late to the party on this thread, so ill try to play catchup, going off wyzdm's list of topics. apologize for the length, but this stuff always intrigues me, so i want to hit each topic:
pausing - i am definitely in favor of allowing pausing. i basically feel like the details wyzdm mentioned, like how long you can pause etc, *should* not be an issue. not to suggest they aren't - but i would hope people would bring some level of maturity to these clan battles, or other clans will simply refuse to play them. anyway, some guidelines still may be needed, but i would suggest making them loose, assuming the large majority of people pause in good faith, not to abuse - until experience proves otherwise! anyway, id go with maybe 5m max pause per game, 10m per prep. going past this is a forfeit of the game/set/all sets, as appropriate. also i suppose excessive pausing would need to be restricted, maybe max of 3 times/game, 5 times/prep? obvious if people hit these limits regularly it would be ridiculous... but i think these limits are reasonable for the sake of having a limit someone can point to in the case when people abuse the system. i would envision an "extended rules" type section at the bottom where all these details 95% of people don't care about can find a home...
game time - i think 15m is probably a bit too short, i would be ok with more like 20m. deciding who won could be difficult, tower control does not imply map control, people go on the offensive and retreat to regroup all the time... so if there isn't a simple metric for who is winning, you pretty much have to call non-obvious cases a draw, which does not accomplish much. maybe a 20m draw forces the next game to be on holloween or swamp? or maybe castle is only used to settle draws? :) or maybe we just let games end in a draw if need be, and sets could still end at 3 games regardless.
- stalling - what do you mean? just extending a game that is clearly over? i think the rules are not the place to solve this. its up to veteran clan members to get their members in line. i have noticed one or two of the SoS guys never surrender, and plan to discuss with them in hopes they will at least do so in clan events - people need to respect the time of others. if its not clear to you its over and you carry it on a bit too long, thats ok, but forcing a statue kill every game is getting ridiculous, especially given the caliber of players in these events - its not complete idiots playing here. i think clan leaders can respectfully let the other side know when certain people are making a habit of stalling, and leave it to the guys on the other side to take care of it. this is still a new thing, it will take some time for the community to develop the desired sense of gamesmanship...
lag/slow speed - hopefully this is another that can be resolved by clan leadership. if you want to have a respectable clan, you bring people who are reasonably mature who dont lag badly or OOS or anything like that. i recognize the clan i am in has some problems with the lag part :) its being worked on... anyway, i suppose any time a 2:1 is in play (2 real seconds per game second) or there is excessive delay (half second or more consistently), either party should be able to cancel the match, and both players can restart SE/close windows/whatever and try again. if someone can't get to the point where they can participate without excessive technical issues, they should have to be replaced. cancelling of the match should have to occur fairly early in the game - is 2m enough? it wouldn't be fair to let people cancel at the end after they already lost... maybe 3m? i dont know.
game map/size - i think maps should be allowed or not, not broken down by matchup. there is too much subjectivity in what matchups have an advantage where, and it varies by skill level of the players. at the map level, if there is a clear case not to play it at all, i think that could justify not allowing a map (except through mutual agreement), but i don't think a map being fair sometimes and not others is a strong enough argument. i measured the size of the maps once, in terms of crawler-seconds from end to end
castle - 16s
swamp/holloween - 19s
desert - 22s
ice/forest/gates - 27s
green hills - i forget but huge, like 36s
there has been a question of desert being the same size as swamp/holloween, so i wanted to test it myself. i just want to point out those maps are, in my opinion, substantially different because the placement of walls on swamp/holloween is not really sufficient to protect the second layer of gold, while desert allows the wall to be placed past the miners on the far side of the farthest gold patches... which is a pretty big difference.
