Skip to main content
Welcome guest. | Register | Login | Post

Qualification groups for Nexuiz and OpenArena 1vs1 tournament

19 replies [Last post]
libervisco's picture
Offline
Joined: 2006-05-04

This is just a proposal/discussion topic. Official groups will be posted on freedomware-gamefest.com.

For Nexuiz 1vs1 we have these signed up and confirmed:

- KIAaze
- Plouj
- mkzelda
- free-zombie
- Jose
- reptiler
- kojn
- FruitieX
- killo
- Fliker09
- kyre
- Rad Ished
- morfar
- dr_Evil
- efus
- GreEnmArine
- Ironwolf
- Thunder_PT
- filipecerk

Now let me tell you there are some incredible players in here. I faced a few and barely made a frag. I watched a demo between FruitieX and free-zombie and at one point zombie said something like "you're everywhere" and I thought, that's exactly how I felt in my practice matches with mkzelda and kyre. Their "secret"? It's map control combined with various skills like bunny jumping and ramp jumping plus, of course, a good deal of experience. Map control is apparently most crucial. It's knowing the map so well that you can practically predict where the opponent is at any time and by your jumping skills pretty much follow the opponent everywhere he goes - hence creating this illusion to the opponent that you're "everywhere". Smiling

Anyway, I am saying this just to point out what kind of players we're dealing with here. Now the job is to split them into most sensible groups for qualifications. My initial idea was to put strong players in their groups and weaker players in their groups, but as was pointed out (in #rainbow on freenode), it would still be fairly predictable.. Basically no matter how we put it, if we know some players are good we could fairly well predict they'll be moving up.

It seems the best option is complete random luck.. asign a number to each group and use the bot to draw out the four random numbers.

And then, you never know, with a bit of luck and some determination even those we assumed as weaker players could end up being qualified.

Thoughts?

The idea is to draw groups of 4 players by saturday afternoon or evening and then immediately do some qualification matches for Nexuiz.

Cheers

libervisco's picture
Offline
Joined: 2006-05-04
Hm since OpenArena 1vs1 is

Hm since OpenArena 1vs1 is so similar I think I'll add OA 1vs1 here too. Here are the confirmed players:

- KIAaze
- plouj
- Skateinmars
- w1zrd
- free-zombie
- Jose
- reptiler
- DeathZone
- Fliker09
- vostorga
- dr_Evil
- Ironwolf
- efus
- Thunder_PT
- filipecerk
- Phenax

Pretty much same thing applies as for Nexuiz above.

Offline
Joined: 2007-09-10
I like the random draw idea

I like the random draw idea since it helps prevent ill feelings (even further).

What I want to point out also is that though the better players will tend to bubble up in the rankings no matter the initial configuration, it would hurt their averages (used for Ultimate Winner) to purposely put good players together.

Maybe after the normal tournament we can do all-star tournaments (with the focus on getting videos) to let the better players have fun going at each other.

And we can even have recorded games where a really good player single-handedly tries to go against a team of players like myself, ie, experts that will pretend to be noobs for the sake of providing balance to the tournament Smiling .

http://www.nuxified.org/topic/funny_videos_and_a_dance_fest

libervisco's picture
Offline
Joined: 2006-05-04
I'd just like to note that

I'd just like to note that if within 10 hours nobody else confirms, confirmations for Nexuiz and OpenArena 1vs1 are closed and I'll assume the 20th player for Nexuiz 1vs1.

We have only 1 more player to go for Nexuiz 1vs1 to have perfect 5 groups and zero players to go to have perfect 4 groups for OA 1vs1. And that's really ideal. Frankly I'm finding myself tempted to just lock it right now, but here, 10 more hours for confirmations and let's see what happens. Here is my wish.. that 4 more confirm for OA and we have 5 groups for OA too. That'd be nice. Eye

Cheers

libervisco's picture
Offline
Joined: 2006-05-04
Thanks. One thing that got

Thanks. One thing that got on my mind regarding grouping is location. If all in one group are from the same continent then we wouldn't have ping issues in qualifications. On the other hand it's unlikely that it will be perfect one-continent-one-group set up even then. That could only happen if we have a number nicely divided by 4. Like, having 2 groups 100% from Europe and 2 100% from US. We already know this ain't happening since we have at least 3 players from Asia..

