Replay: XCOM 2 – Long War 2


I’ve not dabbled with mods for many years. That might sound a little strange considering the high volume of PC games that I get through, but it’s partly because of the high volume of PC games that I get through: I usually play games with an eye to reviewing them, you can’t review them fairly if they’re plastered in mods, and over the last few years I’ve had little time to revisit games I’ve already played to see how they change. With the slightly more relaxed (or less obsessive, anyway) attitude I’m taking this year, though, I have the opportunity to do ridiculous, time-expensive things like reinstalling XCOM 2 along with the recently-released Long War mod for it to see what all the fuss is about.

As a mod, Long War 2 doesn’t do much in the way of the classic mod approach of adding new assets or enemies to the game. Well, no, that’s not quite fair, it does add a fair few new weapons and enemies, but that’s a comparatively low-key change and takes a backseat to what Long War 2 is really here for, which is a massive and comprehensive restructure of the mechanics behind the geoscape and battlescape layers of XCOM 2. The base game tried to run with the idea that XCOM was now a guerrilla organisation that had to work in the shadows to undermine the alien regime, but aside from starting most battlescape missions in concealment it didn’t actually do a whole lot to play to that concept, and in places it was painfully obvious that several of the mechanics (Terror missions, the missions where you can counter one of three Dark Events etc.) were simply hurriedly repainted versions of things that already existed in the first XCOM. On its surface Long War 2 is supposed to make the game longer and more intricate, just as the original Long War mod for XCOM 1 did, but what’s impressed me about this sequel is how nearly every single one of its geoscape mechanics embraces the guerrilla concept, weaving it into the game’s structure far more seamlessly than Firaxis managed.

So, here’s what’s still the same on the geoscape:

  • You still fly around unlocking new regions with Intel.
  • There’s still an Avatar Project counter along with associated Dark Events that you can counter by doing certain missions.

And here is what has changed:

  • Literally everything else.


I really wasn’t kidding when I said this mod was comprehensive: nearly everything that the base game did with the geoscape has been either jettisoned completely, or else altered so drastically that it’s barely recognisable in its new form. Take the regions, for example: instead of being simple passive resource generators once unlocked, they’re now individual resistance havens staffed with named rebels who you can passively recruit, or – more likely – break out of prison in one of the jailbreak missions on the battlescape. Each haven can support up to 13 rebels, each of which can be put to work scrounging up supplies, slowly recruiting more rebels or gathering Intel.

Gathering Intel, again, now means more than just receiving an Intel income. Behind the scenes Long War 2 is periodically spawning prospective guerrilla operations around the world – but you’ll only find out about them if you have enough resistance members sniffing around for leads in the region where the operation has secretly spawned. The way this works behind the scenes is that aside from the familiar Intel currency used to unlock more regions, your rebels on Intel duty are also building up a hidden region-specific hoard of intel. Each operation has a minimum threshold of regional intel required to even detect it in the first place, and the chances of detection get better the more regional intel that you have. Checks for detecting an operation happen every six hours so in theory if you just beat that minimum intel threshold you’re going to get a lot of rolls on whether you detect a given operation regardless, but you still want as many rebels on Intel duty as possible because the higher the chance of detection, the earlier you’ll detect the operation in the first place. And this is very, very important because guerrilla operations do not stick around forever; each of them has a limited shelf-life of around 8-12 days after which they disappear from the geoscape, and so if your intel gathering is too anaemic you’ll only detect an operation with a single day left before it expires.

This is bad because of the new Infiltration mechanic. No longer do your strike teams instantly hurl themselves into action the moment they leap off the Skyranger; instead they must now spend multiple days infiltrating the mission area so that they can strike when opposition forces are at their weakest. The strike team will start building up an Infiltration percentage when they arrive, and it takes around five days for a six-man squad to get to 100% Infiltration. Each operation has a baseline level of opposition (ranging from Extremely Light/Light on most operations I’ve done so far to Swarming for HQ assaults) that determines how many enemies you’ll be facing – but you’ll only do this if you infiltrate your team to 100% or more. You’re free to attempt missions with less than 100% Infiltration if you want, but this will increase the number of enemies over and above the baseline; the one mission I tried at 57% Infiltration upped the opposition from Extremely Light to Moderate, which translated to facing 20-odd enemies in huge pods of 6 enemies each compared to the usual 3 x 3 man pods on Extremely Light. I successfully completed the mission, but at the cost of having to leave a sniper behind to stall the bad guys while the rest of the team sprinted for the Skyranger evac zone along with the VIP.