anyway, i agree with wyzdm and others about removing castle. i thought the point about the game coming down to mechanics, any misclick or small error could decide the game, was a very good one. to me, there are 3 major elements to the quality of a player's game - micro, macro, and strategy (army composition, response to enemy army composition, etc). i think the fair maps strike a reasonable balance between the 3. IMO castle is pretty clearly a micro map, and starting strategy i guess, but those have such a heavy role, with a disproportionately large number of games being decided so early. i think that unbalance is cause for exclusion, but i also thinking allowing a player to strike 2 maps is a perfectly reasonable and acceptable way to solve this problem. also, the order v chaos advantage seems unbalanced to me.
i also think GH is a little unreasonable, by very similar logic. the early game and micro definitely matters in getting the center, that is true for all maps - but games drag on forever on that map, over emphasizing the late game, it just doesn't seem balanced to me. the defensive gimmicks that can pass on GH more than anywhere else also make me not a fan for clan events, which are supposed to be the most "serious" or "professional" of all games. there also seem to be order v chaos imbalances. i personally don't see the order disadvantage that many others do - which is to be expected, i haven't gotten to play much order - but it seems to me the late game favors order while they lose the initiative advantage in the early game. to me, those reasonably offset each other - but i think a lot of people would disagree about the late game favoring order. anyway with so many people seeing order v chaos issues, regardless of what they are, i think its a decently imbalanced map. but again, striking 2 seems acceptable to me - i always strike GH and castle and probably always will.
sets/group sizes - i think 2 or 3 people per team should be the group sizes. when there are 4 or more people, i am in favor of splitting into group A and group B (and C and D, hopefully someday!) - otherwise, it just takes too long. today, it seems preps overly favor clans that are strong at the top, rather than those with a lot of quality players. i am not really a fan of this, although its hard to devise any system where this is not the case. when 5 guys show, only the top 3 affecting the outcome doesn't seem like the right way to do it, to me. a big part of the value of clan events is making clans more attractive to the user base - so these events should be as inclusive as possible.
i originally proposed (in one of the several sos/tz prep chats) having group B be just for fun, not count towards the win. this was partly to allow group B players to double as group A backups, and partly because i thought there was a more established system than there was, and didn't want to suggest too drastic of a change. anyway now that i realize everyone is in the process of figuring this all out now, i think it would be better to have group B count, too. the more we can include people other than the top few, the better! however, losing those people as group A backups would suck - maybe they can get pulled from group B after any set and don't compete their group B sets once pulled into group A? anyone have any better idea? i don't think the # of people in each group, particularly group B, should have to match. what is wrong with a 2v3? i don't see a problem.
i think most people favor going by sets to determine the outcome, not games - meaning if the team with more sets lost more games than they won, they still win. for example, in a 3v3, if one team wins 5 sets 2-1, and the other wins 4 sets 2-0, the first team's record is 10-13. does the team who won more sets, or more games, win? i am not sure either answer is clearly the way to go, but i personally feel sets are better, as a set better indicates which player is better. however, in the event of an even number of sets (as is the case in 2v2, or, if allowed, 2v3 - or even in a 3v3 where the last set cannot be completed by either side), i vote for game count to be the first tie breaker. i am personally ok with draws as the final outcome - if you guys don't want draws, i think a final set or game where winner takes all would be fine.
anyway, by whatever method a single group match is decided, you still have the complication of dealing with 2 groups, if people decide to allow that and have ample members show up. i would vote to still count all sets in the total, recognizing that group B may not play out completely if people leave and groups shift.
actually, i think the whole issue of substitution should be discussed, wyzdm, id add that to the list. can anyone sub any time? do they have to show up on time and be designated a backup? recently a player left and the outcome of the game was decided by the final set, where wyzdm ended up being the sub. without regulation (and im not suggesting they would do this), TZ could have someone leave any time wyzdm couldnt make the start, and have him sub in, which is a pretty big advantage. any clan can do it, just one obvious example... i think in the spirit of getting things going, people should be able to sub without penalty, but we should consider what the "right" way is to do it. some people will want no subs, you forfeit if you leave. that is probably too harsh to get things going, and maybe even down the line, with this being a relatively casual game. maybe some system can be devised, but i cant really think of a good one, hopefully someone has a good idea?