Elo Ratings in Competitive Melee

[DippnDots]

i can't tell who your sarcasm is directed at, but in case of confusion: I'm not in favor of having divisions.

 

[CloneHat]

we should cater to punk-a$$ b!tches especially because they tend to stick with the community and become serious and lasting members with their quality attitudes and desireable traits.

The irony is killer.

 

[ES Lite]

Yeah true i didn't think of people sandbagging to be in a lower division. That could definitely prove to be a problem.

 

[AlphaZealot]

I think with the divisions, you'd actually see people purposely keeping a lower rating so they can participate in the easier division and gain a higher cash pay out than placing top 4/5 in a higher division

This happens in Chess, and when it is discovered that a person sandbagged to not get upped to a higher division, that person can be barred or outright life-time banned by the US Chess Federation. A few bad apples and people cheating is not a reason to not use a system, instead, it is simply a reason to look for and expel cheaters from the community.

 

[Nintendude]

Who will be willing to pay?

I'm also curious about this, Hax. What you propose sounds really similar to what MLG had to offer, but MLG also had a lot of financial support. Any ideas on how to make this feasible?

 

[MattDotZeb]

I approve so hard.
The fact someone is putting a program together to work alongside tio is fantastic. I attempted to do this with some online calculators and it got extremely time consuming and ultimately something I wasn't able to continue.

 

[Zivilyn Bane]

The thing about chess is it is such an established game that they track ratings. Say for example you wanted to enter a U1600 division. If you were rated over 1600 at any time during the last 2 YEARS they would not allow you to enter the U1600. And with chess, only tournaments are rated, and outside of scholastics you see some really high entry fees, average of around $100. So the incentive to sandbag is not nearly worth it and the preventative measures are well implemented. But again, it takes a very established community to be able to use anything like divisions or elo ratings in any way. Sure, we could start tracking players ratings, but to actually use them for divisions or prizes or seeding, it could take years to implement.

The more I think about it the more impossible a USEFUL ratings system seems. Most players would really just not pay attention and it would be so muddy with all the different aliases and general immaturity and unprofessionalism of players and TOs alike that it would probably be pointless.

This discussion has given me a few ideas I'd like to work a little on and share with the community in a different topic.

 

[Hax]

i've been thinking about this a lot and i'm gonna break my post into a few sections

Making this work on a national scale

it's not possible to get this to function when one region, i.e. tristate, has 10x more tournaments than another region, i.e. Alaska. what we would do is set up "regional leaders," designated people that manage the ELO system from within their own region. for example the tristate ELO system would be completely separate from the Alaska ELO system. this is so that everybody is given an equal opportunity to attend just as many tournaments as the people they're competing with to make it to the invitationals.

when it comes time for an invitational, each regional leader submits their ELO rankings to a national leader who verifies the top ranked players and accepts them into the respective invitational tournament. each region would have a different cutoff that corresponds with their tournament attendance (or, a more biased but perhaps more accurate way of measuring this: how many top players attend tournaments in your region). for example, the top 2 Alaskan players would qualify for the invitational while the top 8 tristate players would qualify.

Splitting = double losses

because splitting grand finals kills hype and competition - the opposite of what we're trying to do - whoever splits grand finals gets losses to their respective opponents (as in, they'll lose ELO points)

Getting the money to increase invitational tournies' pots

I don't think getting the money would be that hard. i guarantee that this game's community alone would be willing to donate a couple thousand just to see this happen, or perhaps a percentage of venue fees (i.e. a part of what TO's profit) could be taken out and saved in a bank that would be added to the prize pools at the invitational tournaments. keep in mind, these invitational tournies would have a huge entry fee ($50-$100) so that's a ton of money on its own. I don't know what you guys think about such a large entry fee, but i'm all for it. of course this is all until we get sponsors; keep in mind the chance of us getting sponsors will greatly improve once we give our community a more professional look by using the ELO system.

The LCQ may not be a good idea

One of the reasons we'd be trying to enforce this system is that it would increase involvement in the game/tourney attendance. the last thing i want is someone who couldn't care less about the game, sandbag throughout a season (or not even go to tournaments at all) and then **** the LCQ. I want everybody to work to qualify for these tournaments.

its a dilemma because the LCQ's would also be pivotal in attracting players that did not qualify for the main event. I think the best thing to do would be to make the LCQ EXTREMELY tight, as in only 2-4 players make it out. not 10 players or anything like that.

