As can be gathered from reading my team's (Team USA: Northwest's) group notes page, most of us are getting a little tired of being unequally-matched in World League matches where most or all of us are paired up against players who are rated 200-400 points higher than we are. Some have suggested returning to a system such as the rating divisions that were in place during the World League practice round. I don't like this idea because it is aimed at making the outcome of team matches less dependent on the strength of the teams, and I am not uncomfortable with the idea of a stronger team being able to beat a weaker team. I don't see the fact that strong teams have a better chance of winning as a problem that needs to be fixed. What I do see as a problem is that the current system does not favor teams being consistently matched up against teams of similar strength. I looked at some of the other scores in Group B matchups, and I see that some of the other teams have been having the same problem that we have, i.e. being repeatedly out-rated. I propose some variation of the following system, which is derived from the Backyard Sports series of children's computer games (I played Backyard Soccer a lot when I was a kid), as well as from the system in place at the annual Corus tournament in Wijk aan Zee:
At the end of every World League season, the top 1-3 teams in a group (we'd have to pick a number and stick with it, so as to avoid changing the number of teams per group) move to to the next group up, and the bottom 1-3 teams would move to the next group down.
Here's an example of how this might work in practice: At the end of this season, the top, say, 2 teams in the B group move up to the A group, and the bottom 2 teams in the A group come down to the B group. The bottom two teams in the B group would go down to the C group and would be replaced by the top two teams in the C group. Eventually, the groups would even out so that the strongest teams are in Group A, the next-strongest teams are in Group B, and so on. Unlike rating divisions, this system would establish team wins as a reward for improving as a team, while allowing the weaker teams in a group to move down to a level where they would have winning chances, rather than being forced to repeatedly play matches where they are grossly out-rated and out-ranked.
As I already mentioned, this system is not without precedent in the world of top-level chess, where the annual Corus tournament in Wijk-aan-Zee is divided into groups, and the winner of the B group is traditionally invited to play in the A group the following year.
So, what do you all think? If you like this idea, please respond to tell me. If you don't, please post to tell me why. I welcome both positive and negative feedback. Above all, if you have any suggestions for how to improve this idea, please post them.
I like the idea, and was considering proposing something similar. Also it might make sense to make the Sections a bit smaller because right now there is such a wide range of levels within each section. This would also have the advantage of making the season a bit shorter, and allowing final playoffs, should they be desired.
In the U.S. case, I was wondering if the teams should be reorganized, dividing the U.S. into four teams rather than six. That might give each team a better chance against powerhouses like Team England or Team Philippines.
thanks for sharing it PhillipN ... a brilliant idea !!! ... a good road to take, WL Admins.
This is also done in the IIHF for World hockey championship qualifying, the English Premiership league, and countless other bodies. The only reason why we shouldn't do such a thing, as far as I can tell, is that a lot of the distribution of teams this time around was based off of numbers of players. If a team dropped it may punish a significant number of players, making them unable to play; for instance, in Round 1 Canada vs. England managed 123 players each, while the highest number of players for a team in group B was 74 (Iran vs. Russia).
Although, after looking at it, from what I can tell besides the some instances with with massive participation (see the first three games of Group A, Round 1), most divisions don't have that much difference between numbers participating so it wouldn't be that big of a problem after all.
Thanks for the input, JDudar. That problem had not previously crossed my mind, possibly because our opponents have routinely had about a third to a half of their players (counting just the ones who sign up for the matches) below the "cutoff line." As a result of this, some of our bottom-rated players (800-1000) have been routinely paired up against players rated over 1500 (and occasionally against players over 1600), while the other teams' comparably-rated players have been waiting way at the bottom of their teams' lists (or off playing in the World League tournaments for players who are unable to participate in rounds due to low seeding).
We can't change rules in this WL tournament. This discussion only for next WL. I think that is bad idea becouse give chances to weak teams but the main target of WL - find strongest team.
It's a start, but it doesn't go far enough imo.
One of the key issues is how the size of the team impacts too much on the outcome of matches. It won't matter which division you're in, if you are up against a much bigger group with unlimited boards, then you'll still be hopelessly outgunned.
It's pretty stupid forcing the smaller group's weaker players to play vs midstrength players because of numbers alone. Against Team England, we have some players who are over 750 rating points mismatched. The weaker players of Team England won't even get games. Under such conditions, it won't matter -- even if our top boards do a clean sweep, we are most likely going to lose because of the silly situation at the bottom of the matchup.
Forcing smaller groups (note, not necessarily weaker, but smaller) into the B division doesn't seem all that great to me as a solution.
I think I would prefer rating restrictions, ie make the division between A grade, B grade and C grade apply to player's ratings rather than the whole team. Add a limit to the number of boards played in a match and it starts to become much fairer. This is how it works in OTB meets, I see no reason why it wouldn't work here.
In next WL...for teams which have size >150 players we can play into two reit groups: <1800 and >1800). For teams which have <150 players we can play into two reit groups: <1600 and >1600. I think we need not many reit groups. Two is good.
The goal of the World League as it stands is mainly to give as many people a chance to play as possible, and it probably does not have much competitive significance. Large teams like Team Philippines or Team England are at an advantage because of their strength on lower boards, but creating a bunch of strict rules that would be difficult to enforce wouldn't really change that situation. Creating smaller sections with teams that are similar in strength might make things easier on some of the teams that are currently struggling, and the idea to promote the top finishers to the next section up would give teams something to strive for.