That’s another good thing about having high Infiltration: it means that when you call for an evac the Skyranger will turn up in one or two turns. Low Infiltration means it’ll take three, four, five turns before Firebrand turns up to evacuate the burned and shredded remnants of your team. Low Infiltration also means the enemies will patrol much closer in to your mission objective, as well as causing enemy vision ranges to be larger when your squad is in concealment — and if it’s too low you won’t start the mission in concealment at all. And finally enemy reinforcement drops have been made far more frequent once the shooting starts, making a prolonged fight with Advent forces a losing proposition. You want to spend as much of the mission in concealment as possible, and once you go loud you need to clear out the opposition in just a couple of turns or you’ll eventually be overwhelmed.

Infiltration has two knock-on effects for the rest of the game. First, stealth is now incredibly important. Concealment is far too valuable to be pissed away on getting a free shot in on the first enemy pod you see; instead you end up stalking pods around the map to see if you can potentially get past them without breaking concealment, or if you can catch two together with a rocket or a grenade. Some missions are ideally completed without breaking stealth at all, and — pleasingly — this is now perfectly possible if you drop in a one- or two-man team of the insanely mobile and stealthy Shinobi class. Second, though, and more importantly, while your squad is infiltrating a mission site they’re committed to that location on the geoscape and unavailable for any other activities. Infiltration takes a base of 5-6 days for a 6-person team to infiltrate to 100% – this incidentally means you ideally have to detect operations with at least that long left before they expire, or else make an unwelcome choice between compromising on force strength (fewer soldiers take less time to infiltrate) and facing those 6-8 enemy super-pods, which can quickly lead to things spiralling out of control in the early game. That’s at least five days during which you potentially won’t have access to your finest squad of alien murderers for any other juicy opportunities that might crop up in the meantime. Clearly the usual XCOM approach of having a single A-team with 3-4 reserves being trained up on rotation in case of mishaps isn’t going to work all that well here if the A-team is stuck infiltrating a mission site while the aliens hit one of your resistance havens in a retaliation operation.


This is why Long War 2 features massively expanded soldier rosters along with a named squad management system for convenience. Instead of having 3-4 reserves on rotation I found myself with three entirely separate squads on rotation in the early game, eventually adding a fourth to cover injuries (which take far longer to heal in Long War) and to provide an additional pool of manpower during the set-piece base assaults. I have 28 soldiers on the active duty roster with a further 6 recovering from poison and plasma burns following an unfortunate encounter with a brace of Muton and Viper pods. There are also 3-4 soldiers in training at any one time, either gaining additional abilities in the Advanced Warfare Centre or else being trained up in the Psi Lab or along the new Officer branch in the Guerrilla Tactics School, which confers one-shot abilities that can be used to boost your squad for a turn or delay enemy reinforcements from turning up. There can only be one officer in command on a given mission, but they’ll confer passive Will bonuses to your squad that increase the more missions they go on together, which in the mid-game means Sectoid mindspin attacks have a tendency to just bounce off your fired-up troopers and incentivises keeping the same group soldiers together instead of chopping and changing all the time (although often necessity forces you to do this anyway).

The soldier classes and progression trees have also been significantly changed up, mostly through the expedient of separating out each of the original classes’ ability trees and building a new set that focuses on one specific facet in more depth. Take the base game’s Grenadier, for example, who ran around with a heavy machine gun for suppression and a grenade launcher for demolition; in Long War 2, these weapons and abilities have been split out to the new Grenadier class, who gets the grenade launcher along with a choice between abilities that boost either damaging grenades or support grenades such as flashbangs, and the Gunner, who gets the heavy machine gun and the ability to suppress enemies and reliably destroy cover. (Grenades are basically useless for this now, and only the Gunner’s Demolition ability and the Technical’s rockets will destroy cover 100% of the time.)  The Ranger has been split out into the Shinobi, who gets the sword and the stealth abilities, and the returning Assault class which is built around sprinting through enemy overwatch fire and shotgunning people to death at close range. The Specialist’s drone abilities remain with them, but the rifleman skills go to the new Ranger class.All in all there are now 8 different classes in the game, each of which has a selection of new and familiar skills to choose from, and all of which can be built into a sickeningly powerful murder machine that provides you with an awful lot of options in a fight – if they survive enough missions. Yet another good idea Long War 2 has is to mostly decouple soldier experience from alien kills – you still get some XP for that, but it’s much reduced in comparison to the amount of experience you now get from simply going on a mission in the first place, allowing you to level up new recruits relatively easily without having to faff around arranging killshots for them.


