The paradox of Lotus

#1

Anguibok

As usual, you can't read so I talk. Today Lotus meta :

https://youtu.be/EXpGC-wAsZE

tldw ;

2 most used comp is 2 controller + I Sentinel and 2 initiator + I Sentinel
Best composition is 2 controller + I Sentinel
Sova got amazing result and kinda save the 2 initiator + I Sentinel composition
Best doble ctrl comp : Raze/Vipr/Skye/Kjoy/Omen
Best doble init comp : Jett/Brch/Sova/Kjoy/Omen
Fade can replace Skye, Vyze can replace Kjoy
The paradox is that Sova strong (60% NMWR significant), 2C-1S comp are strong but they are not play together
Another paradox is that agent have horrible winrate with precise comp and amazing winrate with otther (wish sisnt happen in other maps)

What do you think about Tejo on this map ?

PS : Adding script in com

#2

Hades_Loves_Rb

12

Frags

+

–

Raze/Fade/Omen/Viper/Killjoy run it back

#15

Charism

0

Frags

+

–

Literally the goat comp

#19

Hades_Loves_Rb

2

Frags

+

–

Wait actually now that astras buffed and omens nerfed I could see a return of Astra over omen-the best teams on the map (fnc/champs eg) ran it so I think it might be more viable than omen especially on teams like 100T where Cryo can play Astra but not Omen

#30

Anguibok

0

Frags

+

–

With Tejo, I wonder if Vyze isnt better now (Fade-Vyze, of Tejo-Vyze)

#20

Zaba

-2

Frags

+

–

viper is shit now sadly

#24

H3ENnZ

4

Frags

+

–

Shes still good on Lotus

#28

Anguibok

0

Frags

+

–

Not that mush, 54% Non Mirror Winrate for Viper

#35

Zaba

0

Frags

+

–

In 3 maps only 1 team picked her and 0% winrate (vs bbl btw)

#36

Anguibok

0

Frags

+

–

She wasn't nerfed on preseason so no reason to believe her strengh in the last split have changed. You don't judge a pick on 3 maps ^^ (especially since the top team hanst played yet since they arrive in round 2)

Only reason I can justify people pick her less it's because pick her less and less in soloQ and doesn't know to play well enough for proplay, but she was strong after her nerf so no reason to beleive she is weak now (Especially with Tejo that can counter KJ ult that was countering Viper ult)

#37

Zaba

0

Frags

+

–

she will not be good. source: trust me

#3

1243

3

Frags

+

–

nah skye ain't all that, fade clears
skye flashes are so scarce now, no info

#26

Anguibok

0

Frags

+

–

She got 60% Non Mirror Winrate in +SC composition

#4

isumorw

0

Frags

+

–

could replace sova since his dart was near useless on any site of the map and he was used for his shocks are drone.
tejo would replace him

#5

1243

2

Frags

+

–

sova was never meta, so it's hard to replace him because no one plays sova

#6

isumorw

0

Frags

+

–

yea lowk sova aint even meta on the map only like kc last year played it for a bit

#7

josef_ow

0

Frags

+

–

team secret also played sova on lotus the whole year (i dont get it tbh, viper is so fucking strong on this map and not having it makes your defense so weak)

#8

isumorw

0

Frags

+

–

i think they was just more familiar with sova

#9

sideshowsbaldhead

0

Frags

+

–

except the chinese teams that ran neon breach cypher sova omen on every single map

#22

Anguibok

0

Frags

+

–

16% Pickrate and an amazing 60% Non Mirror Winrate (Pvalu 2%, meaning they are only 2% chance this is because of random, because the sample size is big enough)

I have no idea why this is good but he is T.T

#23

cloudberry

0

Frags

+

–

Same reason he was good on Fracture. He's not played on this map, so no one knows where the darts land, so everyone gets scanned by them and he gets way too much value

It doesn't mean he's actually a good agent on Lotus fyi

#29

Anguibok

0

Frags

+

–

16% it's not THAT low.

But yeah you are right good NMWR isnt 100% sure it's the best choice, but that mean in any case, he is massively underated, and more team should play him, he is the only one that allow 2 init comp to be viable..

I still think best comp is 2 controller comp (Maybe with 0-sent since KJ suffer from Tejo)

#10

Laundry

0

Frags

+

–

Can u review the cloudberries lotus comp? :)

#11

sideshowsbaldhead

1

Frags

+

–

we’re not even running the triple initiator anymore bro this is just hating 😭

#14

Laundry

0

Frags

+

–

Mb wasn’t trying to hate😭just wanted to see what she thought of the comp

#12

cloudberry

2

Frags

+

–

We ran it in ONE SCRIM y'all (we couldn't run it again because it was clear some of our guys weren't comfy on their agents)

#13

Anguibok

0

Frags

+

–

What is this comp exactly ?

