If I’m in Elite Smash how good am I?

[zipzo]

When you can lose several hundred thousand GSP for one loss that should speak volumes about how little it means to unlock elite smash or fall out of it.

 

[peekpeek]

When you can lose several hundred thousand GSP for one loss that should speak volumes about how little it means to unlock elite smash or fall out of it.

Losing a lot of GSP in the middle ranges of the rank spectrum doesn't mean anything at all about Elite smash. Yes, when you're in the densely populated part of the MMR spectrum like 0.5M to 2.5M, Rank swings rapidly, because a single average Elo swing is enough to swing past hundreds of thousands of players. But it is way more stable on the extreme ends, and stabilizes significantly well before the Elite cutoff.

All Elo distributions are median-dense, and Ultimate's is especially so because it assigns default values to players for characters they haven't even played yet.

 

[ConOne]

Data seems to suggest it's closer to 3.5%. There was a lot of experimentation and data collection regarding this in the "Elite smash?" thread.

Nice still very happy with that.

Yeah I’m pretty sure

Well I'm very happy with that.

 

[zipzo]

Losing a lot of GSP in the middle ranges of the rank spectrum doesn't mean anything at all about Elite smash. Yes, when you're in the densely populated part of the MMR spectrum like 0.5M to 2.5M, Rank swings rapidly, because a single average Elo swing is enough to swing past hundreds of thousands of players. But it is way more stable on the extreme ends, and stabilizes significantly well before the Elite cutoff.

All Elo distributions are median-dense, and Ultimate's is especially so because it assigns default values to players for characters they haven't even played yet.

Missing the point entirely.

If you really think one single game no matter how badly played validly justifies downranking a person in skill-identification beneath hundreds of thousands of players worse than them prior to that match, you have missed the entire boat.

GSP is a poor method of determining player skill, period.

 

[peekpeek]

Missing the point entirely.

If you really think one single game no matter how badly played validly justifies downranking a person in skill-identification beneath hundreds of thousands of players worse than them prior to that match, you have missed the entire boat.

GSP is a poor method of determining player skill, period.

GSP is Rank. GSP is Rank. GSP is Rank. If your rating drops past a bunch of people, it should go down. What should Nintendo do? Just straight-up lie to players about what their Rank is? Reduce the k-factor to something abysmally small so that it takes hundreds of consecutive wins for people to rise to the top ranks?

I totally agree that people care too much about GSP, but it is Rank. It is assigning a sequential # to players based on their Rating that is used for matchmaking. It is the most basic-ass concept that can be given to a player to show how they stack up against other players.

The "original sin" is that players have GSP/Rating/Rank for characters they haven't even played yet, so just a few games can lead to wild swings for players that are in that same range of rating. I don't know what your proposed solution is. Massively reducing the k-factor to make people feel better, despite hosing matchmaking? I'm curious what your solution is. I think Rank is much more comprehensible than Rating, but I guess people disagree.

 

[zipzo]

GSP is Rank. GSP is Rank. GSP is Rank. If your rating drops past a bunch of people, it should go down. What should Nintendo do? Just straight-up lie to players about what their Rank is? Reduce the k-factor to something abysmally small so that it takes hundreds of consecutive wins for people to rise to the top ranks?

I totally agree that people care too much about GSP, but it is Rank. It is assigning a sequential # to players based on their Rating that is used for matchmaking. It is the most basic-*** concept that can be given to a player to show how they stack up against other players.

The "original sin" is that players have GSP/Rating/Rank for characters they haven't even played yet, so just a few games can lead to wild swings for players that are in that same range of rating. I don't know what your proposed solution is. Massively reducing the k-factor to make people feel better, despite hosing matchmaking? I'm curious what your solution is. I think Rank is much more comprehensible than Rating, but I guess people disagree.

To be quite clear, the OP asked in the title, what does his GSP say about his skill, so I would say that my declaration is rather on topic and relevant, despite your point (which I don't wholly disagree with).

 

[Kookie]

All it really means is you're used to playing terrible connections and camp.

Play battle arenas. That's where good people play.

 

[Dark 3nergy]

Play battle arenas. Learn how the game engine works, find yourself. Then when you're ready head to offline local and level up. Elite mode is not a true representation of skill, just a number grind.