There doesn't only have to be 1 invitational tournament

The world championships at the end of a season doesn't have to be the only invitational tourney.. we could have huge tournies every few months that would have ranking cutoffs, i.e. 1800+ only, 2000+ only.. etc.

There are no divisions

the higher ranked players should still compete with the lower ranked players at non-invitational tournaments. it doesn't make sense for people that worked hard to be good at the game to win far less money because they aren't competing with people that aren't as good..

 

[Acryte]

This happens in Chess, and when it is discovered that a person sandbagged to not get upped to a higher division, that person can be barred or outright life-time banned by the US Chess Federation. A few bad apples and people cheating is not a reason to not use a system, instead, it is simply a reason to look for and expel cheaters from the community.

jesus, how is this worse than splitting -_- ? Also the fact that different skill level players can get payouts is nice, I mean, I remember when Cosmo was going to a tourney and he asked mendoza (his roommate who plays samus) if he was going and he was like, nah, I'm not gonna make top 5 so why would I go. This is a point where he was decently skilled and probably could have made money if he went, provided he fell in a different ELO category, and he may have actually attended then. Who knows. Sounds to me like it would increase player attendance, but also increase the attendance vs spectators when people show up too. Like people may be more inclined to pay into the tourney if they believe they can get it back or win some in a lower ELO bracket payout.

I wouldn't mind seeing divisions in small tourneys (aka amateur tournaments that aren't based off pools results), however, I woudn't want to see them at a national unless it was like before, where people who get knocked out of the main tourney go to the amateur, and I think pools survival and pools ranking is better for that because people have to work harder and not get guaranteed a legit seed in the amateur tourney just because their ELO is higher even though they sandbagged it to hell and back in pools.

EDIT: hax has alot of great ideas. And yea, so far I've seen the melee community is constantly willing to donate to players to attend tournaments etc. so I wouldn't be surprised if they pooled extra cash around to make an epic pot so that all the players in the tourney want to try extra hard for that moneyz if it was like that year's championship tournament.

 

[Zivilyn Bane]

Hax, I think you're misunderstanding the true purpose of elo ratings. When you say

it doesn't make sense for people that worked hard to be good at the game to win far less money because they aren't competing with people that aren't as good..

You're implying that the pros are winning less money because they money is being distributed to a larger array of players.

However, the thing is that the divisions INCREASE attendance and thus the pot. I can pretty much guarantee you that the USCF system, and a system I'm working on right now for Melee, will increase attendance by allowing amateurs the shot to win money. It's like this: you might run a tournament with top 3 payout and hype it a 100 people show up. 20 bucks each and that's 2 grand. Or you can run a tournament with a pro and amateur division, give 75% of the winninngs to the pro division, and payouts to top 3 or 5 in both. You're going to double your attendance and the pros are going to end up winning just as much or even more than they would have otherwise.

Also, it doesn't make sense for players to have double losses of Elo when they split. Basically an elo system is a scale, and there must always be someone winning and someone losing. In chess, people draw. In that case they gain/lose half as much, or sometimes neither player will gain.

Splitting is just a form of what the chess community calls the "Grandmaster Draw" in which two or more players that are winning a tournament will take draws and split the pots. There is really no rule to detract from this action and the risk is too high for chess players to play for the win usually. It's just something that happens. And honestly, if the first and second players in the tournament want to split their money, it is wrong and immoral for anybody to try to stop them, no matter how much you really wanted to see that grand finals set. It's all about the money and if they don't want to risk it they really don't have to. Plus for one, the top two players in a large enough tournament to merit the want for splitting aren't really going to care if they lose elo.

 

[Hax]

However, the thing is that the divisions INCREASE attendance and thus the pot. I can pretty much guarantee you that the USCF system, and a system I'm working on right now for Melee, will increase attendance by allowing amateurs the shot to win money. It's like this: you might run a tournament with top 3 payout and hype it a 100 people show up. 20 bucks each and that's 2 grand. Or you can run a tournament with a pro and amateur division, give 75% of the winninngs to the pro division, and payouts to top 3 or 5 in both. You're going to double your attendance and the pros are going to end up winning just as much or even more than they would have otherwise.

oh, so every tournament would have 2 different tournaments (amateur/pro) and 1/2 of the amateur pot would go towards the pro pot? I'd have to see this happen to believe that pros wouldn't be losing money from this. i'm not really convinced because over here the typical excuse for not going to tournaments usually never has to do with "I know i can't win so i'm not gonna come;" it usually has to do with the person actually not being able to go lol

also, if you ran an amateur bracket and a pro bracket then how would the system work? amateurs would be ranked in completely different dimensions than pros.. how would there be invitational events if a pro's ranking can't even be compared to an amateur's? how would you even determine whether someone is a pro or an amateur to begin with without being biased?

i think the better thing to do is just have invitational events that only pros can attend (for higher stakes). this should be the "goal" players seek to gain through the ELO system, rather than qualifying for a pro bracket that runs simultaneously to an amateur bracket.

