FICS Teamleague

Board

Teamleague Forum

Section layoutT86Index ->

posted at 2022-10-19 19:07 by pchesso

Hi smallblackcat and all,

19 teams, is this it?

I submitted RW_Amarilla and RW_Naranja the way I did, hoping +100 rating points difference would be enough. It isn't.

Is there hope to get a late-joining team in the lower rating ranges, leading to 20 teams total, and possibly 4-4-4-4-4? Then I can swap HaStudent and antvrs (possibly yet another pair), so the new fixed average ratings would be RW_Verde 1824 and RW_Amarilla 1787.

If there is no late-joiner, I can merge the teams RW_Amarilla and RW_Naranja, leading to 18 teams total, and possibly 5-4-4-5, or similar.

posted at 2022-10-19 19:54 by kurumim

Hi, pchesso.

That difference is not enough simply because there aren't enough teams to build a fifth section. In addition to that, even if we had another team in the 1700s, with 4-4-4-4-4 and the new averages you've considered for RWVerde and RWAmarilla, Verde would no longer be in the third section and a clash would be inevitable.

The only team missing from the previous season is LightSpartans, whose averages are usually around 2000.

In an 18-team scenario, 5-4-5-4 and 4-4-6-4 would put MRM and MPM in the third section, which would be a disaster for them; 4-6-4-4 would avoid it at the cost of a one-player clash, and 5-4-4-5 has no clashes, although it may be a bit strange given how close NewBlood/Erythro and MPM are.

posted at 2022-10-19 19:55 by pchesso

Thanks, kurumim.

Regarding 18 teams: Yes, I noticed, too. I edited my post towards 5-4-4-5, but that was probably after you saw it.

Regarding one more team rated e.g. 1750, leading to 20 teams with possibly 4-4-4-4-4, I cannot follow you. Yes, that would put two RW teams in section 4. What's the problem? In that scenario, I can't see any player clashes except one (TM, MRM), which might be resolvable (and might appear in a couple other scenarios as well).

posted at 2022-10-19 20:17 by kurumim

Yes, I meant two RW squads would be in the same section, but indeed there wouldn't be any player clashes between Verde and Amarilla, you're right.

I talked to maxysgm (from VirtualRed) last week and it looked like they would be back, but I guess they're still struggling to get a full team.

posted at 2022-10-20 03:32 by KRMCHESS

For the MRM/TrickyMove clash I can switch keresch (from MKM) and mcstorytaller around so that is fixable. I plan to do a change but as mcstorytaller in in MPM, MRM and TM I figured I'd work out which team to remove him from before doing it in case there's another clash. I'll also add that if MRM and MPM did end up in the same section one of the teams would have to withdraw as there are only 7 different players over the two teams. For quite a while a thought I'd only enter 2 teams and then I got a late entry so I could just increase it to 3.

posted at 2022-10-20 16:19 by KRMCHESS

Sorry for the double post but I was looking at the Rainbow Warriors situation. So looking at it logically what pchesso has to do is aim to get one of his lowest two teams over 1834.50. When I checked his 3rd and 4th team they have an average rating of 1805.5 so the only way he can achieve that is to get player from his second team. This does seem a bit complex. If it does help for the running of TL I could do the following:

1. Move Keresch from MKM to MRM
2. Move mcstorytaller from MRM to MKM
3. Move Gargin from MPM to MKM

First two switches I think are enforced due to clashes. 3rd might make TL work better. Point would be that for the 3rd change it would lower the rating of MPM to 1754 so it lowers the target to get out of a 4 team division to 1781. So looking at the teams if you were then to swap AsDaGo/HaStudent and pifpaf both teams would be around 1800ish.

Does that sound like a sensible plan? There would be no clashes and a 4 team lower section would work. So in that scenario it would presumably be a 5-5-5-4 unless it works better to change two of the 5's with a 6-4 split.

If it would be helpful let me know and I'll do the changes

posted at 2022-10-20 20:40 by smallblackcat

I'm a bit late arriving to this thread, but I see that discussion has progressed. Assuming the lowest two RW teams are merged, it seems logical to have either a 4-6-4-4 or 4-4-6-4 split, depending on how that helps avoid clashes. I'm not opposed to some kind of split with 5s and 4s either, but I always think those are less desirable.

posted at 2022-10-20 20:56 by pchesso

RW_Naranja can be deleted.

posted at 2022-10-20 21:10 by smallblackcat

Las Naranjas are no more. Kenny, perhaps you can advise on what split works better for your teams regarding clashes. Ideally I'd like to finalise sections tomorrow.

posted at 2022-10-21 02:51 by KRMCHESS

I've put through the change so the mcstorytaller doesn't clash in MRM and TrickyMove and I figured the MPM change does make better sections.

If you approve the changes I've made all I think the clashes are resolved from my team's viewpoint. So I think both a 4-4-6-4 and a 4-6-4-4 split are viable.

posted at 2022-10-21 03:11 by pchesso

Thanks, KRMCHESS!

I much appreciate the shuffling ideas in your previous post, but I'm also fine with what we get now.

posted at 2022-10-21 03:32 by smallblackcat

Changes have been made, and I can confirm that either section split works. I'll finalise sections tomorrow, and do pairings soon afterwards.

posted at 2022-10-21 08:56 by kurumim

To avoid any crazy arms race, I'd say the alternative method of setting section caps, i.e., adding 20 points to the top team's average, looks much better for the present configuration than the traditional method of taking 0.25 from the average of the last team in the upper section.

posted at 2022-10-21 21:16 by smallblackcat

Well the alternative method has now become the standard method as far as I am concerned. I went with a 4-6-4-4 split, as the teams are marginally closer together in rating this way. Also this way Kasparov is essentially u1900 and Alekhine u1800 which looks neat.

posted at 2022-10-22 09:27 by kurumim

I mentioned it because in the previous season all section caps were set using the old method. Of course it all depends on the gaps between sections, and as of now the new method is pretty effective in keeping the original balance.

Thanks once again for your top-notch work, sbc, and good season to all!

posted at 2022-10-22 22:17 by smallblackcat

Pairings are up. As always, let me know if you spot a mistake.