Absolutely Lukewarm take: YSHELPER infographics should include constellation levels by default
I see a lot of people use YSHELPER stats as evidence for META discussions. And while the merit of even using usage rate as a metric to judge META is a whole seperate discussion, the average constellation level of a character can be very impactful on both their strength AND their teamcomps and should always be included for the full picture.
For example: the majority of people who clear spiral abyss with Neuvillette have his C1, which increases his synergy with Furina and ups his interruption resistance and Furina herself has an average constellation level in spiral abyss of 1.7, with C0 only having a 2.9% higher part of the usage rate ratio than C2.
Usage rate is also often used against TC comparisons when almost all TC's will assume C0 5 stars.
This makes direct comparisons using usage rate inherently flawed without constellation context because units like Xilonen, Citlali, Skirk and all the nod krai units are used most often at C0 (the nod krai units are to be expected because they're still in their shill era) than any other constellation level. You can't argue for example that "Furina is more META than Xilonen because she has a higher usage rate" when they have disparities in constellation level. This isn't me asserting the META value of any of the units I'm using as examples BTW, it's simply me pointing out the flawed logic because these types of analyses completely lack nuance.
Does Wriothesley only having a 16.9% usage rate make him "not that far behind Skirk" because Skirk's usage is 25.5%? No, because in this abyss Wriothesley has an average constellation of 4.1 and most that use him have his C6, while most Skirk users used her at C0. (That's how those solo Wriothesley clears happen BTW because I keep seeing this conversation come up everytime the stats come out).
There's also a conversation to be had about using usage rate as a metric for META in the first place but that's a conversation for another day.
For example: the majority of people who clear spiral abyss with Neuvillette have his C1, which increases his synergy with Furina and ups his interruption resistance and Furina herself has an average constellation level in spiral abyss of 1.7, with C0 only having a 2.9% higher part of the usage rate ratio than C2.
Usage rate is also often used against TC comparisons when almost all TC's will assume C0 5 stars.
This makes direct comparisons using usage rate inherently flawed without constellation context because units like Xilonen, Citlali, Skirk and all the nod krai units are used most often at C0 (the nod krai units are to be expected because they're still in their shill era) than any other constellation level. You can't argue for example that "Furina is more META than Xilonen because she has a higher usage rate" when they have disparities in constellation level. This isn't me asserting the META value of any of the units I'm using as examples BTW, it's simply me pointing out the flawed logic because these types of analyses completely lack nuance.
Does Wriothesley only having a 16.9% usage rate make him "not that far behind Skirk" because Skirk's usage is 25.5%? No, because in this abyss Wriothesley has an average constellation of 4.1 and most that use him have his C6, while most Skirk users used her at C0. (That's how those solo Wriothesley clears happen BTW because I keep seeing this conversation come up everytime the stats come out).
There's also a conversation to be had about using usage rate as a metric for META in the first place but that's a conversation for another day.
Commentaires
When people say it's a popularity contents they don't mean fav units. They mean it's a known issue about it being the most popular meta units, not the best for that mode. If a character is really good but not as popular they will rank very lowly. It's places like keqingmains say never use it to judge Mera :(
Dont see the negative in adding that, but idc about meta so take my thoughts like im Ei in a kitchen
I will be using this all the time from now on.😭
Step 1)Thinking that META discussions are the path to enlightenment on the intricacies of Genshin Impact
Step 2)Realizing that a lot of people engaging in the META discussions are referencing flawed statistics and sometimes chasing falsehoods <------- You are here
Step 3)Zooming out to realize that the entire META discussion debacle is a huge self-referential merry-go-round and doesn't necessarily help players or provide accurate insight
Step 4)Finding out that META discussions are fueled by content creators / websites cashing in on people's fear of doing suboptimal things, and further fueled by fandom of these content creators who take everything they say at face value
Step 5)Meditating on the fact that nobody has a clear or common understanding of what they mean by META and that they are using it as a buzzword to feel like they have some sort of mastery over the game
Step 6) True enlightenment
But then look and behold, it's a team with ineffa who do heavy lifting, neuvilette is c1r1 that lets you to use furina without disturbing his passive (huge amount of damage) and personal damage from sig, which is, quite frankly, doubles his output.