#16

Laundry

0

Frags

+

–

Breach, Fade, Gekko, Vyse, and omen

#17

1243

0

Frags

+

–

remove vyse add raze, and you win
breach raze fade combos go crazyyy
stun nade off a prowler
raze ult stun off a prowler

#21

cloudberry

0

Frags

+

–

That's just one of the main comp ideas with Gekko instead of Cypher at that point

#18

Galactc

0

Frags

+

–

Depending on if you're good or not neon > raze the map is really big so better rotates and stacks to theoretically win more defense rounds

#27

Anguibok

0

Frags

+

–

With the nerf, I don't think enon is viable anymore :/

#25

cioccolat

0

Frags

+

–

in the stripped club. straight up "jorking it". and by "it", haha, well. let's justr say. My peanits.

#31

n1zf

-1

Frags

+

–

this is why everyone hates bronzilians

#32

cioccolat

-1

Frags

+

–

uncultured swine

#33

Anguibok

0

Frags

+

–

Full script for people that prefer to read :

+++Hi vlr, still based on 2024 split 2, today gonna analysis Lotus meta, and we have got a big sample size with five hundred and forty nine games, so more than 1 thousand compositions, so they are a lot to say.

As usual lets start about archtypes, to give you a small reminder, archetype is the shape of your composition, for exemple for the class nomenclature we consider every composition got 1 duelist, 1 initiator, 1 controller, and the first letter of the roles of last 2 agents define the name of the archetype. For exemple 2 initiator, 2 controller and 1 duelist archetype will be call « IC ». The 2nd nomenclature I use for archetype, is based on the 5 main spell, 1 mobility agent also called diver, 1 flasher, 1 reveal, 1 sentinel, 1 controller. A composition where for exemple flash is missing will be call 0-Flash, or 0-F. Classical is when the 5 roles are covered by 5 agent, with the last letter is the name of the class that cover flash. And Hybrid is a composition where 4 agent cover 5 spell, and the last letter of the archetype is the name of the class of the 5th agent.

The 2 most used class archetype on Lotus have both 44 % Pickrate. IS-composition, and SC-composition, so basically 2 composition with a Sentinel, and the last agent being either a Controller or an Initiator.

Lets start with the archetype that is the most used in valorant in General : IS composition, wish is eavy link with Classical I composition, a composition where the 5 main spell are used and the flasher is played by a Initiator. Thoose archetype have both 44 % Pickrate, 47 % Non Mirror Winrate, but doesnt have enough data to be sure it’s significant because it’s not that far from 50 %

The other most used archetype is SC composition, so doble 2 controller composition, still 44 % Pickrate, but 56 % Non mirror Winrate, wish reach the significant theeshold. Meaning this a really great composition. We can subivide this class archetype into 2 different spell archetype, the most used on is 0-Flash archetype, 37 % Pickrate and 52 % Non Mirror Winrate, it’s not good enough to be significant. The oether archetype is Hybric-C, so a composition were 3 agent cover Diver, Flash, Reveal and Sentinel, and with 2 controllers, it have 9 % Pickrate, but a really amazing 60 % Non Mirror Wirate and is really close to reach the significant thrshold but it lack so data.

Every other archetype are under 5 % Pickrate, so it may be better to stick to the SC composition, but Lotus have an enough big sample size to start making some opinion about archetype under 5 % Pickrate.

The biggest one is 0-Sentinel composition a little bit less than 5 % Pickrate, a Non Mirror Winrate of 43 %, it seems bad but when we look in detail more than half of composition are made of 2 duelist and 2 Initiator composition that is kinda a troll archetype that have an horrible Winrate on everymap because you lack of defensive power, and still in 0-Sentinel archetype, you have IC archetype so composition with 2 initiator and 2 controller got indeed 61 % Non Mirror Winrate, again we lack of data, but if you want to go without a Sentinel, this may be the solution.

I have nothing to say about 0-Reveal composition, 2 % Pickrate isnt enough to make a judgement and for one the Non Mirror Winrate isnt in negative, but I still think it’s a troll archetype everywhere.

If we look at class archetype then we have DS composition, a composition with a little bit more than 3 % pickrate and 32 % Non Mirror Winrate that is significant even if the sample is low. Never play doble duelist on Lotus.

We can end the archetype anlysis on Lotus with doble Sentinel archetype 3 % Pickrate for 45 % Non Mirror Winrate, it’s not bad enough to be significant, but again, why play this archetype when you can play SC composition that seems more reliable ?

