Valorant: Thoughts on Split (re: T1xNSG Invitational)

On the shorter side this time, but had some thoughts on Split after doing some quick reviews on the T1xNSG Invitational.

General thoughts: this was a nice tournament, because with all due respects to the other billion invitationals, this really is the first one where it doesn’t feel like purely a collection of pubstacks — there were still a couple, but mouseSpaz/Gen.G/Prospects have all put down decent early records in open competition, Sentinels were having what I believe is their first outing, T1 put out what portion of their roster is currently confirmed (as Team Brax), and the format (double-elim and no group stages) meant that the teams who were more in the way of filler (sorry, Sonii and Kurt) didn’t get too many games anyway.

The one thing that was a little disappointing for me was Prospects (one of the two open qualifier entrants) going out so early — they ran into eventual champions Gen.G in upper bracket, and took a narrow loss to Team Shroud in lower bracket. I say it’s disappointing, because they’ve tended to be one of the relatively few teams that don’t seem to hate Split enough to always have it firmly as third choice among the current collection of maps. They actually even qualified for this tournament with a 13–3 on Split in the grand final of the second qualifier. (I talked a bunch about how Split was probably going to be an IGL’s dream on my really, really, really early takes article, so kudos to Prospects IGL JoshRTz on their performances there.)

In the end, of the 9 matches on stream, we only got one on Split — the lower-bracket Bo1 final between mouseSpaz and Team Brax (which I’m just going to call T1 from here on out; be thankful I’m not going with SKT T1 B). The game hasn’t quite been out long enough for the community to completely develop all the stereotypical opinions about every single aspect of it yet, but as best I can tell, the general perception of Split is that it’s D/CT-sided, and probably to an unhealthy degree, and that’s why it’s generally bottom of every team’s pick list.

That game…pretty much reinforced everyone’s fears and derision about Split. mSpaz jumped out to a 10–2 lead in the first half on D, only to nearly give it back entirely in the second half — T1 won the eco, mSpaz clawed back the next couple of rounds to make it 12–3, and then T1 ripped off six straight at match point before finally falling 13–9.

23 rounds played, an 18–5 balance in favour of the defending team. Not the best given the perception of Split!

It’s actually a shame. I did go out of my way to watch the Prospects-Team Shroud game, too; the best I could do was a Hiko PoV (nothing on the Prospects side). It was actually a pretty back-and-forth game! Prospects rallied back from an initial 4–0 deficit on offence to make it 6–6 by the half, but gave up the pistol and then could just not win a duel against Relyks’ Phoenix to save their life (the one-time eUnited CS rising star put up 20K 5D in the second half, straight-up manhandling opponents on A site entries); they kept it close in spite of that, but as mentioned earlier, it was a 13–11 win for Team Shroud. Not exactly the cleanest performance on either side, but still, a decent matchup.

Obviously, two games are far too small of a sample to take too much from, but watching both has given me a few thoughts on Split in competitive.

Yes, it’s (most probably) D-sided.

I’m going to keep this to one paragraph. Despite the Prospects result, I would be shocked if, once the APIs are running and the stats start coming down, the map is anything other than D-sided, and to an extent that’s at least a little worrying. Long sight lines everywhere, too many things for attacking teams to trip over, and verticality is inherently a D-sided factor (see Nuke pretty much always being the most CT-sided map in CS).

A Main is a trap.

I didn’t run any numbers on this, but from what I saw, it felt like there were two ways in which teams would get messily dismantled on offence on this map for the most part. One was rushing B Heaven through mid, which, whatever — nobody’s under any illusions about the risk there, and the reward is generally a 4v2 and B site or similar.

The other way was by slowing down in or near A Main. Yes, Valorant (like CS) is in a huge part about map control. Yes, you can go any way you want from A Main technically. But going forward, pushing the enemy off A Main, and doing something like this (mSpaz from round 18 of mSpaz-T1):

Unless you can rush straight through A Main and secure Heaven and site swiftly, this honestly feels like a worse place to put yourself in than even B Main. There are just too many long lines that a defending team can safely take in even a scenario where they 100% know that they’re going to be on the retake. This is perhaps a little bit obvious of a statement ultimately, but I think it’s worth mentioning nonetheless.

B Main needs a Lewis and Clark.