Also, it doesn't make sense for players to have double losses of Elo when they split. Basically an elo system is a scale, and there must always be someone winning and someone losing. In chess, people draw. In that case they gain/lose half as much, or sometimes neither player will gain.

to my knowledge, a split in chess happens when two players get caught in an infinite loop of moving certain pieces back and forth. that's different from splitting before a game even starts. I could be wrong about this but even then, why should we allow something that kills the hype and competition we sought to create? splitting is selfish

edit: I have the solution!! the people in grand finals can decide to split the money prior to their game, but then they should still play a game for ELO points. that way we'd get to see a competitive game while they happily split their money

 

[AlphaZealot]

Sorta random, but Hax I'm quite impressed with your posts on all this.

I think I might try something with my (Brawl) tournament this weekend, where everyone pays $10. I'll take $2 of it and leave that for an AM bracket, and use the other $8 for a Pro bracket. I always do pools followed by bracket, so when Pools finish half the people will get seeded into the pro bracket and have 80% of the pot for their chance of winning, and the other half will have 20% of the pot available.

The payouts would look like:
32 people, $310 (I don't pay entry to my own tournaments)
$248 goes toward the Pro pot, creating the following payouts
1st: $119.04 (48%) - vs $148.80
2nd: $54.56 (22%) - vs $68.20
3rd: $34.72 (14%) - vs $43.40
4th: $19.84 (8%) - vs $24.50
5th: $9.92 (4%) - vs $12.40
5th: $9.92 (4%) - vs $12.40

$62 goes toward the Amateur pot
1st: $29.76 (48%)
2nd: $13.64 (22%)
3rd: $8.68 (14%)
4th: $4.96 (8%)
5th: $2.48 (4%)
5th: $2.48 (4%)

I would probably do some tweaking though...maybe 35/30/20/15 for AM? Leading to
1st: $21.70 (35%)
2nd: $18.60 (30%)
3rd: $12.40 (20%)
4th: $9.30 (15%)

 

[Zivilyn Bane]

Ok to answer. The idea of an amateur/pro bracket with separate payouts is an adaptation that I'm working on designed to increase turnout at tournaments without using elo ratings as the cutoff, or even at all. In chess it works because it's such an established game, but in Melee it just is not going to work. Thus, I've personally given up on the idea of elo ratings in melee at least for short term goals of increasing tournament attendance. Maybe one day melee will be at that maturity level, but not now.

And a Draw in chess is a completely different thing than a "Grandmaster Draw." Lets say for example, a large chess tournament's open division has 550 players. There are say...9 rounds (swiss system, like most chess tournaments). In the final round there are 6 players all tied with 7 wins and one draw each, and of course they're all Grandmasters. IF they played it out lets say one guy wins (thus one guy loses), and two of the games draw. This gives one guy clear first and the entire amount of the first place prize. Sounds great but in the real world these six players will very often agree to a draw BEFORE the game even starts and all six will split the prize money that was intended to be distributed to the top six entrants. It happens all the time and because of the fact that it's the players option to draw then there is no use forcing them to play. They tried making rules like "no draws before the 25th move" to allow for players gaining advantages in the openings and early middle game and thus want to play for the win once the advantage has been established. It didn't work though and people would just play safe openings or play a game and agree to a draw anyways.

So the thing is, lets say Pro player A and B attend a national tournament and pays $800 for travel expenses/food/entry/ect. They're now playing eachother in finals. First prize is 1200. Second prize is 600. They are extremely evened match and it could go either way and provide for a fantastic set and epic proportions and lots of entertainment. But is OUR entertainment really worth THEIR money? I mean, it's money they've already won. And if the two players agree to split and take 900 each, they have both profited. There's really nothing we should be able to do about it.