If you can use lunchtime at work and get matches amongst co workers those are great. I am one of those lucky few who can play matches during lunch time.

 

[KingDadada]

I don't really agree with the "just play battle arenas" advice. If you have a group of similarly skilled or better friends to play with then sure. But when I've joined public arenas it's mostly just me facerolling worse players whereas quickplay produces more even/difficult matches.

It's true that skill levels in elite smash vary greatly, but not as greatly as skill levels outside of it.

 

[ConOne]

I don't really agree with the "just play battle arenas" advice. If you have a group of similarly skilled or better friends to play with then sure. But when I've joined public arenas it's mostly just me facerolling worse players whereas quickplay produces more even/difficult matches.

It's true that skill levels in elite smash vary greatly, but not as greatly as skill levels outside of it.

Agree with you I find battle areas annoying- the lag seems more consistent and you can wait for a really long time only for the arena to be shut down before you get a match.

 

[KingDadada]

Agree with you I find battle areas annoying- the lag seems more consistent and you can wait for a really long time only for the arena to be shut down before you get a match.

Yeah that's the other con, I usually join ones with only 2 or 3 slots. Spectating randos just isn't as useful as playing for getting better. You'd be better off watching tournament footage/pros/etc.

 

[Dark 3nergy]

I don't really agree with the "just play battle arenas" advice. If you have a group of similarly skilled or better friends to play with then sure. But when I've joined public arenas it's mostly just me facerolling worse players whereas quickplay produces more even/difficult matches.

It's true that skill levels in elite smash vary greatly, but not as greatly as skill levels outside of it.

To each his / her own.

My experience is the exact opposite of yours.

 

[peekpeek]

To each his / her own.

My experience is the exact opposite of yours.

The MMR system in Ultimate, despite oddities, does a good job of placing you against players for a 50-50 win chance. Only at the very very high and low ends does it get notably funky.

 

[Oneiros5321]

Well if you literally read what I said I tried shielding and you can’t do anything about it. It locks you end until you’re at 80% which is ridiculous. How is spamming a move and getting free damage acceptable?

All it really means is you're used to playing terrible connections and camp.

Play battle arenas. That's where good people play.

Last time I checked Arena, there was only 1 arena on 1v1 no item open to public and it was full =')

 

[Dark 3nergy]

The MMR system in Ultimate, despite oddities, does a good job of placing you against players for a 50-50 win chance. Only at the very very high and low ends does it get notably funky.

Wait, the match making system is regional/country specific right? I forgot about this.

That might explain some of these differences. If I go to arenas, que up a search, I get hit with a TON of East Coast arenas. Some of them even specifically say they're EC. My Match making for my Quick Play just isnt that great compared to other regions.

 

[peekpeek]

Wait, the match making system is regional/country specific right? I forgot about this.

That might explain some of these differences. If I go to arenas, que up a search, I get hit with a TON of East Coast arenas. Some of them even specifically say they're EC. My Match making for my Quick Play just isnt that great compared to other regions.

The matchmaking used in Quickplay/Background is trying to match all of the specific settings, in addition to matching on the underlying MMR and on your connection quality. So that 3.4M GSP (per character) is the giant melting point of what is essentially 10+ different ladders. Not only is there the geographic split, but there's also the "hard" division between Stock 1v1, Time 1v1, and various FFA settings. There may be good people close to your MMR that you will never match with because of game mode and distance.

There's a reason no other game does Global Rank. Sora/Namco are only doing it here because they've taken a total step back from dictating what are the sanctioned game modes.

 

[Dark 3nergy]

The matchmaking used in Quickplay/Background is trying to match all of the specific settings, in addition to matching on the underlying MMR and on your connection quality. So that 3.4M GSP (per character) is the giant melting point of what is essentially 10+ different ladders. Not only is there the geographic split, but there's also the "hard" division between Stock 1v1, Time 1v1, and various FFA settings. There may be good people close to your MMR that you will never match with because of game mode and distance.

There's a reason no other game does Global Rank. Sora/Namco are only doing it here because they've taken a total step back from dictating what are the sanctioned game modes.