People like to bring out usage data without understanding anything about character's default power output.
For dire he needs more investment though, either on himself or on Furina and Xilonen. You're right about that. Ineffa and Columbina's necessity depends on the content. Durin is also content dependent.
I’ve also seen lots of C0R0 clears against the current crab boss, obviously because it was designed for him.
Yes I agree that Neuv is starting to fall off compared to modern DPS, but he is very much still a good choice in Stygian when the content calls for a hydro DPS (because he is still the second best hydro DPS).
Cactus is literally ineffa's shill - while I don't want to call it ineffa team, since it's not exactly full wheelchair, but calling it "i cleared it with neuvilette", while having an off-fielder that supposed to be an ultimate shill boss... is rather unfair.
Also, while sac jade isn't a 5 star weapon, it's much stronger than usual 4* alternatives.
Not having ineffa was a HUGE downside for me. Granted, i tried fischl instead, but even she came short. I tried to look tons of "guides" and just "clears" on youtube. Specifically ones marked "without ineffa", because otherwise all of them are easy as pie (with neuvilette or without). And every single one had something that i lacked, maybe sig, maybe c1, sometimes both (and people were presenting it in build demonstration like it's not a big deal), or that damn sac jade (yes, i never bought bp).
In the end, i cleared it with yelan-XQ-fischl-sucrose, and yes - it was much easier than whatever neuvilette could do instead.
If Neuv could do it with **Xingqiu** of all characters, that means his DPS is good enough to clear the boss. Remember I am missing an entire Draconic Stack AND Furina. You can swap it for R5 Prototype Amber + another character that Neuv synergizes with better than XQ (Xilonen for example) and he’ll clear just fine.
You should also understand that without ineffa - even going through hat phase is much harder. With ineffa it's basically automatic even with just barbara as hydro.
Next time cactus is there - just try to bruteforce it with "electro-charged", and then tell your opinion on neuvilette.
This guy cleared DIRE with C0R1 Neuv, no Ineffa, 5 cost team. And yes he is a skilled player with cracked builds. But Fearless is more than doable for a less skilled lower invested player. NO INEFFA.
Yes Ineffa is the best character for this boss. No Neuv isn’t lacking the DPS to take it down. We can **both be right** in this conversation you know.
I fully agree that i just didn't have enough skill for that, same with absurd "dire clears with only 4\*" that only top 0.01% can perform.
My point was that ineffa makes that boss a joke, and clear with her can't really be accounted to neuvilette. And that without ineffa - i had much more success with a powerful, yet outdated hydro-combo yelan+XQ just to capitalize on electro-charged and fischl. There is a huge gap between - "it's possible to clear with that character" and "it's easy to clear with that character". Neuvilette is definitely a former.
And yes, just look at this video - it's not just neuvilette who has sig, his fischl is on aqua simulacra. And honestly, even xiphos R5 is a huge part of it too, since in that team, my ER was a huge pain the butt when i tried it. Solo hydro neuvilette that doesn't have sig with energy restoration, and even fischl is solo electro and still wants energy. And even then - he was incredibly close to losing.
I guarantee you, f2p equivalent of that team will not get a half of that damage output, which also means that it will not get fearless done.
https://youtu.be/fWcMq_Jnh0M?si=SVjUx41thuCHBbMr
This is DIRE. With a FREE TO PLAY TEAM. Fearless is totally doable with F2P Neuv. I literally prepared this vid in advanced cause I knew you’d nitpick the other one.
Once again, yes Ineffa is the best character for this boss. No, Neuv is not “bad” at clearing this boss. You clearing it with Yelan and Xingqiu doesn’t make Neuv bad at clearing.