Now the idea of forcing them to play it out anyways is a good one and I believe it's been tested before. But the fact is, once they agree to split it really just becomes a friendly without the intensity of the stakes at hand. Oh well though? The only thing you can really do about it is try to win next time instead of letting somebody else get there and split. Hax, you're obviously more capable of that than most people, and the fans love a player who doesn't like to split. But if you did split, I for one wouldn't lose any respect for you or any other player that does. Unless you're trying to force your opponent to split with stall tactics or whatever (you know who you are lol).

EDIT: Just saw AZ's post.

Wanted to ask a couple of questions. For one, how do you determine which division the player is put it? I am working on something right now that gives the player the choice, however it is a system that really only works with Nationals and slightly higher entry fees.

And in my opinion, as long as first place in the Ama gets less than third place in the Pro, it's a system that could work pretty good AS LONG as the point is to increase attendance, and as long as it can be proven through various statistics (doesn't have to be immediately) that it is indeed increasing attendance.

 

[AlphaZealot]

The tournament is capped at 32 entrants (which I'll likely reach). 4 pools of 8, top 4 advance to Pro bracket, bottom 4 advance to AM bracket.

 

[MK26]

AZ, once someone comes up with an ELO plugin for TIO, would it be possible to amalgamate all of the ranked tourneys to create an ELO ranking for Brawl?

Also Hax, I don't really think its necessary to change the k-value for larger events, becausue your rating will already fluctuate more since you're playing more games against more people, and what really matters is who you beat and not where you beat them.
Unless im wrong and its normal practice to change k-value based on the size of a chess tourney

 

[AlphaZealot]

It is not normal practice to change the K-value.

 

[Vts]

for the amateur bracket how are you basing on who gets in it?

is it just people who didn't get out of pools or people who got out in first rounds in bracket also?

 

[Rappster]

http://en.wikipedia.org/wiki/Claude_Bloodgood
this is why we can't absolutely trust elo ratings.

 

[PEEF!]

I made a little comic for you guys.

The mind of amateur smash players according to the people posting in this thread:

 

[MK26]

^thats why lower ranked people are allowed to enter the higher ranked levels

also, the last line is 'forget that, i wanna beat scrubs and win some money!'
same face tho

 

[Divinokage]

LOL, that is hilarious. =P

 

[crush]

I love you peef

 

[Hax]

i dont have time for a full response right now but i will say that its wrong to deem the community incapable of utilizing the ELO system on a national scale. we have extremely responsible community leaders scattered all over the country that i'm sure would be willing to manage the ELO system for their given region (it doesn't have to be just 1 person per region, either). i'm not even a TO and i'd be willing to do it for tristate.

peef's comic illustrates one of the problems with having amateur brackets, but the real problem is that amateur brackets cannot coexist with the ELO system. a side tourney for those that don't make it out of LCQ's at invitational tournaments is fine (just so those people have something to do during their stay), but we need to be able to incorporate every tournament match played into the ELO system so that everybody can be ranked together. like I said before, if we constantly have amateur brackets then there are essentially two different dimensions to the rankings, and the system wouldn't work. this is why the side events that I'm envisioning would not be ELO rated, as it's unfair to let people artificially boost their rank while the pros compete in a different tournament

 

[Stev]

I love seeing Reddit's influence in other areas. <3 Peef

 

[Bones0]

To be frank, I don't see the division format utilized in chess being viable for the Melee community. It's simply too easy for someone to sandbag to win an amateur tournament because many people who are very skilled don't necessarily take tournaments very seriously so they wouldn't have a very high ranking. I even had a few great players tell me at Pound they half-wished they had simply got knocked out of Round 1 pools because they were confident they could have won the tournament.

Until the community has a well-established elo system in place (which is a pretty long-term goal from what I can tell; like at least a year from now), we should try to find substitutions for incentives to attract lower skilled players. Just as a random idea, what if tournaments had amateur brackets, but instead of winning money, players who win the amateur brackets could receive some sort of lesson from the top 3 players from the tournament?

It wouldn't have to be anything crazy; you would just need the cooperation of the few best players to improve the community by giving a little back. I think it would be great for players who are starting to reach that slightly above average point in their game play to get direct feedback from players. The top 3 of the main bracket would just sit down for a good 30 minutes and give some thorough lessons/advice to the top 3 players from the amateur bracket. This would at the very least get rid of any incentive to sandbag because players who are closely matched with the top players won't be too interested in getting lessons from them, but newer players would have a pretty cool prize to be shooting for.