Yeah I would not see some of those people due to my settings. Gotchya.

No wonder why they patched it first week the game launched. I haven't touched those settings since day 1.

 

[Veigar]

Grats, but I wouldn't factor it in any meaningful discussion.

Nothing is stopping me from just using my sister's switch to match my switch and then constantly rematch to get liek 10 trillion GSP effortlessly

 

[Father Kelso]

dont know why this posted here

 

[chaos_zero]

It's likely that you're really good. The people I'm playing in Elite are generally pretty great.

Anyone else saying anything else is...mmm...can you taste that...salty.

 

[KingDadada]

Grats, but I wouldn't factor it in any meaningful discussion.

Nothing is stopping me from just using my sister's switch to match my switch and then constantly rematch to get liek 10 trillion GSP effortlessly

I'm pretty sure the matchmaking would stop having a gsp impact or become so minimal you'd have to play a preposterous amount of games to raise your gsp higher. You'd also have to get lucky enough to match against yourself.

Regardless, even if the matchmaking system can be exploited that is fairly irrelevant to the question. That would be like someone saying it's irrelevant I can just pay a pro smasher to play on my account and get all my characters to elite.

 

[VnMidnight]

I recently just hit Elite with 2 characters I main and I find it very tedious even trying to climb up there. I would agree using Anther’s as a way to gain better experience. The issue with quickplay is that 75% are usually Fire Emblem characters, K Rools that can’t use anything besides specials, and projectile spammers that run to each side of the stage and play extremely passive (This gets annoying). Quickplay also picks between two players rule set so you could play a match with characters that use items and gimmick stages in which Anther’s is usually a comp rule set. I would highly recommend Anther’s for those trying to get into the comp of Ultimate.

 

[Oneiros5321]

Play battle arenas. Learn how the game engine works, find yourself. Then when you're ready head to offline local and level up. Elite mode is not a true representation of skill, just a number grind.

This ^^^...a good example happened to me yesterday. A Bayonetta sitting at something like 3 200 000 GSP, all they were doing was side B into combo.
Which got me the first 3 games. After that, I adapted to it, won with just one stock so this person rematched me (probably thought it was luck).
But the next match, I had 2 stocks left at the end of the fight so instead of adapting and trying something different, they left.
Probably to find someone else who would fall for his "technique".

All of that just to say that a lot of people are going to do that just to reach Elite Smash. They will use a cheap technique that requires barely any skill and no understanding of the game whatsoever and rematch only the people who have a hard time countering them.

Honestly the best representation of your skill is your opponent. He's he good? If he beats you, can you adapt? And how fast can you adapt?
Also if he adapts to your game, can you mix up to surprise him? Lots of variable to look into to know if you're a skilled player and GSP might be a part of it, but it's a small part in my opinion.

 

[Crystanium]

On January 2, 2019, I went online and played Quickplay. Within about an hour and a half of online play, I ended up in Elite Smash with Samus. I've been there since with her. However, my Ridley hasn't been consistent. I've been invited and kicked out of Elite Smash with Ridley. Admittedly, he's not my best character and I've stopped using him temporarily, since I feel he's not as good as he should be. Still, I fought someone who went by the name TrinityFox. This player used Palutena and three-stocked my Ridley. From what I recall, this person's GSP was 3.2 million. I cannot say for certain if that means the player should be in Elite Smash. Needless to say, I began to wonder if this person is in Elite Smash and I'm not with Ridley, does Elite Smash matter?

I asked this question at the Samus Discord and I was told that those who are in Elite Smash can still play against those who are not. I don't know the validity of this assertion, but if such is the case, then I find Elite Smash to be meaningless. After all, those in Elite Smash should be fighting others who are in Elite Smash. It should be exclusive. If it's not exclusive, then GSP and Elite Smash are irrelevant. I'm fully aware of how GSP can be irrelevant, since plenty of people can manipulate the system by having stages not set to Final Destination or Battlefield. They can have items on and even time people out with two stocks. Anyway, now that you know why I'm asking this question, I'd like to know the answer and to know if the assertion from the Samus Discord is true.