I'm super surprised, and concede on the point that neuvilette is still totally viable end-game clearer. If i kept trying, while adjusting rotations little by little - would probably clear myself (in the same way as i was gungho wasting the entire day to clear tulpa with tighnari).
That said, if somebody ask me whether to pull him for the endgame, i'll still say categorical "No!", since i'm still sure that it's not replicable for the majority of community, whether due to skills or the quality of artifacts (his RV is pretty nuts, but i'm not sure that it is supposed to bring THAT much difference)
Also, I do not imagine most players could clear dire with that team. But the difference between Fearless and Dire is something like 3m+ HP, so that team should have enough DPS to clear fearless even if your artifacts aren’t as good.
I’ve already posted a video of someone else that has done Dire with a 3-cost team (C0 Neuv Proto Amber, C0 Furina Festering Desire, C0 Kazuha Fav Sword, C6 Fischl Fading Twilight). It’s definitely possible to clear Fearless with a f2p Neuv team.
https://youtu.be/fWcMq_Jnh0M?si=SVjUx41thuCHBbMr
And stagger res? With ineffa's electro shield against an electro attack?
But in the end it doesn't really matter now does it? You can get a general idea of what works well from this. Community and character PvP is just nonsense people do for fun.
Also usage IS meta. Calcs, spreadsheets and even speedruns etc are cute and all but what actually works consistently and reliably will always be infinitely more meta defining.
Sure this chart could be more clear about the constellations, but people need to realize that their acronym for meta, “most effective tactics available” isn’t actually what meta means and is a backronym used to simplify the term for people new to metagaming.
Meta is literally “exploiting the tools you have to find what gives you an advantage”. It’s not “this guy has/gives you the highest DPS”. People yelling about how Zhongli isn’t meta and “he’s a DPS loss” when for the longest time he’s helped players facetank and reach 36-stars without even trying, and that is EXACTLY what meta is about; finding the easiest possible path to victory and max rewards. Jean cheese in stygian is meta for the exact same reason.
Yeah, it has always been kinda interesting to see that the genshin players for some reason have come up with their own meaning for meta lol. Somehow spreadsheet dps is more important than actual in game performance, which makes absolutely no sense.
>“most effective tactics available” isn’t actually what meta means and is a backronym used to simplify the term for people new to metagaming.
But that IS what meta is generally used for (well, outside of genshin at least lol). Yes it's a backronym but it is precisely because it's a backronym that it tells you what meta is actually used for as that backronym was literally MADE to explain the word "meta". Sure you can go into the details of what meta actually means but it's irrelevant when the backronym gives you everything you need to know in the context of a video game.
>Meta is literally “exploiting the tools you have to find what gives you an advantage”. It’s not “this guy has/gives you the highest DPS”. People yelling about how Zhongli isn’t meta and “he’s a DPS loss” when for the longest time he’s helped players facetank and reach 36-stars without even trying, and that is EXACTLY what meta is about; finding the easiest possible path to victory and max rewards. Jean cheese in stygian is meta for the exact same reason.
I've argued with so many people that Zhongli is among the most meta characters available for YEARS back in the days before Xilonen etc existed lol. The shield was SO MUCH better for most players than doing a bit more dps that it wasn't even a fair comparison, which is why his usage was incredibly high for a "comfort" character.
Yell it from the rooftops.
Meta is just what people are using. No more, no less. Meta and Optimal are completely different concepts, though often there is a degree of overlap.
What people use IS what's the most "optimal" way of playing generally speaking (or close to it). Usage and meta go hand in hand and what's consistently good or reliable way of dealing with content gets high usage and ends up being meta.
Some INDIVIDUALS may perform better with more min/maxed raw dps teams etc but that doesn't make it meta. Because those teams may not actually work very well with most people, making them sub optimal for the majority. If most people get consistently better results (as in clear content, not simply doing more dps) by for example using a healer vs going healerless with say Nefer teams, then the team with the healer is the actual meta comp rather than the "stronger" healerless team.