In fact, if the amateur bracket is held as a sort of separate tournament after the regular tournament finishes, the top players from the main bracket could even coach the top amateur players as they play out WFs, LFs, and GFs against each other.

This whole idea has some solid benefits to it:

- Amateur brackets will no longer have sandbagging veterans
- Amateurs will still have a goal to shoot for with a prize that is significant to them
- Veterans can potentially increase attendance of amateurs (to increase their prize pot) by giving some solid lessons/coaching afterwards
- Up and coming players will be able to develop their game more with the increased amount of feedback from experienced players

 

[Hax]

during the amateur brackets the amateurs should be able to use X-factor and give their controller to a pro for 20 seconds

 

[Divinokage]

during the amateur brackets the amateurs should be able to use X-factor and give their controller to a pro for 20 seconds

LOL, oh man this thread ftw. I would definitely X-Factor, PP. lol.

Anyways, Bones.. you know not every pro is suited for coaching/teach things to lower end players. I don't see how it could be a good motivator as I'm pretty sure we already do our part in the character boards for example.. and I mean if you really wanted to play someone then you simply ask them. The community is simply not big enough, it's not like the top players are like the NHL where all you can do is watch. If you want advice then you ask someone as you play.. and if that person cannot give you advice then leave it be. I definitely don't want to be forced to do something I don't want to do. However, I don't mind helping either when needed. Though finding motivation for the amateurs.. seems rather tough.. lol. I really can't think of anything right now that could be good.

 

[Zivilyn Bane]

http://en.wikipedia.org/wiki/Claude_Bloodgood
this is why we can't absolutely trust elo ratings.

First off, this is completely irrelevent. Unless the elo syetem is effectively implemented, a really good smash player goes to prison, teaches all the inmates, and rigs it himself, it's not ever going to happen. It is not a good example of how elo cannot be trusted, but is actually a pretty decent example of why elo will not work with today's melee community. "Closed pool" tournaments are what they call it. For example, where I live in Springfield Missouri there is a melee/brawl tournament every month. Our melee team is actually competitive and travels, however the brawl team is just pretty much locals with no interest of improvement or competing. In fact these tournaments are free. If we tried an elo system for our Brawl players, our best guy would be rated as high as anyone else in the community given enough time. And amongst smash, there are plenty of closed pool communities.

Another good example is the fact that we use double elimination brackets. In Chess, they use a swiss system. In swiss, every player plays the exact same number of games. So if there are 30 entrants, there would likely be 5 rounds. They try to match players with the same amount of wins and losses after each round and no two players can play eachother twice. What it does is give everyone an even ground. In double elim, some players only play two games, while some can play a dozen in the same tournament.

i dont have time for a full response right now but i will say that its wrong to deem the community incapable of utilizing the ELO system on a national scale. we have extremely responsible community leaders scattered all over the country that i'm sure would be willing to manage the ELO system for their given region (it doesn't have to be just 1 person per region, either). i'm not even a TO and i'd be willing to do it for tristate.

peef's comic illustrates one of the problems with having amateur brackets, but the real problem is that amateur brackets cannot coexist with the ELO system. a side tourney for those that don't make it out of LCQ's at invitational tournaments is fine (just so those people have something to do during their stay), but we need to be able to incorporate every tournament match played into the ELO system so that everybody can be ranked together. like I said before, if we constantly have amateur brackets then there are essentially two different dimensions to the rankings, and the system wouldn't work. this is why the side events that I'm envisioning would not be ELO rated, as it's unfair to let people artificially boost their rank while the pros compete in a different tournament

Hax, when you say "amateur brackets cannot coexist with the ELO system" you are basically refuting the primary use of using an elo system to begin with. If the goal of your elo system is just to give players numbers corresponding to their skill level, it will fail. The reason is just like Peef is saying, and that's that most people just don't care and won't pay any heed to them anyways. We'd be better off implementing a class system that doesn't use a numerical value, but rather a letter based class system like in Yu Yu Hokusho or Chess. That is, S class, A class, B class, ect. It's easier to administer and keep track of and players would probably support it more.

However if your purpose of elo ratings has a goal to increase tournament attendance, it does have the ability to exist with amateur brackets because that's how you increase the attendance. See my previous two or three posts and it's all explained there.

