Here’s an idea for tourney seeding rankings. Please applaud or bash at will.
Teams receive weighted points for each win or tie, based on the division of their opponent.
WINS
A = 6 points
B = 5 points
C = 4 points
D = 3 points
E = 2 points
F = 1 point
TIES
A = 3 points
B = 2.5 points
C = 2 points
D = 1.5 points
E = 1 point
F = .5 point
Ties are fun, especially when teams from different divisions play. Slappers (an A team) tied HFB (D team) earlier this year. In this system, HFB earns 3 points for tying an A team, Slappers earn just 1.5 for tying a D team.
This weighted point value could then be used as is, added to a team’s weight (weighted points + 6,5,4,3,2,1) or multiplied by a team’s weight (weighted points * 6,5,4,3,2,1). Since all teams have now played nine games, here are the resulting rankings…
Here’s an idea for tourney seeding rankings. Please applaud or bash at will.
Teams receive weighted points for each win or tie, based on the division of their opponent.
WINS
A = 6 points
B = 5 points
C = 4 points
D = 3 points
E = 2 points
F = 1 point
TIES
A = 3 points
B = 2.5 points
C = 2 points
D = 1.5 points
E = 1 point
F = .5 point
Ties are fun, especially when teams from different divisions play. Slappers (an A team) tied HFB (D team) earlier this year. In this system, HFB earns 3 points for tying an A team, Slappers earn just 1.5 for tying a D team.
This weighted point value could then be used as is, added to a team’s weight (weighted points + 6,5,4,3,2,1) or multiplied by a team’s weight (weighted points * 6,5,4,3,2,1). Since all teams have now played nine games, here are the resulting rankings…
i like it!
this is actually similar to something i’ve been playing around with, though i used different points values. I figured each team plays 2 games in the division above and 2 games in the division below. (except A and F teams who just play 2 games out of their division, not 4 like other teams)
the points i found worked out nicely (without additions or multiplications) were:
WIN ——-
A - 10
B - 9
C - 8
D - 7
E - 6
F - 5
TIE ——-
A - 5
B - 4
C - 4
D - 3
E - 3
F - 2
seems like this way a dominant C team can seed higher than a middle of the road A team. Though, I suppose it’s open to discussion whether or not that really reflects the relative strengths of the teams.
i think i really like the last scheme that brendan proposes with the multiplication by weight. looking at the rankings that generates, they feel to me like the closest to accurate.
Here’s an idea for tourney seeding rankings. Please applaud or bash at will.
Teams receive weighted points for each win or tie, based on the division of their opponent.
WINS
A = 6 points
B = 5 points
C = 4 points
D = 3 points
E = 2 points
F = 1 point
TIES
A = 3 points
B = 2.5 points
C = 2 points
D = 1.5 points
E = 1 point
F = .5 point
Ties are fun, especially when teams from different divisions play. Slappers (an A team) tied HFB (D team) earlier this year. In this system, HFB earns 3 points for tying an A team, Slappers earn just 1.5 for tying a D team.
This weighted point value could then be used as is, added to a team’s weight (weighted points + 6,5,4,3,2,1) or multiplied by a team’s weight (weighted points * 6,5,4,3,2,1). Since all teams have now played nine games, here are the resulting rankings…
I like it, too. We get to keep playing in fun, competitive matches week to week with the opportunity to actually fight for a better seed outside of that pre-determined range we’ve been given this season.
The New & Unimproved HFB Mandy - 12 March 2010 10:51 AM
Brendan - 12 March 2010 10:21 AM
Here’s an idea for tourney seeding rankings. Please applaud or bash at will.
Teams receive weighted points for each win or tie, based on the division of their opponent.
WINS
A = 6 points
B = 5 points
C = 4 points
D = 3 points
E = 2 points
F = 1 point
TIES
A = 3 points
B = 2.5 points
C = 2 points
D = 1.5 points
E = 1 point
F = .5 point
Ties are fun, especially when teams from different divisions play. Slappers (an A team) tied HFB (D team) earlier this year. In this system, HFB earns 3 points for tying an A team, Slappers earn just 1.5 for tying a D team.
This weighted point value could then be used as is, added to a team’s weight (weighted points + 6,5,4,3,2,1) or multiplied by a team’s weight (weighted points * 6,5,4,3,2,1). Since all teams have now played nine games, here are the resulting rankings…
I like it, too. We get to keep playing in fun, competitive matches week to week with the opportunity to actually fight for a better seed outside of that pre-determined range we’ve been given this season.
I’m not sure whether it is the same thing that Travis is proposing, but the best solution I can think of so far involves assigning different point values for wins and ties against teams in the various divisions,* so that a teams’ playoff seeding is determined not just by their end of season record but also by how they did against the easier and tougher opponents.
* For instance, a win against an A team is worth 10 points, a win against a B team is worth 8 points, a win against a C team is worth 6 points, etc. And then a tie against an A team is worth 6 points, a tie against a B team is worth 4 points, etc.
I support this system, although I wonder whether you really need to multiply a team’s “raw score” by a “weight” (because you’ve already taken into account the relative “value” of the team’s ties and wins by alloting a different point value to them).