And this is exactly what the backronym tells us so it's the easiest way to define what "meta" means in the context of gaming.
No, what people use is what people use. Optimality is a theoretical concept. The theory can influence the praxis, but meta is a measure of the praxis only.
The backronym is worthless and only serves to confuse. It is strictly wrong.
The idea of “technically optimal but not for most people in practice because skill issue” is not an argument against optimality. The optimal strategy is still optimal (optimality is a theoretical concept), but it is not meta (a measure of usage and only usage).
We are agreeing on the premise, but I really want to hammer home that Meta is completely separate from Optimal, any overlap is coincidence, and “most effective tactics available” is a worthless description and should be removed from the lexicon. More accurate would be “most used, sufficiently effective, tactics available”, but just “most used” is the clear simplification of this.
You can theorize what is optimal but what is ACTUALLY optimal or close to it for the average player is shown by usage. Of course it isn't 1:1 comparison as some people will intentionally play sub optimal teams or just don't know what would make their experience better. But in general what tends to perform better than others eventually starts showing up on charts and lists etc.
Back in the day Zhongli for example was objectively more optimal choice in most scenarios than pretty much anything else simply because he made it practically impossible to fail any content he was used in and dps requirements practically didn't exist. And this continued for a long while even to the whole "Zhongli is a dps loss" era. And because of that he was always in the 60-85%+ usage ranges despite being constantly dismissed as somehow not meta character.
>The backronym is worthless and only serves to confuse. It is strictly wrong
The backronym is literally created TO fit what "meta" means in the context it's used in gaming. It just straight up is the easiest way to explain the meaning of meta. You cannot explain "meta" without it working with the backronym.
>The idea of “technically optimal but not for most people in practice because skill issue” is not an argument against optimality
There is no true "optimal" that applies to everyone equally. Someone with major skill issues will find defensive characters significantly more optimal. Someone who is willing to reset multiple times to get that perfect run will find defensive characters mostly useless (well, these days they're also offensive so this doesn't really apply much anymore lol). You cannot claim something is "optimal" if the average player gets WORSE performance from it. That would be objectively a sub optimal team to suggest.
You can say something is optimal FOR dps for example. But that doesn't make that specific team the optimal meta team.
> but I really want to hammer home that Meta is completely separate from Optimal
No, they're essentially the same. Meta generally means "what are the most optimal ways do succeed".
>is a worthless description and should be removed from the lexicon.
No, it's the perfect way to explain meta without having to go in depth with it. It perfectly explains what the word is used for and it does it very effectively.
This is wrong. Optimal is theory. Usage does not inform optimality, it informs meta.
> Back in the day Zhongli for example was objectively more optimal
Zhongli was never optimal, he was comfortable. He was meta. He was not optimal.
> The backronym is literally created TO fit what "meta" means in the context it's used in gaming. It just straight up is the easiest way to explain the meaning of meta.
It was created because people like forcing acronyms. It is wrong. “Meta is what people use” is far easier to explain, and is actually correct to boot.
> There is no true "optimal" that applies to everyone equally. Someone with major skill issues will find defensive characters significantly more optimal.
Yes, there is a true optimal. Application to individuals necessarily has left the discussion of optimality. Optimality is theory. You are no longer discussing theory, you are discussing praxis, and thus you are not discussing optimality. Someone with major skill issues will find defensive characters more *comfortable*. An individual does not find something “more optimal”, that’s a completely nonsense phrase.
> the optimal meta team
Another nonsense phrase.
> Meta generally means "what are the most optimal ways do succeed".
No, Meta means “what people are using”. Nothing more.
===
Since you’re continuing to be argumentative, purposefully missing my point to incorrectly pick nits, despite starting off in agreement with the premise, I’m going to block you.
Even if Yasuo has 30% play rate in Silver/Gold I still wouldn't call him meta meanwhile characters like Rumble are barely played and have been part of the competitive meta for years. Same thing goes for some fighting gamen.