 

[Madison Turner]

Elite Smash is a bit of a mystery because I don't believe Nintendo has said much about how it really works, but from what I can tell, people in Elite only play others in Elite, and people in regular quickplay play against others in regular quickplay. I believe that you're locked into whatever you queue up as originally, so if you're right on the border between both and queue up originally in regular quickplay and win a bunch, you'll still be playing others in regular quickplay even if your GSP would put you in Elite, and if you're in Elite and lose a ton, you'd still be playing people in Elite despite not having enough points anymore.

Regardless, Elite doesn't technically matter, it's just a way to signify that out of others playing quickplay, you're generally winning more, or doing better against others who are winning more. If there weren't an Elite, but the GSP still worked exactly the same, barely anything would change except that the system wouldn't tell you that you're in the higher echelon of players.

To me, it sounds like that Palutena player was in regular quickplay when they queued up initially, and may have won enough games (sometimes you only need to win one) to have enough points to be in Elite smash, but just kept re-matching instead, which as far as I'm aware will keep the regular quickplay queue going.

 

[Dilan Omer]

Well it means ur good technically but to get really good play LAN and go to tournaments

 

[pollo20x620x6]

I would say the benefits of elite are, you're less likely to run into lag or free for alls.

Most serious players hook up their switch to a lan connection or use seriously good wifi. In my experience, most lag-tastic match were found in regular online, but disappeared shortly after making elite. The occasional laggy match comes in every now and then, but not as much.

And most elite players are more likely to have their settings set to 1 v 1s with tournament rule stocks and level selections.
The moment I made elite, I stopped seeing free for alls as frequently as I did before. This was pre patch.

Elite smash mattering as a badge of honor or accomplishment? No.

 

[Sean²]

There are a ton of ways to abuse your way into Elite, unfortunately. Stupid rulesets, 1up flags. They're a gamble but people do it. You can also only play FFAs and get in just by winning lots of those. Also, regional matchmaking can help or hurt you. I've said it before, but if you're in a region with tons of high level players, you're going to have more issues getting in and staying in. If you're in a region with weak presence, you'll get in and stay in easier.

I've found making arenas turns out to be a way better experience. Titling them like, "(Your Region) Competitive". It's is a way better way to practice, because you can practice tech on people and not have the loss of GSP looming overhead. Normally good people will join - way better than the average quickplay fare. And you're less likely to roll the dice with someone lagging hard on potato wifi. If a lagger joins, you can boot them.

So it means you're pretty good at your preferred modes. But drudging through the dregs of laggy, campy, quickplay to get that one good player with no lag only to have them leave after one or two games, pales in comparison to getting a good arena going. So in all honesty, considering there's an immediately better alternative built into the game, it's fairly worthless. Unless you're really into the trophy factor of it.

 

[leafgreen386]

From what I recall, this person's GSP was 3.2 million. I cannot say for certain if that means the player should be in Elite Smash. Needless to say, I began to wonder if this person is in Elite Smash and I'm not with Ridley, does Elite Smash matter?
.

If you were previously in elite smash with ridley, and just lost to get "kicked out," it will continue to pair you with elite smash opponents until you leave the quickplay menu (i.e. going all the way back out to where you select "quickplay"/"elite smash" before selecting "solo"). Elite smash players only get paired with players in the elite smash queue, unless they sit in a very long lobby. However, a long lobby is usually because they've fallen well below the elite smash threshold due to repeated losses, so if you ever get matched with a player like this, it wouldn't matter anyway.

If you're actually in the normal quickplay queue and you fight someone who should be in elite smash, it may be because they clicked "rematch" on their last opponent, who dropped, and they never exited back to menu. If you exit back to menu, it will forcibly kick you out of the normal queue mode so that you select elite smash.

 

[Mikazuki]

Still not tough enough for the salty spitoon.

 

[Sean²]

I got well beyond the threshold of Elite Smash with a character I'd never touched before today. I'm nearing 3.6 million with Mii Gunner.

How? My Mii Gunner's GSP was at about 1.5 million, just based on the prorated GSP you have for unplayed characters. Queued up at that GSP. Rematched everyone as long as they would stay. Got no unwanted matches or lag, all fairly close to my preferred rules. This was lucky, I guess. I didn't drop a single game. Never backed out a single time while on my win streak, either.