Teams do not tend to do much early on in or near B Main. A quick scan of positioning in the mSpaz-T1 game at (or close to) 0:00:

I actually think this is something of a mistake, on both fronts. I understand why there’s a hesitancy; the Cypher setup heavily disincentivises you from going into that tunnel on D (more on that later), and on O, anyone who’s ever played CS at any level understands the dangers of setting up shop in an area like B Main; no way forward without a lot of utility to block off multiple long-range sightlines, and if your guy watching on the backside goes down, you’re all dead regardless of what you do. From what I managed to see of Prospects, even they don’t want much to do with it.

I still think there’s a bunch of reasons to up the default aggression on it, particularly on offence:

1. The barrier positions mean that in terms of rushing, I think it’s just a straight-up 50–50 despite the extra gap on both sides (and you can use utility to make almost sure it’s going to be a 1v1 without too much effort).
2. Yes, Cypher utility gives them enough warning to start rotating on most pushes. That’s fine. If you can spend a piece of utility to block off sight lines on window or site, I tend to think you can push up, trigger, back off, and have plenty of time to make decisions from there.
3. On the attacking side: it’s a hard site to smoke, but not really any harder than A in my view. Yes, you’re going to have to commit a lot, and yes, it’s a reasonably good site for defenders trying to retake (and you’ll struggle to get more than one pick with how it’s mostly being played right now), but it’s not THAT bad.

I wrote a lot about Split B in the nades article anyway, but I feel like there’s a lot more that can be done here, and there’s probably a need to test the boundaries here in a way that doesn’t seem to happen even in high-level pugs at the moment. It isn’t being mapped out; it’s terra incognita. It’s an issue, and even if you don’t share my perchant for aggression and quick play, it seems like there’s a bit of a collective complacency developing around the whole area to an extent that’s less true on other maps.

Is Cypher the right pick?

As absolutely everyone has already mentioned, there already seems to be a pretty consistent meta emerging with agent picks — almost every team uses four of Breach, Brimstone, Cypher, Viper. Excluding the two filler stacks, they were picked 142 out of a maximum 148 times in this tournament.

Four of the exceptions were T1 on Haven (excluded Brimstone in favour of Phoenix/Sova); the other two were on Split, with Shroud dropping Sage for Phoenix/Omen, and mSpaz dropping Breach for Phoenix/Viper (a pocket pick of the aptly-named hazed).

Both options are interesting. I tend to think Split is Omen’s best map; his ability to bring extra smokes is very welcome, he has relatively good aggressive ult teleport spots for pretty much any site attack/defence, and his short-range teleport works well with the amount of verticality on show here. Viper is a little weaker in absolute terms, but the way her wall works is great on offense in particular, and her ult is great for mid control here.

Anyway, you will notice: Cypher was picked in every single game by said six teams (Team Kurt did not pick him at all, but Team Kurt even got stomped by Team Sonii, and that’s all I’m going to say about that). Watching the games (and particularly from Hiko’s PoV), what I’ve been wondering is this: is Cypher actually the right pick on Split? I know to speak of this in certain circles will have me dethroned or detained, but hear me out here. If you’ve seen Hiko stream ever, you have some idea of the Cypher defence on B split:

The camera position in particular can vary, but the idea here is fairly simple: if they dare poke their heads through the tunnel out of B Main, they can be immediately spotted (and the cage can be triggered to possibly get a kill from window or similar), while Cypher (given that he doesn’t actually have to be anywhere near the camera to see) is then given licence to sit as deep as window itself and assist in mid if kills go down there (which, particularly in pug play, tends to be a far more common venue for early action in a round than anything around B Main itself).

We can debate the merits of this approach; as I mentioned a little further up, I’m personally sceptical that it’s THE best, indisputable thing to do on B Long/Main, but clearly, it’s hardly bad. The problem I have, however, is moreso this. Cypher has poor offensive utility in general, and Split is his weakest map in that regard. Trapwire is an anti-retake/push weapon on offense, and there are just too many tight corners for it to really do the job even in that regard on Split. Cyber Cage is the only smoke without a ceiling, which, as much as it’s great for defending here, is a nightmare given the vertical advantage defenders have on both sites. Even the camera does a relatively poor job of spotting anything actually useful on A from Main (or thereabouts); you have to have a decent amount of push already to make good use of it.

My instinct is that Cypher’s offensive utility on Split specifically is poor enough that it might actually not be worth the tradeoff. Again: I understand your calls of blasphemy. I even first started really getting interested in Cypher as an agent by watching Hiko layer things up on Split B. But I think it’s at least a legitimate question. Besides:

If Hiko is top-fragging as Cypher, something’s gone a little off with the universe.