randomness (the top seed meets the bottom seed with some probability, the second to the bottom with a smaller probability, the third to the bottom with an even smaller probability, etc.). I'm a statistician, so I find doing this fun, but I'm sure others have much simpler methods.

Anyway, just wondering what people do and what they think of the various methods.

2. ### Bill J Producer

Joined:
Oct 27, 2001
Messages:
3,970
0
Trophy Points:
0
For both of my tournaments I used the completely random method. For some strange reason I would feel as if I were "fixing" the tournament if I did it any other way.

By the way, George, I find your method very interesting. I had no idea that your method was that complex.

3. ### Scott Weinberg Lead Actor

Joined:
Oct 3, 2000
Messages:
7,477
0
Trophy Points:
0
I do a random seeding, except in that Comic Book one I did. I doctored the brackets to make the most interesting or appropriate matchups, not because I cared who the winner was.

Like Bill, I'm also impressed with your seeding "formula".

4. ### SteveGon Executive Producer

Joined:
Dec 11, 2000
Messages:
12,251
1
Trophy Points:
0
I ike pairing up similiarly-themed films but if that isn't possible, then it's pretty much a random affair.

5. ### Rain Producer

Joined:
Mar 21, 2001
Messages:
5,015
0
Trophy Points:
0
Totally random.
Why? Cuz then nobody can bitch.

6. ### george kaplan Executive Producer

Joined:
Mar 14, 2001
Messages:
13,063
2
Trophy Points:
0
You may have a good answer when they do, but they can and will still bitch when the two best movies (randomly selected) get pitted against each other in the first round.

7. ### Steeve Bergeron Cinematographer

Joined:
Feb 3, 1999
Messages:
2,541
0
Trophy Points:
0
Real Name:
Steeve Bergeron
Wow, George, I'm also impressed by your method. I think only a statistician would come up with such a complex method.

But, why are you so concerned about having the two best movies in the final round? They will have to meet at some point anyway. I mean, even in sports, it happens all the time. Just look at the NHL for example, rarely you'll have the two best teams at the end of the playoffs.

Personally, I think it's the journey that is really important. I have absolutely no problem with someone using the random method.

8. ### george kaplan Executive Producer

Joined:
Mar 14, 2001
Messages:
13,063
2
Trophy Points:
0
I'm really not. It's more a concern about them not meeting in the first round. Basically, you'd ideally like the better films to meet later on. In the end it doesn't matter, but lots of us (myself included) complain about first round match-ups.
Actually, way back when I first did these, I seeded them publicly and when the #16 seed beat the #1 seed I called it an upset. But then lots of people would complain that it wasn't an upset, cause my seeding was wrong. Now, I just seed them and keep the results to myself .

Joined:
Feb 26, 2002
Messages:
1,721