Even in Genshin there's a difference between casual meta (YShelper though 1k sample size is a joke) and Speedrun meta, it's just that ppl don't make the distinction and use meta as a general term
VGC has its own meta that changes every season, with what other players are using affecting who is used as well in order to counterplay what is most used.
And then there is Smogon singles format; while not official it is a widely accepted format. You will have Pokémon played above their tier even though they are terrible in their own tier, just because they specifically counter what is being used in the higher tier. Shedinja in DPP Ubers for example completely destroys Kyogre who can’t touch it, but Shedinja itself is a terrible mon while Kyogre was scarf Water Spouting and one shotting everything that wasn’t immune.
In Genshin’s case, now with Milliastra’s PvP stage, Noelle + Columbina is OP because she doesn’t die. Mavuika is an inconsistent nuke.
people don't question things like user ratings for consumer electronics or IMDB user reviews for movies as much as they question usage rates, but what's the difference really? it works on the same exact logic
That's not a problem with Yshelper but people who just take a screenshot of it and post it on Reddit for upvotes.
- 37,5% C0
- 35% C2
- 18,6% C6.
- 9% the remaining 4 constellations combined.
That simply proves that a lot of people decided to pull for his C2, it's not like they really needed those constellations to clear the SO, you simply can't disable constellations that you already own. I think it's the same for Neuvillette.
Characters like Wriothesley with a 94,6% usage rate at C6 of course proves that people can use him only at C6,
That and, people who are getting higher cons for characters are more likely to care about meta and use a site like this. Others not so much, I'd think.
Durin is popular because he has some of the best cons in the game,
If more people are using Furina then yeah she is more meta. Meta and "which character is better at C0" isn't the same thing.
YShelper is the only reliable one we got at this time.
Goodness I hope the people who work on those don't burnout cause tbey're a godsend to the community.
I think the current statistics is fine as it is: clean, no clutter, visually pleasing, easy to read.
The constellation info is still available to those who want to dig further.
Directly from the post btw
…
RAIDEN? A tier? While Arlecchino isn’t??
> Usage rate is also often used against TC comparisons when almost all TC's will assume C0 5 stars.
This is an inherently flawed argument for a number of reasons. Firstly c0r0 characters are used for the sake of setting a *baseline* for the character, it was never meant to be used to compare what different characters and teams are capable of. Secondly, theorycrafting is subservient to gameplay; the entire purpose of theorycrafting is to help people to understand the game and play better. To make the assertion, then, that the actual gameplay is irrelevant and can’t be used as an argument against calculations is just ridiculous. A third flaw is that, if a constellation has high ownership (as you noted, c2 Furina being only slightly less used than c0 Furina) then evaluating *only* c0 Furina isn’t correct, as it’s dismissive of the way many players actually play the game.
> This makes direct comparisons using usage rate inherently flawed without constellation context because units like Xilonen, Citlali, Skirk and all the nod krai units are used most often at C0 (the nod krai units are to be expected because they're still in their shill era) than any other constellation level.
You can't argue for example that "Furina is more META than Xilonen because she has a higher usage rate" when they have disparities in constellation level. This isn't me asserting the META value of any of the units I'm using as examples BTW, it's simply me pointing out the flawed logic because these types of analyses completely lack nuance.
I think the actual issue is that, much like with most of the community, you look at the usage rates as a way of doomposting rather than just taking them for what they are. Citlali having a lower usage rate than Furina doesn’t mean Citlali is worse than Furina, it simply means that Citlali has a lower usage rate in the current abyss. Period. That’s it. That’s all it means. This is the same reason why I thought it was so stupid how the community doomposted Nahida when her usage rate was really low. Like yeah, in an abyss where you have Gluttonous Yumkasaur King on one half and the Freeze mek on the other Nahida just didn’t really have a place. And guess what? That doesn’t mean she’s bad!