I was deeply impressed by all of these new and altered mechanics and how they supported each other as well as the core fiction that you’re a guerrilla operation going on multiple hit and run raids that require forward planning. The base geoscape design of Long War 2 is one of the most elegant pieces of design I’ve seen in quite some time, and they can lead to some thrilling missions – sending in full 10-man squads to clear the 45-odd baddies populating HQ missions gets downright apocalyptic with the amount of fire and shrapnel that gets sprayed around. Unfortunately I have a bone to pick with the way the campaign has been balanced, and one that might seem a little unreasonable given the name of the mod: it’s too damn long. I accept that having three times as many squads means you now go on three times as many missions; I don’t think that that in and of itself is a bad thing, especially since the tactical decision space created by the new stealth and concealment mechanics is more fulfilling to engage with. I think Long War 2 is actually good enough to support sending your troops on 90 missions in a campaign (as opposed to the 30-ish of the base game).

What it can’t do, though, and what no XCOM game could ever do, is remain engaging through 150+ missions — and this, sadly, is what Long War 2 forces you to do. I am 40(!) hours into my Long War 2 campaign and I’ve just about hit the midpoint with a few regions liberated and the Shadow Chamber being built, and I haven’t even started on the first blacksite mission I got just after the game started. I welcome the meatier mechanics, the more detailed research trees and a more expensive and detailed manufacturing process that’s meant I’ve still been gradually teching up this whole time instead of just plateauing for long periods; there’s always been a sense of forward progress in my campaign. At the same time I look at the game and wonder if it’s really worth spending another 40 hours getting to the end; when I do almost nothing else with my free time for 3 weeks that isn’t playing XCOM1 and I still only make it halfway through a single campaign, I seriously have to question the point of doing all that insanely good work on the design and then asking so much of the player that you ensure that only the tiniest fraction of the people who download Long War 2 will ever see it all the way through.


Oh, and also Long War 2 is, if anything, even worse than the base game at obfuscating its mechanics. It doesn’t explain a single goddamn thing about them, which is really weird considering how well put together they are; I have a suspicion that various regions increasing their Advent troop strength in response to my guerrilla activities is pulling resources away from the Avatar project given how slowly it’s progressing, but I have no way to link the two to prove the cause, and I wouldn’t have had a clue as to how gathering Intel related to unlocking missions if I hadn’t watched a couple of Youtube videos on the subject. There’s a specific chain of missions that lead to each region being liberated from Advent control, but the only way you can tell the difference between the mission that kicks this chain off and the generic guerrilla operations that otherwise spawn all the time is that the liberation chain will have the objective “Find a lead” and the generic operation will have the objective “Find a Lead”. I know Firaxis are the current market leaders at fucking up literally everything to do with communicating to the player how they should play their games, but just because Long War 2 is built on top of one of their products it’s no excuse for Pavonis to do the same thing here. I enjoy the more complex systems of Long War, but they really, really need to be explained to the player far better than they currently are, especially considering the huge time investment the mod is asking them to make – as it is you’re making far-reaching strategic decisions with almost no background information informing your actions, making it very easy for misunderstandings at the start of the game to eventually scupper a campaign 20 or 30 hours in.