Edit: Oh and we don't even have an Nex or OA server in Asia (so when asian players are in a match we'll probably have to even the odds by having the other player play on to him unfavourable lag.. so if one must be laggy, both shall be - and then just do your best (I've seen players saying they do well on lag over 100 or even 200ms).

So it might not be worth it compared to just doing a random draw..

About your other ideas above, sounds nice, we'll see how things proceed. I especially like the second idea. Smiling

Cheers

Offline
Joined: 2007-09-10
>> One thing that got on my

>> One thing that got on my mind regarding grouping is location. If all in one group are from the same continent then we wouldn't have ping issues in qualifications.

I've been completely insensitive to this issue when thinking about groups and qualifiers. My lack of experience with these games shows.... or maybe I am a really advanced player!

Offline
Joined: 2007-11-21
first of all, I'd suggest

first of all, I'd suggest to have like 2 stronger and 2 weak players in a group. So I favor that the weaker players get sorted out in the group stage. It would be stupid to put 4 strong players in one group, as good players would be thrown out of the tournament before even getting into the single elimination...

As I've promised to give you a judgement, here is my list of...

... (very) good players:
- kojn
- morfar
- GreEnmArine
- kyre
- killo
- FruitieX

...players that are not new to the game but will definetely be beaten by the strong players (so they are "weak" or "average"):
- mkzelda
- reptiler (not sure about him, but I think I heard his name already)
- Rad Ished

...absolute newbies that will probably not have much chance against any of the players mentioned in the 2 lists above:
- KIAaze
- Plouj
- free-zombie
- Jose
- Fliker09
- Ironwolf
- Thunder_PT
- filipecerk
- dr_Evil
- efus

I tried to have some sort of "ranking" for the first list. I cannot provide this ranking for the 2 other lists, though.

Of course, no garantuees that these lists are right Smiling

Offline
Joined: 2006-03-28
I'm also not so sure about

I'm also not so sure about myself.
It might be good that I have experience in other shooters (which would be where you might have run into me before) but I have no experience in these shooters. Of course shooters mostly are shooters. You might have a few different weapons and the maps might be different, but it's still pretty much about reacting quickly when you see something move.

The biggest problems I see for myself are the lack of knowledge of the maps and my horrible ping, which really plays against me.

Well, we will see. I will try to be there nonetheless, and I'm sure it'll be fun.

Offline
Joined: 2007-09-10
Sample grouping

For all intents and purposes (lol), I've been categorized properly.

Here are five groups. The list is built straight from your post going in order and trying to spread out the members of a group from themselves. Actually, the twist I spun in was to put the players in snake-like fashion, ie, once I reached group 5, I kept taking names in order but now working my way back towards group 1. This allows the tougher opponents to get the easier opponents.

Group 1:
kojn
KIAaze
Plouj

Group 2:
morfar
Rad Ished
free-zombie
efus

Group 3:
GreEnmArine
reptiler
Jose
dr_Evil

Group 4:
kyre
mkzelda
Fliker09
filipecerk

Group 5:
killo
FruitieX
Ironwolf
Thunder_PT

We could also try a(nother) random selection of some sort (maybe a javascript function can serve as a model usable elsewhere.. or not). We could try alphabetizing. Actually, the best is some sort of indeterministic method or else people that want a particular shuffle will vote for the particular method that gets close ("oh, I want reverse alphabetical but only after removing the first and third letters of the username.. yup").

Though this list may work, we may want to at least flip to see who gets to be in the group of three (this might be a disadvantage or advantage).