> Does Wriothesley only having a 16.9% usage rate make him "not that far behind Skirk" because Skirk's usage is 25.5%? No, because in this abyss Wriothesley has an average constellation of 4.1 and most that use him have his C6, while most Skirk users used her at C0. (That's how those solo Wriothesley clears happen BTW because I keep seeing this conversation come up everytime the stats come out).
And again, this is you overvaluing these stats. Solo Wrio is a pretty well known meme at this point.
> There's also a conversation to be had about using usage rate as a metric for META in the first place but that's a conversation for another day.
Honestly it’s so weird that you say that, because that’s literally what meta gaming is. It is the “game within the game.” It’s looking at what tactics are most commonly used so that you can copy that idea to garner your own success. So usage rates are, by definition, the most effective way of seeing what characters are meta.
Additionally, like I said to someone else, this post is NOT about the idea that C0 is the only criterion that should be used to analyse META. It is about the fact that, when analysing META, imbalanced constellations and investment between characters provides an incomplete picture . A C2 Furina being used more than a C0 Xilonen tells you nothing about how a C0 Furina compares to a C0 Xilonen or a C2 Furina to a C2 Xilonen, hence why I say Constellation context should be added. If the argument here was to only include C0 units, I would have simply said "YSHELPER should only consider data using C0 units" the way that (I think) prydwen does for ZZZ
I'm not using these stats for doomposting no but your third paragraph doesn't make much sense to me. You say that usage rate simply indicates how often a character was used in that abyss and shouldn't be overestimated past that but then say in the final paragraph that usage rates are by definition the most effective way of seeing which characters are META. This seems contradictory to me but I may also just be misunderstanding the point you're making.
Since this is a PvE game while the usage rates *are* indicative of what the meta is (by definition) they also aren’t that important and shouldn’t be taken as such.
For example you seem to want it clearly communicated that the majority of players that are using Neuvillette have his C1 and that's the only reason he is still viable — but the majority of players using Mavuika have her C2 or higher. So is Mavuika also only viable and usable at C2 or higher otherwise she's worse than the other DPS on the chart that are used at C0?
That's a rhetorical question but you hopefully see what I mean that adding the cons really only shows you which characters are popular with the player base and have been invested in.
TC assuming shit investment level for end-game content users are the one who are putting themselves out of the META discussion. They need a common ground, cool, but their choice ends up being super unrealistic and far far away from the reality.
TC takes are sometimes (not all the time) irrelevant for META discussions because they don't take into consideration the reality of :
\_ level of investment
\_ level of skills and consistency
\_ character availability
\_ device capacity
\_ sometimes even game mechanics (and Hoyo shillings)
Having said that, TCs themselves inherently know how shit their own assumptions are regarding end-game content, they are faced every day with the contradiction between their results and how people really play the game illustrated by YSHleper data. The problem is more about people taking evrey thing they say like the gospel and repeating it everywhere fooling new players or inexperienced players into making bad decisions regarding pulling priorities or investments.
That's why, it's always great to make fun of all thoses parrots every time one of their famous TC takes doesn't translate into the reality of end-game content META. If it can force them to put context on some of their "relative" opinions, it will avoid the spreading of misinformation.
someone with a true understanding of the meta never plays low investment c0r0 teams with shit artifacts because that is either a sign that someone is 1. a new player still barely getting started and shouldn't be rushing endgame or 2. someone who is bad at resource management and more suited to be a casual player
knowing how to properly vertically invest is how to stay ahead in the powercreep rat race, not scrambling to pull and build and then retiring a team within a year's time, and that's something western TC don't teach new players how to do
high usage rate / low average constellation: constellations are underwhelming across the board, maybe vertically investing in teammates is better
low usage rate / low average constellation: weak / clunky from c0 all the way to c6, hard skip
low usage rate / high average constellation: c6 is a game changer and has drastic improvements, either skip or go all the way to c6
Yes, usage rate doesnt equate to meta(epecially considering that it favors low ownership rates), but if a character is consistently used a lot and character is consistently used very little, it doesnt take a genius to realise that the former performs better, after all there is a reason why one is picked waaaaay more than the other.