Season 10 Ending

[nabo.]

Season 10 will be ending end of this month (Oct 30-31). Exact time will be posted as date approaches.

If you have any new interesting ideas or suggestions to start a new season, now would be a great time to post. Dont hold back~

[dweiler]

I would like for next season to try randomized teams every game, and the one who has the most positive win-loss record in the end wins. I am really curious how it will turn out to be. For example, if top Elo players play vs bottom Elo players and the bottom get good rolls and the top not, will the top still win, etc? And will randomized teams give a different ranking? It's a known fact that the way of measurement can have a big influence on the outcome, and after 10 seasons it's maybe time for a different measurement.

[nabo.]

@mickeythemousie

What do you mean by "randomized teams". Do you mean sorting teams via shuffling instead of autobalanced by elo?
Winner will be highest elo, yes?
Basing it on w/l can be difficult since number of games played differs for everyone.

[dweiler]

@mickeythemousie

What do you mean by "randomized teams". Do you mean sorting teams via shuffling instead of autobalanced by elo?
Winner will be highest elo, yes?
Basing it on w/l can be difficult since number of games played differs for everyone.

Yes, I mean without Elo, just shuffling and having the w/l stats of the players as their ranking.

I agree that this can be a bit of a problem. For example, a player with 600-500 has a score of 100, while someone with 80-10 seems to be better, but has 70 score. But on the other hand, activity should be rewarded too, so maybe it is good that the 600-500 one will win the season in this case. OR if there are better suggestions to rank that's cool too, I just like the idea of a 'random' instead of 'balanced' season.

[nabo.]

@Mickeythemousie

So you want teams randomized every game and something like +1 point for win -1 point for loss and player with highest score end of season wins? Or, you wanted elo which considers score difference between both teams, whether elo is accurate measure of players' skill or not instead of the +/-1 points.

[dweiler]

Yeah, the first one you mention, every game is the same as !shuffle, and the player with the best w/l at the end wins. So someone with 30 wins, 10 losses has 20 score, and with 10 wins, 30 losses has -20 score.

[nabo.]

@MickeyTheMousie

Last question: Based on the +/-1 w/l rankings, which would be better?

Would it be better to 1) just shuffle randomize teams or 2) autobal based on score (wont be elo, but w/e value you end up which is based on your w/l)?

[dweiler]

I would like to see 1) just shuffle randomize teams.

The reason for me is like this: in the current system all games are balanced in such a way that the skill is '(as) evenly (as possible)' distributed across the teams. In other words, the odds are usually around 50/50 of winning the game based on the players who you team up with. Therefore, the thing that decides most games is not the skill of the players, but the rolls. If you randomize the games, the skill factor of the players becomes bigger again, and the rolls will get a less important role. It seems like a good thing to me to see games being decided on skill again instead of rolls.

[nabo.]

In my opinion, there is nothing wrong with trying different rankings time to time. So, I think its a great idea.
In dihl, we use +/-1 w/l per game and +/-2 extra points for captains. We vary rankings every season by changing to lets say extra +/-2 for captains +/-1 for co-captains vs +/-1 for captains only, etc.

As long as the rankings is fair for everyone, it should be interesting and perhaps a new challenge spicing things up.

Edit: Don has brought to my attention about leave and dc penalties. To make this work, we will probably have to go with +/- 15 points.

[Diablo_]

I'm not necessarily against this suggestion (tho I would prefer the current one), but before making such heavy changes I highly suggest to make a poll for it, or the drama will be huge (which will increase anyway as in "omg I don't want to play this, the teams are so unfair").

[nabo.]

Feel free to make a topic

[cam_breezy]

i think this being a clan we should practice more customized games with captains, its sad that 99% of games are hp prophet. they don't even bother trying different stuff. we didnt do fun week this season and we dont even do tournaments. where is the clan play?? might as well just open a 1700+ bot for the better players instead of even having lihl.

[nabo.]

i think this being a clan we should practice more customized games with captains, its sad that 99% of games are hp prophet. they don't even bother trying different stuff. we didnt do fun week this season and we dont even do tournaments. where is the clan play?? might as well just open a 1700+ bot for the better players instead of even having lihl.

-Use !challenge more if you want captains games.

-1700 will not work. Think about it.

-As for your desire to want to play other modes, we could add !votemode if its hard to communicate to others you want another mode...But you know, if others wanted to play sdgmcb, they would have played it by now since as long as everyone agrees, you can play other modes aside hpgmcb phgmcb.

[eldryan]

the top 10% can be rewarded some sort of recognition, instead of just the top 1. I think that would be a more accurate recognition, as the top 1 is often in flux in various parts of the season and anyone in the top 10% is really still in contention. Can also reward anyone over 1200 ELO, as that's generally a distinguished finish mark.

I think the idea of shuffle is dumb. w/l will be based more off of team selections instead of rolls and micro play, which is stupider. I'm 90% sure most of LIHL agrees. if you want to try top players vs bot players, go captain mode for +.5 - 29.5, no need to do it 100 times. I did this myself when I got bored at 1635 ELO, and started playing me + gog vs bot players. We won like 4 for every one we lost, and I dropped 100 ELO very quickly.

[SLSGuennter]

I doubt this is a good idea, since having 52% wins and simply spaming games could make you #1 ... As long as you are above 50% wins, you can get high ranking while being worse then a lot of other players, which play less games.