Also, if we use a point system of 4 - 2.5 - 1 or 4 - 3 - 2.5 - 2 - 1, I think we should be able to just as easily go with group sizes of 3 or 5 if those are preferred.

Oops. One thing I forgot to check (it will take a while if I do so) is to verify the times available to make sure these groups are practical. Also, we should consider ping.

Has someone started a matrix with players name, location, and availability? Actually, availability should be more flexible than location so we should first consider location (for ping). And we may want to consider that some may not mind ping that much. Also, libervisco (and/or someone else) stated somewhere for one of the games that we might just ignore ping and simply handicap everyone such that their effective lag is about the same -- or something like that Smiling .

libervisco's picture
Offline
Joined: 2006-05-04
Thanks GreEnmArine for the

Thanks GreEnmArine for the chart and thanks Jose for making a first working proposal based on it. I'm afraid there's not much time before we should have at least one solid group. I'm hoping we have some matches tonight.

First of all, since nobody additional confirmed in the last 10 hours I am officially closing the confirmations for 1vs1 OA and Nexuiz tournaments. I will be taking the position to fill in the fourth spot in the currently 3-player group.

As for the location and availability factors.. I'm really not sure if it's worth taking further time on that, because my gut tells me it wont pay off much. Most people have specified their availability on the weekends and there's just about equal chance for anyone to be or not to be available at any given time. It's up to us to cease the times they are available and have those matches then.

The location.. well.. it's gonna be hard putting everyone of the same location into the same group all the time. I guess it wont hurt to do a locations chart though (but that would take a bit of time to gather up valid data), at least just to know up front if we'll have to do a ping-fix match or a normal match. But.. yes, in any case, ping-fix system is there and it will even the odds as much as they can be evened when the lag between two players is too different.

I think your groups are good Jose. One thing I just noted which luckily ended up rather well in it is the way you ranked killo. From demos I recorded FruiteX might actually beat killo effectively, so putting these particular two in this group isn't looking like two stronges against two weakest at all. It's fairly balanced.

Of course, at the end of the day and the beginning of matches all assumptions should be silenced. We can do our best with grouping, but it will be matches who will say who is the best. Eye

libervisco's picture
Offline
Joined: 2006-05-04
Trying to put together

Trying to put together groups for OA as well I'm thinking.. actually knowing who is a stronger player kind of biases us to engineering groups ourselves based on that perception. For OpenArena I have a bit less insight into who are the strongest players and hence have trouble putting the groups together same way as it was put for Nexuiz..

At least for OA random still seems like the best option, if not for Nexuiz..

free-zombie's picture
Offline
Joined: 2006-03-08
I myself am all for

I myself am all for completely random groups. If they are unfair, uh, bad luck. You can't make them completely fair anyway. If you categorize players by how good they are beforehand, you could just as well skip the group phase.

I can't really say whether my placement in that list is correct. I'm no complete newbie at either game, but most of what I play is Nexuiz Minstagib, which requires a skill set different from what it required here...

libervisco's picture
Offline
Joined: 2006-05-04
That's a good point, but

That's a good point, but I'm kind of in between two fairly equally strong points though. On one hand, yes, seeding strong players in each group has a chance of having a predictable result. On the other hand having random groups could put 3 or four great experienced players in one group meaning that they'll disqualify each other before the very tournament starts while the weaker players proceed to the tournament instead and again get predictably thrown out..

I think this is what Jose has been talking about. Weed out the strong players at the beginning and you'll have a less fun tournament to watch further on, because you'll be watching strong players owning weaker ones instead of real one on one matches.

And considering that there is still justification for having qualifiers even if we seed the strong players in that it poses a certain amount of challenge to the strong ones and an early chance for weaker ones to at least make some sort of a stand, at least prove they're better than the other "weaker" players.. It's rather compelling..

On the other hand, you never know what random could come up with.. it could very well strike a balance between these two issues.. but then again it might not. You just don't know..