I feel a little bad putting the boot into Long War for being too long, especially since it’s very easy to forget that it is “just” a mod rather than the full-blown (and far superior) remake of XCOM 2 that Long War 2 feels like. Pavonis have some crazy talent and it’s good that they’ve made the jump from mod team to independent developer, as I feel that they have some really good ideas about how to make strategy games that could end up creating something special if given the chance to blossom properly in a standalone game. However they suffer from a not-uncommon malady amongst developers of all stripes these days, which is that they lack somebody who’ll tell them when they’ve gone too far and when to reign the project back in scope. I prize focus above all else in the games that I play, but while Long War 2’s mechanics are admirably well-focused (for the most part) there was absolutely no need to bloat the game length the way they have here, other than that it was a sequel to a mod called Long War. I’d be very interested in a mod for this mod that adjusted the pacing so that it took half the time — the Short Long War, if you will. As it is, Long War 2 heroically staves off blundering into the trap of becoming a repetitive slog for a very long time, but that it does eventually fall in regardless is entirely its own fault and ensures it just misses out on being a version of XCOM 2 that I’d always choose over the base game.

  1. And Earth Defence Force. But we’ll get to that in good time.
Tagged , , , , , , ,

4 thoughts on “Replay: XCOM 2 – Long War 2

  1. everblue says:

    The original Long War mod had a “dynamic war” feature that allowed you to adjust campaign length. Maybe they will add that in due course.

    • Hentzau says:

      I hope so! If not, I’m fairly sure somebody will genuinely make a Short Long War mod, or a Dynamic Long War mod. While the idea of modding a mod does bleed over into the fractally recursive, something that’s surprised me about LW is how there’s a whole host of other mods that are specifically designed to work with it. So it’s definitely a possibility.

  2. ilitarist says:

    Entertaining read. Glad to have you back.

    You are right about mods. There are too many good games now to be bothered with mods that are not patches/UI fixes/graphical enhancements. I may be fan of Paradox games and I may be interested in Game of Thrones mod for Crusader Kings 2… But then I try it and see that uncolonized lands are presented as immortal character with a special portrait who hates everyone and is susceptible to a special “Colonize” diplomatic action. Other mods have kitchen sink approach to feature development and have no sense of taste – such as Civilization 4 Fall from Heaven 2 mod which featured Guybrush Threepwood as a hero of one of the factions (not an Easter Egg, but a real Secret of Monkey Island protagonist with ripped VA lines) and Inquisition units are being a copy of Monthy Python sketch. Yep, things like that remind you why you need publishers and marketing and limits on artistic freedom.

    You talk about focus in games and it reminds me of my problem with XCOM 2 which I started but stop playing. Even on Normal it was hard and unforgiving. Some time ago I completed XCOM 1 on Normal/Iron Man (and failed to progress much on Classic/Iron Man several times) so maybe it’s me degrading and becoming old (27 is one step in the grave basically). But it requires some special sort of focus and attention to details. It feels like a game that expects you to have fun on your second playthrough or read guides on your first. Which feels wrong cause it’s also an AAA game aimed at all kinds of people. You describe XCOM 2 LW as it amplifies this problem: it’s like it’s made to be enjoyable after you master it and want long campaign to optimize everything and have all kinds of challenges.

    Meanwhile I’m playing Disgaea which is like a Japanese XCOM. Not that it’s easy – but it splits a campaign into chunks that can be digested, allows backtracking for grinding and is something you can play both when you’re relaxed and when you want a challenge.

    • Hentzau says:

      I’ve already castigated both XCOM and its sequel for that on multiple occasions, as both campaigns are actively set up to trick newbies into making the wrong decisions. XCOM 2 LW is, if anything, even worse as it deliberately obfuscates all of its systems and effectively makes you fight blind. You’re probably right in that it seems like a game that’s targeted at the people who developed and tested it rather than the vast majority of people who are actually playing it. Or in other words, Early Access Syndrome.

      And yeah, on the subject of mods in general I’m a little split. On the one hand I don’t want to be irrationally prejudiced against them and things like Black Mesa and Long War do prove that modders can be at least as talented as the people who made the base games. On the other, I did go through a heavy period of playing mods from about 2000-2005 — basically the heyday of mods in PC gaming — and oh my god I played some absolute stinkers. Some modders are very talented, true. Most, however, are not, or at least are still learning. Arguably the same is true of actual game developers, but overlaying a professional game development process over the top does have a tendency to increase quality and filter out most of the idiocy; it does also tend to make things rather more bland than they would be otherwise, but that’s a price I’m willing to pay in most cases.

Leave a Reply