If I had backed out at all, I feel like I would have had a way worse time getting there. Because of the way the system works, it kept matching me with people in the 1.5 million GSP range. I'm assuming there aren't a lot of Mii Gunners online, as a lot of these guys were actually decent, but didn't seem to know what to do against my BS. Maybe those in the higher GSP range would have figured me out quicker.

Nintendo held my hand all the way to Elite Smash. I didn't even have to try and make stupid preferred rules to aid my cause. Another reason to push that Elite really shouldn't be a goal for those trying to truly improve at the game.

 

[Prince Fury- Roaring Lion]

I've found that a lot of Elite players either have very solid gimmicks, or a really good game plan. Most of the matches I have played in Elite have taken considerably more effort to pull off wins. I believe that if you can consistently maintain Elite status with a character then you must be pretty good. That being said, I have also played very difficult matches on the threshold of Elite as well, perhaps even a bit better since the performance was unexpected. While I do not necessarily believe that you have to be good to make it to Elite, I believe you have to be better than the average player to remain there.

 

[leafgreen386]

I got well beyond the threshold of Elite Smash with a character I'd never touched before today. I'm nearing 3.6 million with Mii Gunner.

How? My Mii Gunner's GSP was at about 1.5 million, just based on the prorated GSP you have for unplayed characters. Queued up at that GSP. Rematched everyone as long as they would stay. Got no unwanted matches or lag, all fairly close to my preferred rules. This was lucky, I guess. I didn't drop a single game. Never backed out a single time while on my win streak, either.

If I had backed out at all, I feel like I would have had a way worse time getting there. Because of the way the system works, it kept matching me with people in the 1.5 million GSP range. I'm assuming there aren't a lot of Mii Gunners online, as a lot of these guys were actually decent, but didn't seem to know what to do against my BS. Maybe those in the higher GSP range would have figured me out quicker.

Nintendo held my hand all the way to Elite Smash. I didn't even have to try and make stupid preferred rules to aid my cause. Another reason to push that Elite really shouldn't be a goal for those trying to truly improve at the game.

Sure, it's possible to get into ES by cheesing your way there, but it's a real question of if you can stay in after doing that. Should matchmaking account for your change in rating as you win and lose matches, even when you press "rematch?" Yes, it should, and hopefully it will in the new 2.0.0 update. But no matter how high you cheese your GSP, once you leave the queue and enter ES proper, you'll be facing off against a much higher caliber of player, which if you're not up to snuff, will eventually knock you back down. The system is self-balancing in this regard.

Just because you "can" cheese your way into ES, doesn't mean you can't also get there legitimately. I've said it before and I'll say it again, the GSP system only really works when you don't rematch people, and get a new opponent every game. This will cause you to be matched based on your new GSP each game, resulting in fairer matches across the board. Rematching someone you can beat over and over doesn't prove anything.

 

[pollo20x620x6]

I have 3.4 mil and have never performed decently at a tournament. And this was 2 years ago when I had no life. All I'd do is play smash. I'd do even worse now that I work and socialize, but even then I have 3.4 mil.

 

[Sean²]

Sure, it's possible to get into ES by cheesing your way there, but it's a real question of if you can stay in after doing that. Should matchmaking account for your change in rating as you win and lose matches, even when you press "rematch?" Yes, it should, and hopefully it will in the new 2.0.0 update. But no matter how high you cheese your GSP, once you leave the queue and enter ES proper, you'll be facing off against a much higher caliber of player, which if you're not up to snuff, will eventually knock you back down. The system is self-balancing in this regard.

Just because you "can" cheese your way into ES, doesn't mean you can't also get there legitimately. I've said it before and I'll say it again, the GSP system only really works when you don't rematch people, and get a new opponent every game. This will cause you to be matched based on your new GSP each game, resulting in fairer matches across the board. Rematching someone you can beat over and over doesn't prove anything.

No, I understand this. It was just a little experiment to see how fast I could get there.

I just think people need to focus less on getting into Elite and more on just playing the game.

 

[leafgreen386]

Your numbers for the top of the ladder are pretty off. High level tournament players are a tiny percent of the overall playerbase. Consider that the difference between 3.3M and 3.5M is 200,000 players. The number of players in the entire world that could compete against top level players is less than 1% of that.