If you just want to do a circuit culminating in a national/international tournament, just do it and leave elo out of it because it's not going to have an effect and again, most people won't care. They'll likely be very inaccurate also because the fact is smash players don't travel like chess players do. And without that mesh pot of communities intertwining then you're just going to end up with the best players in each region having the highest ratings all close together even though their skill level might be way different.

 

[Nintendude]

First off, this is completely irrelevent. Unless the elo syetem is effectively implemented, a really good smash player goes to prison, teaches all the inmates, and rigs it himself, it's not ever going to happen. It is not a good example of how elo cannot be trusted, but is actually a pretty decent example of why elo will not work with today's melee community. "Closed pool" tournaments are what they call it. For example, where I live in Springfield Missouri there is a melee/brawl tournament every month. Our melee team is actually competitive and travels, however the brawl team is just pretty much locals with no interest of improvement or competing. In fact these tournaments are free. If we tried an elo system for our Brawl players, our best guy would be rated as high as anyone else in the community given enough time. And amongst smash, there are plenty of closed pool communities.

I honestly think that this is a very minor problem. First of all, it would take a very long time for a closed pool's top player to get to the same level as the actual top players. Unless that pool remains in complete isolation for quite some time, I have a feeling that the closed pool's top player would end up somewhere between above average and pro.

Second, this would only really matter for seeding, the issue being that the top player may be overrated. I say "may" because if a region is so isolated, is it necessarily right to assume that their top player is overrated? That's personal bias, which the system aims to avoid. Additionally, is inaccuracy due to shortcomings of the rating system worse than inaccuracy due to seeding with personal opinions? I believe that overall, going strictly by ratings, no matter what circumstances they were attained under, is more accurate than opinions.

 

[Hax]

Hax, when you say "amateur brackets cannot coexist with the ELO system" you are basically refuting the primary use of using an elo system to begin with. If the goal of your elo system is just to give players numbers corresponding to their skill level, it will fail. The reason is just like Peef is saying, and that's that most people just don't care and won't pay any heed to them anyways. We'd be better off implementing a class system that doesn't use a numerical value, but rather a letter based class system like in Yu Yu Hokusho or Chess. That is, S class, A class, B class, ect. It's easier to administer and keep track of and players would probably support it more.

However if your purpose of elo ratings has a goal to increase tournament attendance, it does have the ability to exist with amateur brackets because that's how you increase the attendance. See my previous two or three posts and it's all explained there.

If you just want to do a circuit culminating in a national/international tournament, just do it and leave elo out of it because it's not going to have an effect and again, most people won't care. They'll likely be very inaccurate also because the fact is smash players don't travel like chess players do. And without that mesh pot of communities intertwining then you're just going to end up with the best players in each region having the highest ratings all close together even though their skill level might be way different.

dude have you been reading my posts at all? the goal of the ELO system i'm envisioning certainly isn't to rank players without a purpose; as you said, almost nobody would care. the goal would be qualifying for a fantastic tournament at the end of each season with huge cash prizes and the title of world champion up for grabs. if this isn't enough incentive for someone to attend tournaments and improve his/her ranking, then i don't know what is.

i am extremely experienced with this and know for a fact that it would work. as some of you guys know i also play Pokemon (which uses the ELO system) competitively, and in Pokemon a season consists of tournaments of increasing K-value until it comes time for the world championships at the end of a season (which only top ranked players are invited to). Pokemon USA always strives to make this event a haven for Pokemon players; from having tons of side events to an awards ceremony at the end of the tournament, the world championships has been a success every single time. having been one of the few to qualify and play in this event in 2005, 2008, 2009 and 2010 i can assure you that it is a huge honor and a great experience to play a game you love at such high stakes. an event of this caliber is exactly what i have in mind for smash when i attempt to link it with the ELO system, and i think you're underestimating how good of an idea it is

 

[PEEF!]

On the topic of amateur brackets in Smash:

 

[G. Vice]

Lol. Peef may be mean, but his little comics are quite funny and entertaining :D

 

[Zivilyn Bane]

First off I apologize Hax if it appears I haven't been reading your posts, I have been. I have a lot of mixed emotions about elo ratings and it's hard for me to express all of them in words. What I really mean is this: What you're describing sounds like a Tournament Circuit of events culminating in a "world championships." To me it sounds a lot like the MLG Pro Tour when they ran melee. I'm not sure if you were around to compete in any of those, which is irrelevant but just a side thought. The MLG Pro Tour basically did 5-8 regional events which gave points to players based on standings and then did an invitational at the end of the "season" with invites to the players with the most points. Now, these points are not to be confused with elo ratings. These points are more like something used in Nascar. Now I'm not a fan of Nascar in the slightest, but I do believe that this is a better system for organizing what you're describing than elo ratings. What I assume the OP means when saying "elo ratings in competitive melee" is that he wants to have every single (or the vast majority) melee tournament be elo rated like chess.