I support this system, although I wonder whether you really need to multiply a team’s “raw score” by a “weight” (because you’ve already taken into account the relative “value” of the team’s ties and wins by alloting a different point value to them).
After looking at these numbers a bit, I think the WEIGHTED * WEIGHT ends up being the best solution of the three. The raw WEIGHTED score allows for too much movement up and down the rankings (HFB, the #1 D team, is tied with the #2 and #3 A teams and Slappers, the #8 A team is tied with Army, a middle of the pack D team).
Sure, there are a couple anomalies, like The Hustlers, but I think that’s mostly due to an incorrect preseason division placement, which will surely be corrected next season, provided you guys come back and don’t request a weak schedule.
I support this system, although I wonder whether you really need to multiply a team’s “raw score” by a “weight” (because you’ve already taken into account the relative “value” of the team’s ties and wins by alloting a different point value to them).
After looking at these numbers a bit, I think the WEIGHTED * WEIGHT ends up being the best solution of the three. The raw WEIGHTED score allows for too much movement up and down the rankings (HFB, the #1 D team, is tied with the #2 and #3 A teams and Slappers, the #8 A team is tied with Army, a middle of the pack D team).
Sure, there are a couple anomalies, like The Hustlers, but I think that’s mostly due to an incorrect preseason division placement, which will surely be corrected next season, provided you guys come back and don’t request a weak schedule.
i totally agree with the WEIGHTED * WEIGHT being the best solution so far. if you wanted to tweak it a little more, you might try using different point values/scales to do the weighting, and see how it falls out.
I support this system, although I wonder whether you really need to multiply a team’s “raw score” by a “weight” (because you’ve already taken into account the relative “value” of the team’s ties and wins by alloting a different point value to them).
After looking at these numbers a bit, I think the WEIGHTED * WEIGHT ends up being the best solution of the three. The raw WEIGHTED score allows for too much movement up and down the rankings (HFB, the #1 D team, is tied with the #2 and #3 A teams and Slappers, the #8 A team is tied with Army, a middle of the pack D team).
Sure, there are a couple anomalies, like The Hustlers, but I think that’s mostly due to an incorrect preseason division placement, which will surely be corrected next season, provided you guys come back and don’t request a weak schedule.
Fair enough. Frankly, I like any of these options better than the current seeding methodology.
Looking at the way this ranks all of the teams I think it does a pretty good job. after 9 weeks we all have a pretty good idea who is playing well and I would say all of these ranking options are way more accurate then the A then B then C… we have now.
-
I do like the divisions having some sort of weight though, it gives a team playing a hard schedule some motivation to continue there instead of dropping down and blowing people out.
de-jon/ - 12 March 2010 10:42 AM
seems like this way a dominant C team can seed higher than a middle of the road A team. Though, I suppose it’s open to discussion whether or not that really reflects the relative strengths of the teams.
I think last night shows that a dominant C team should be higher than a middle of the road A team. (Skeets over slapers / SLE over danza)
Brendan - 12 March 2010 01:49 PM
After looking at these numbers a bit, I think the WEIGHTED * WEIGHT ends up being the best solution of the three. The raw WEIGHTED score allows for too much movement up and down the rankings (, the #1 D team, is tied with the #2 and #3 A teams and Slappers, the #8 A team is tied with Army, a middle of the pack D team).
I think HFB or Army would be very competitive with the bottom of the A division.
Weighted * Weighted does a nice job of creating a balanced tournament without having to utilize a seeding committee like kickball. Teams are able to improve their seeding beyond their division, while also preventing too many upper teams from having to compete in the earlier tiers of the tournament. The divisional multiplied values actually seem to work out pretty nicely in that list, but the values could easily be tweaked if it was felt that they were creating too little/great of a change.
Most importantly, looking forward to future seasons, putting a system like this in place before the divisions are finalized allows all the teams to decide what they’d like their season to look like while also knowing how it affects their seeding, which should also help in creating a balanced and fun tournament/season.
I think last night shows that a dominant C team should be higher than a middle of the road A team. (Skeets over slapers / SLE over danza)
Keep in mind that a couple upsets in a 270 game season might not be solid justification for the inadequacies of a ranking system.
Cowhide - 12 March 2010 03:01 PM
Weighted * Weighted does a nice job of creating a balanced tournament without having to utilize a seeding committee like kickball.
To me, that’s one of the most important goals. Remove as much subjectivity from the seeding process as possible while simultaneously delivering a balanced tournament based on the results of the season.
I think last night shows that a dominant C team should be higher than a middle of the road A team. (Skeets over slapers / SLE over danza)
Keep in mind that a couple upsets in a 270 game season might not be solid justification for the inadequacies of a ranking system.
Not to quabble nor puff the Skeets up as grander than they are, but it should be noted that SLE and Skeets are a combined 16-1-1 with a week left where they will face each other. They are also a combined 2-0 vs. A teams and 2-0 v. B teams. And 2-0 vs. D teams. The one loss came of these two teams was in a short-handed game. So, perhaps Ballerina’s conclusion isn’t based on a “couple upsets”. I haven’t seen all the scores yet, but I did note that another C team tied an A team and that a D team beat a B team.