Here, let me fix your list. Assuming scores very near 3.5M are the current highest, and that a score just under 3.4M is enough to get into ES, that would make the current ladder something like this:

=== One or more standard deviations below the norm ===
<0.750M: Very casual
=== Within one standard deviation of the norm ===
0.750M-2.750M: Casual
=== One standard deviation above the norm ===
2.750M-3.4M: Intermediate (includes the very bottom of elite smash)
=== Two standard deviations above the norm ===
3.41M-3.45M: High level
=== Two and a half standard deviations above the norm ===
3.45M-3.5M: Very high level (you can probably go to a tournament and have some close games, but won't place)
=== Three or more standard deviations above the norm ===
3.5M+: Top level (you actually have a shot at placing at tournies, but there's still a large skill discrepancy even at this level)

There's not enough data on precisely how many players are on the ladder to further subdivide it as it gets close to 3.5M, since we're talking fractions of a percent at that point. There's a substantial difference between the top 20 players and the top 2000 players, but it's really hard to represent that without accurate measurements. Let's say the highest player right now is at 3.523M. That means a top 2000 player would be 3.521M. It looks almost the same, but the difference in skill is night and day. Also, there's the fact that the system will take a long time to place a top level player where they belong, so it's possible that an actual tournament level player is currently sitting sub-3.5M, just because they haven't played enough games yet.

 

[Xquirtle]

Its just a straight ranking so theres really no way to derive the difference in "skill" between two players. Smash has a nearly infinite skill ceiling, and the best players are many many times better than people ranked even just a few thousand below them. We also have a massive bias injected by the starting GSPs of characters that have never been played. I personally have a small set of like 20 characters that I would ever mess around with online, while the rest of them have zero games played. For me, all of those characters are starting at 2.5 mil and slowly increasing over time. So i'm occupying that 2.5 mil rank across 70% of the cast that I have zero play time on. This results in a gigantic pile of idle fake ranks across the middle of the spectrum. This was done to keep all of the GSPs in line with the total population, but I'd imagine that it is inherently oppressive to the lower end players that actually play a character. You are basically being compared to and outranked by millions of people that haven't even played a single game on your toon.

At the end of the day, the system is trash. I'd say that there is a loose correlation between GSP and skill that gets more extreme towards either end. My personal experience is that degenerate play styles are still highly pervasive from 1.5-3 mil, but they do taper off as you get closer to and over 3 mil. By this, i mean players that have no game plan other than their B button. Not to say that there aren't good "honest" players in the 2 mil range, but the highly degenerate play styles can still survive at those rankings. You just probably won't find a 3.2 mil Ganon that has no approach other than side B, and tries to follow his side Bs with more side Bs

 

[FartyParty]

Maybe you can generalize about players' skill levels below 1 mil and above 3 mil, but from 1-3 mil I say it's kind of a crap shoot because all rulesets are lumped into the same ranking. Somebody around 1.5 mil who plays exclusively 1v1, no items, bf or omega only is very likely better than somebody at 2.5 mil playing FFA matches or 1v1, random stages w/ hazards.

 

[Kobalt]

So... what would it say of me as a player if I manage to get more than 10 characters into Elite Smash? (currently working on #20 atm) 3-stock, 7-minute matches on Battlefield stages, no items, no FS meter.

 

[leafgreen386]

So... what would it say of me as a player if I manage to get more than 10 characters into Elite Smash? (currently working on #20 atm) 3-stock, 7-minute matches on Battlefield stages, no items, no FS meter.

It means you're good enough for elite smash with more than 10 characters. You have enough general smash skill to perform at least decently well on multiple characters, but it doesn't say much tbh.

The difference between players inside and outside of elite smash is like night and day, and that's because the GSP cap in non-elite is the minimum GSP for elite. For a competitive player, getting into elite smash isn't the goal; it's the starting line. Unfortunately, the ranking system doesn't do a good job of telling you exactly *where* you are in elite once you get there. Players only a few thousand GSP apart can be vastly different in skill level as you get near the top, as there's fewer players there to separate you. The only way to find out where you truly stand as a player is to go to tournaments.