As far as Pokemon USA goes, I did a little research and it appears to be an affiliate of Nintendo probably backed by a tremendous amount of money and sponsors. Something like this would definitely work if we had an organization with the financial capabilities organizing it. Basically if MLG or other company decided to run a melee circuit they would undoubtedly be able to establish and maintain an elo system to be used in each of their events. They could even do like the USCF and require memberships to be able to enter the tournaments that are rated.

I just honestly can't see the community getting the job done without some kind of Business type organization in place to do so.

Hax I will say though that I am extremely knowledgeable when it comes to elo ratings and how chess uses them. If you are going to pursue establishing them in melee I can't say that I agree that it will work, but I would be happy to provide any assistance if you need it.

If anybody is looking for more information on various ways to increase attendance in melee tournaments, check out my topic.

 

[Nintendude]

I want to once again reiterate the fact that ratings and rankings are not the same thing. An invitational tournament is using points to rank people and then inviting those who reach a certain rank. Ratings are meant to function as an objective way of comparing the relative strength of many different players to seed or divide into classes, and so people can start saying something like "I'm an 1850-rated Falco" instead of "I think I'm pretty good I've beaten so and so in tourney." What seems to be different about Hax's proposal is the way points are distributed. It's no longer directly related to how you place, but rather who you beat and at what events you beat them at. It's a hybrid system.

I would support a ranked circuit but I am more focused on the ratings at this moment.

 

[Beat!]

I want to once again reiterate the fact that ratings and rankings are not the same thing.

Thissssssssssssssssssssssssssssssssssssssssssssssssss

 

[CloneHat]

An ELO system rates players. A rating is something to brag about and strive to improve.

These ratings make it easier to implement ideas such as ranked circuits, but we lack money, skill, and motivation to make such things a reality. The ELO makes the community more attractive to larger organizations like MLG, but I can't see Melee players putting on yearly skill-based circuits with "huge cash prizes" (from where?) without outside help.

Before thinking about these things, we should focus on having all players rated, and the ELO system commonplace at all major tournaments.

 

[1048576]

I think the idea is that you give enough money to the amateur pot that it becomes worth it for the bad player to attend your tournament, but not so much that it becomes worth it for the good player to exclude himself from the good player pot.

I'm having difficulty thinking through whether there is actually guaranteed to exist a sweetspot here where both scenarios come to fruition, or whether the two will overlap and trying to create the optimal pot will still keep bad players from attending and cause good players to sandbag.

 

[Hax]

Zivilyn, what do you think of a $50-$100 entry fee invitational tournament? that alone would attract all of the top rated players (according to the ELO system) without requiring any outside funding. keep in mind that this is worst-case scenario

and nah, I wasn't around for the big MLG tournies (i was 12 when i went to MLG 06 in new york haha). but a high stakes tournament is definitely something i want to experience, and many people want to relive.

the MLG-style points system would certainly be easier (it wouldn't even require software), but the ELO system is far more accurate. when the ELO system is used, every single match is counted which is good because a player can still be rewarded despite he/she not placing highly. if someone sends a pro to losers and then loses the tournament while the pro goes on to win the tournament, the player should still be rewarded for his victory. that's just one example

 

[Bones0]

I don't see why you would need an ELO system in place to host an invitational tournament. It's pretty obvious who the top players are, so if you want a high stakes tournament just invite all the good players and make the registration fee $50. Problem is, no one except the top few players who are going to win money will bother attending. The players at the "semi-pro" level won't bother wasting all the money when they can play the same players at a regular tournament with a standard entry fee.

 

[Zivilyn Bane]

You might consider running circuit events that set aside a certain amount of money from their entry fees to be used towards the prizes in the culminating event. I know Darkrain started doing $1 entry fee tournaments every weekend at his house before Event 52 and was able to come up with an $800 OOS bonus.

Hax I'll try to make a bigger post and add more later when I have time.