#34

Anguibok

0

Frags

+

–

2nd part

+++Ok, let’s look at agents, I Usually I do a tierlist, but for this time I want to break out this rôle per roles because I think it’s more relevant. But before we start, i’ve seen somethind really really weird, usually when I look at the agent, some agent are strong on the map and other agent are weak on the map, whatever the rest of composition is, but on Lotus it seems it’s not the case, some agent seems horible with a precise composition but amazing with another composition

Let’s just start with the most complex rôle to analysis in general, the agent that is not the main diver, not the main reveal, not the sentinel not the main controller : the last agent that people usually call flex, and that usually define the archetype of the composition, and I hate that term, usually on this cathegorie the debate is Breach and Kayo versus Viper, and on this map it’s pretty clear, it’s 42 % Pickrate for breach and 45 % NonMirror Winrate wish is almost significantly bad, versus Viper and her 40 % Pickrate but a 54 % NonMirror Winrate, she seems to be the good choice here, but they are other intresting option, it’s the cathegorie with the most diversity afterall. With the alternative flasher you, Phonix, Reyna and Kayo have almost no plays, next. About 2nd smoker, Astra is the other choice against Viper, 5 % Pickrate, but 55 % Winrate before her up, so expect her to see mush more on the replacing either Viper or even the main smoker. But now we still have 3 agent in this categorie, Iso that have a really low pickrate, Sage that also have a low pickrate but again, a good winrate and finally Deadlock, with 5 % Pickrate and a Winrate under 50 %. Conclusion, the choice of this 5th agent will define the archetype of your composition, and as we see in archetypes data, the best choice seems to be a 2nd controller, so either Viper or an Astra. But you see with other categories, things will be a little bit more complex.

Now lets look at reveal, and you know how I want that every team got at least one reveal, And weirdly enough, this seems the most complex rôle and the most intresting one to study, usually, interraction between agent doesnt matter, wish mean usually you just pick the agent that is the best in their rôle on a specific map, but thats not the case here, you see… On Lotus, reveal create a paradoxe because the agent that are used in the best archetype have a bad winrate, and the agent that have the best winrate is picked only in the worst main archtype, so the best choice of reveal highly depend on the rest of the composition. (Add the winrate of the 2 archetypes)
Fade is the one that have the most pick with 59 % Pickrate, but a Nonmirrorwinrate slightly under 50 % she is the most versatile of the agents, but she seems to have no situation where she is the best choice. The good surprise here, is Sova wish have 16 % pickrate and 60 % NonMirrorWirate that IS significant, they are only 2 % of chance that his amazing winrate is because of luck, but what is weird… Is that Sova is never played with the doble controller archetype that he is supposed to be the best, he is played with the doble initiator archetype, wish mean he allow this archetype to be good, and even Breach to have his chance thanks to him. Then we have Geko with 16 % Pickrate and Skye with 11 % Pickrate and both got a bad winrate so they are supposed to be not a good pick BUT… When they are played in a doble controller composition, Gekko have 55 % non MirrorWinrate and Skye ave 60 ! So conclusion : as reveal you can pick Fade by default but that may not be the best choice. If you want to go doble initiator you have to pick Sova. And if you want to go doble controller, then Skye may stay the best choice.

But now let’s look at the divers, that also see some paradoxes, and the most picked agent is Raze with 74 % Pickrate but an horrible nonmirrorwinrate of 43 %, but when we look closer we can see her bad Winrate is only with doble initiator composition, and she got 43 % NonMirrorWinrate with doble initiator composition but And she got 54 % Non Mirror Winrate with doble smoker composition and is the best duelist with that composition. Neon is in positive with 19 % Pickrate and 54 % Non Mirror Winrate, buuuuut, Neon was hardly nerf so…. RIP… Jett is in negative with 8 % Pickrate, so you probably think she is bad buuut in fact when you look in detail, you can see when she is solo Duelist in doble initiator composition she has 61 % Non Mirror Winrate, and she is 24 % Non mirror Winrate when she is played as a doble Duelist, so Jett isn’t bad, Her Winrate was just been destroyed by doble duelist pick. And you know what is the funniest : when she is pair with Sova and Breach, her winrate reach 65 % Non Mirror Winrate, Better than when Raze is pair with Sova. The same Phenomenon kinda occur for Yoru, but he has less data so better stick to Jett. Conclusion even if she Raze have a significant bad winrate… Well, is be the best option here if you plays her with 2 controllers, but if you want to play Sova with Breach… Then it’s seems it’s better to go to Jett…

Let’s continue with the easiest rôle to analysis on Lotus : Usually the sentinel debate is Cypher versus Killjoy and the debate is really clear, Killjoy is far better, she has more pick, have a good Winrate while Cypher have a bad winrate before his nerf. The 2 of them cover 89 % Pickrate, but some other agent can be intresting option, Chamber doesnt have a lot of play but did receive a lot of success, and Vyze wasn’t release yet, and I do beleive she is strong. Deadlock wasn’t played as a solo Sentinel so I won’t talk about her now, and composition without a Sentinel didn’t do great, except maybe Doble Initiator and Dobol Sentinel archetype. Conclusion pick Killjoy is the default choice, but if you want a flasher, Vyze may be the best option right now, Chamber is also an option if your sentinel love the agent.

Lets look at main smokers, Omen was 97 % Pickrate, and got a good winrate, but he was nerf, so does he stays best pick ? Well Brim and Harbor didn’t saw anypick, Clove have 5 % pickrate and an excellent winrate, she might be an hidden broken pick, but not sure. And Astra… Well, on this map, weirdly enough she barely had anypick in solosmoker, but with the up she might saw more pick in that domain. Conclusion as main smoker pick Omen, Astra or maybe even Clove depending on your preference.