Also, as it was just pointed out to me, another reason qualifiers even with seeded strong teams aren't pointless is because there aren't that many strong players. We'll have whole 12 of them moving on. This WILL include some weaker ones too in the very tournament too!

Offline
Joined: 2007-11-21
Here is my UPDATED list

Here is my UPDATED list of...

... (very) good players:
- kojn
- morfar
- GreEnmArine
- diomedes
- kyre
- killo
- FruitieX

...players that are not new to the game but will definetely be beaten by the strong players (so they are "weak" or "average"):
- mkzelda
- reptiler (not sure about him, but I think I heard his name already)
- Rad Ished
- killgenerals
- cortez

...absolute newbies that will probably not have much chance against any of the players mentioned in the 2 lists above:
- KIAaze
- Plouj
- free-zombie
- Jose
- Fliker09
- Ironwolf
- Thunder_PT
- filipecerk
- dr_Evil
- efus
- Chinker
- 2xtreme
- Spyro_boy

libervisco's picture
Offline
Joined: 2006-05-04
Here are the final Nexuiz

Here are the final Nexuiz 1vs1 groups and matches: http://www.freedomware-gamefest.com/nexuiz_1vs1

Offline
Joined: 2007-09-10
I don't like to make

I don't like to make assumptions about players. Ideally we would have matches to determine the seeding/placement in the qualifiers. Most tournaments *do* give higher seeds the benefit. Here, the best we might be able to do is to judge volunteered demos and to rely on the grapevine. Basically, we are using a subjective (call it voting) system to rank the players initially. We all know that the winner of a tournament tends to be a high ranked player.. but not always.

If someone objects, I would think that they would be given a second look.

libervisco's picture
Offline
Joined: 2006-05-04
Final OpenArena 1vs1 groups

Final OpenArena 1vs1 groups and matches: http://www.freedomware-gamefest.com/openarena_1vs1

Group 5 is missing a player. If noone fills it soon I will. Eye

But we don't accept any registrations any more, nor confirmations for this game, except that one lucky player. I'm soon to move on to grouping CTF's. I can't guarantee anyone who confirms for OA or Nexuiz to be accepted anymore.

Btw, OpenArena groups were drawn randomly because we don't have as much insight into how good most players are. It had to be random.

This thread is now close. I'll open a scheduling thread for OA 1vs1 shorly.

Offline
Joined: 2007-09-10
How will the final

How will the final tournament work if it starts with a number of players which isn't a power of 2?
Currently for OA, it will be 5*2=10 players and for Nexuiz, 7*2=14 players. (2 finalists per group)

libervisco's picture
Offline
Joined: 2006-05-04
There exist brackets for

There exist brackets for those two:

http://www.darters.com/brackets/10_team_single_bracket.htm
http://www.darters.com/brackets/14_team_single_bracket.htm

And yes this means that some would have to play the first level of matches while others wouldn't. Which ones will those be? The weakest qualified runners (depending on their points) or random (if points end up selecting them all equal).

Offline
Joined: 2007-09-10
Mmh, interesting. I never

Mmh, interesting. Smiling
I never heard about the double-elimination process before (nor triple, swiss and round-robin ^^).
According to wikipedia, the double-elimination tournament is "extensively used in computer gaming tournaments (most famously by the Cyberathlete Professional League)".
http://en.wikipedia.org/wiki/Double-elimination_tournament

Quote:

Which ones will those be? The weakest qualified runners (depending on their points) or random (if points end up selecting them all equal).

It should be the strongest against the weakest in order to avoid elimination of good players early on, while weaker players can progress further. (the goal is to have a realistic player classification after all)
(if they all have the same number of points, yes, random)

Here's my seeding proposition:
http://img142.imageshack.us/img142/875/10teamsinglebracketuh6.gif
http://img57.imageshack.us/img57/7683/14teamsinglebracketxm7.gif

Comment viewing options