I agree; we have had a similar problem vs. the larger teams in our group. Oftentimes we have fewer than two-thirds the size of the groups we are playing, and we have a similar problem with the rating matchups. If the system I proposed were used, the teams that have more boards (that is, the larger teams) would tend to move up to the higher divisions as they would be harder to beat due to their large size, as you said. Also, teams that have an extraordinarily high average rating (such as Team Russia, with an avg. rating over 1700 and whose lowest board in our 47-board match is rated over 1700) would be able to play against the large teams, rather than easily winning all their games against teams that have a more normal rating distribution.
One other thing to take into account as you consider my suggestion is the fact that some teams have their "size" dictated by their participation level as much as by their actual size. Our team has over 150 members, but we've never had more than 56 players sign up for any one match, and our participation numbers have been falling (from 56 to 47) as more people get the idea that it's pointless to join matches when you always end up losing both your games because you were paired up against opponents who are rated 200-400 points higher than you are. There are some teams in our group that I think we can beat, but we had a tough draw early on (USA: Southwest, Iran, Romania, and Russia) and people are getting tired of playing against much-highe - opponents so that I don't know whether we're going to have enough players to win once we start playing easier matches. Plus, a very strong team that is not quite as large as the other A teams (but whose players are strong enough that 95% of them would not be significantly out-rated if they were to play a larger team) would be able to play against teams that pose more of a challenge rather than torturing smaller teams that have a more normal rating distribution.
You think you guys had a tough draw. We played Team England!, Team Philippines and Team Australia in the first three rounds! What a bloodbath!
OMG, now that is a tough start!
Phil, under your proposal, what, Team England and Team Phillipines would be in group A alone? No other team can come close to their size, so I guess so.
Do you see how your solution just doesn't help enough?
How about 8 teams in each section instead of 12?
how about just flipping a coin each match so strong teams wont have an 'unfair' advantage.
In other words, if the WL organizers were to implement this idea, and decided that the top/bottom 3 teams should move up or down respectively, then the 3 teams that had to move down to Group B at the end of the season (that is, the lowest-ranked 3 teams) would be replaced by the top 3 teams in Group B, so as to keep the number of teams per group at 12. If they decided that only the top 1 team in the group moves up and the bottom 1 team moves down, then what would happen is the bottom team in Group A would move down to Group B, and the top team in Group B would move up to Group A, so as to keep the number of teams in Group A at 12. Some have suggested having fewer teams per group, and I think that the principle behind my idea (i.e. maintaining the group sizes while evening up the matches simply by having the bottom teams in Group A and the top teams in Group B "switch places" at the end of the season, and the bottom teams in Group B and the top teams in Group C, etc.) would work even if the number of teams per group were changed, although the competition would be better balanced if the ideas were implemented in this way: the smaller the group size, the fewer the number of teams that move up/down at the end of the season. In other words, if the number of teams that get to advance at the end of this season is 3 per group, then the number of teams that would get to advance at the end of an 8-teams-per-group season would be 2. If 2 teams advance out of 12, then if the groups were shrunk to 8 the number of teams that get to advance would shrink to 1.
Does this answer your concern, BorgQueen?
8 teams per section with 2 teams moving up and down sounds pretty good.
It's great to see all this discussion! Thank you for your input, everyone. I heartily agree that one of the biggest problems faced by the WL is the disparity and size and therefore ratings.
I think it's certainly a plausible idea to move a couple teams up and down at the end of each season. Earlier on I personally was a little more focussed on getting teams of as close to the same size as possible together in groups, but as we know the actual size is far less important than the active size of each team, and there is no better way to determine which teams have more active players than others than by seeing which teams tend to come out on top most often.
As soon as we start regrouping teams by their number of active players next season, I think the size and ratings disparities will become a lot less of a problem.
However, teams like Team England, I really don't have a solution for, and that's something I'd really like to discuss here. As it stands, Team England and Team Philippines stand to sweep all their games without much opposition, and it doesn't look like that's going to change unless somebody comes up with something good that doesn't seriously limit the number of players allowed to play.
Any team playing Team England or Team Philippines is going to lose a lot of games on their lower boards, but maybe we can give people other goals to strive for: special trophies for most brilliant game, most impressive upset, best individual performance and/or most valuable player on each team. We could have a thread where people nominate games or players for special recognition each round or at the end of the tournament.
Unfortunately, my idea would probably result in some teams being bumped down to levels where their bottom-board players would be unable to play (most of the time, anyway, as such a team would probably not be much larger, in terms of active members, than the top teams in the group to which it was bumped), and exceptionally strong teams (such as Team Russia, with something like 2/3 of their active members rated over 1700) would work their way up to levels where their opponents would have similar ratings (their average boards would be evenly matched against their opponents' above-average boards), but their opponents' average players would, unfortunately, be cut from the match. But as is Team Russia's bottom players are being consistently cut from matches while their opponents' bottom-board players are out-rated by 750-800 points! Either way, somebody gets cut from the match. I can't think of any way to deal with this problem, other than the World League tournaments (beginning a day or two after the World League matches so that lower-board players will have a chance to check to see whether they ended up in the match) for players who get cut from their matches.
In light of the scenario brought up by Masterful_Forfeit (a team being bumped down to a group with smaller teams), the difference it would make is this: If the team gets bumped down, its lowest-board players would play in the WL tournament rather than the match, but if the team doesn't get bumped down, its lowest-board players would play in matches where they would usually be out-rated by at least several hundred points.
One solution I considered to the problem of unequal team sizes was the possibility of having team standings be affected by their players' placement in the World League tournaments (in addition to their team wins & losses), but this would directly benefit the larger teams, who would be able to enter more (and higher-rated) contestants in the WL tournaments.
Any suggestions, guys?
Join Chess.com for free to add your comment! Already a member? Then login now to comment.