The reason for this test is to offer a better understanding of just what "enough anti-tank" means in the modern metagame. I've noticed that a lot of people, especially newer players whose equally new friends don't own a dozen tanks yet, really don't intuitively get just how vital anti-tank firepower is now. It's one thing to say "you need railguns" or "mech lists have lots of tanks", but how do you quantify it? That's the purpose of this article. It can't give you a precise answer, but it can at least give you a rough sanity check on your strategy.


Step 1: Write your list.

Take your standard strategy and mix of units, and write it at 2000 points. Don't worry too much about fine-tuning every little detail, this is a general estimate, not a substitute for careful analysis and playtesting. And also don't worry too much if you don't play at 2000 points, it's just an arbitrary value chosen to simplify this exercise. You'll obviously have to make minor changes in strategy as you scale up or down, but as long as your mix of units is roughly the same percentage-wise, the answer to this test is still a good general guide.

Step 2: Study this list.

This is an example mech IG list. One thing to keep in mind is that this is NOT a well optimized tournament list (it's actually pretty bad), nor does it come even close to maximizing vehicle spam. It's just an example of an "average" mech IG list, the kind of thing you could expect to face in a random non-tournament game against even non-competitive players.



Step 3a: Analysis (simple):

For the simple version, just ask yourself one question: does this scare you? Look at your army, and try to explain just how you plan to deal with this list. If you don't like the answer, then you don't have enough anti-tank firepower. If you face an opponent like this, you are going to lose the game, and probably get tabled.

Step 3b: Analysis (complex):

This is the more complicated, but more informative version:

1) Do the math and calculate your average number of penetrating hits against AV 12 (we'll approximate temporary damage as good enough) in two shooting phases, and remove that many vehicles of your choice.
2) Remove an additional vehicle for each Piranha squadron or spare suicide Devilfish you have as movement blockers (don't count Kroot, as they can be tank shocked away).
3) What is left is what will be reaching your army. Are you happy with the number of melta guns that will be inflicting instant death on your crisis suits? Are you happy with that many pie plates? Etc.
4) Repeat these steps, but you only get a single shooting phase, and it takes two hits to remove a vehicle (smoke save). Are you still happy?
5) Repeat these steps, but anything in a Vendetta (scout move) or with range over 12" gets to shoot you on the first turn. Still happy?
6) Remember how many very generous assumptions I gave you, such as always killing a target with a penetrating hit, or not losing any shooting units before they fire. Are you starting to feel a little worried yet?

If, at any point in this analysis, you are not happy with the results, you don't have enough anti-tank units. Re-write your list and try again.

Step 4: Conclusion:

The current metagame is extremely mech-heavy, and you need a lot of anti-tank firepower to have any chance of winning. Even poorly optimized mech armies can put a ton of vehicles on the table, and it can take a surprising amount of firepower to destroy each of them. Worst of all, given our poor defensive ability, failure to stop all of (or at least the vast majority of) the incoming threats will leave your opponent with enough firepower to massacre your units.

One important result of this test is that you should now see why Broadsides and Piranhas are so important. If not, try repeating it with a list that uses sniper drones/Hammerheads/etc instead of Broadsides, or spends lots of points on Fire Warriors. You'll very quickly find out that you just can't kill the tanks fast enough, and what gets through is way more than you can survive.

So, hopefully this has provided a better understanding of the mech threat, and what it takes to stop it. Just remember that this is a very generous test, and a real competitive opponent will be much more difficult.

This would hurt. Much.
I would say you could do this kind of test with most one-sided armies, like Deathwing or a Tyranid gaunt spam. If your army can take all sides of the spectra (tank spam, model spam and heavy infantery spam) then you have a good army indeed. It's really a thing to aspire for.

This is a harsh test. Fair, but harsh. Just to illustrate how difficult this really is:

The one-shot-against-smoke test (3b-4) is basically impossible without Piranhas. An army of 24 Broadsides generates 18 hits, 12 penetrations, 6 unsaved penetrations, for a total of 7 vehicles remaining to start killing them (along with 3 that are still possibly ready for some action, and probably a bunch of Meltaguns that survived their transports exploding). 15 Sunforge+ and 9 Broadsides bring down 7 or 8 of the 13 vehicles by this test, and of course the Sunforges can only do that by already being in range. 15 Deathrain+ will drop two vehicles (and some change). So, if you buy this test, there is no such thing as shooting them down fast enough to not need blockers.

An army with 3 Piranhas, 2 Fish, and 9 Broadsides penetrates about 3 vehicles, blocks 5 (by this test - blocking is pretty hard to quantify) and still has 5 to deal with, plus maybe 1 ready to return to the game. It's still a fight, but a much better one than those without blockers.

A final observation: 2 Hammerheads and 3 Broadsides will penetrate an amusing 1.4 of these vehicles, leaving 11.6 for the rest of the army to deal with. How long can we really entertain the idea that Hammerheads hold a candle to Broadsides?

First of all, nice article. My list includes 10 suits with plasma/missile pods and fusion blasters, a hammerhead and 2-4 broadsides. I can assure you it does not pass this test, since I merely optimized it to fight marines, so all I have to take out are the rhino's... I faced a very similar army to this one before, and got a draw as follows:

We played starting from the corners (damn, what's that called again? Had a break for too long...). I set up all the slow units on one side and the fast ones on the other, then infiltrated my kroot to prevent him from scout moving his vendetta's too far. First turn I moved all my units to the slow side, thereby effectively denying half his army to join the fight, and by moving my piranha's flat out the second half never really got to my forces. So basicly I moved closer to his first half, which then blocked his own tanks, forcing them to drive around. I guess you could pull off the same in a pitched battle, although it would be rather hard in dawn of war.

Conclusion: you could base your army on it, but once you meet the tyranid horde, you're pretty much screwed, so you should still keep some balancing in mind. Hmm, how would Heatwaves work against vets...

Thanks Peregrine. I think this demonstrates a reality of 5ed.

I think the point Peregrine is making with respect to swarms is that most people do not face them often. Most areas do not have quantifiably large tyranid or horde ork players. In larger gaming circle armies are heavily mech.

Claiming it is a 'vehicle spam' list is not correct either. I can take VERY non optimal lists for SW, BA or IG that include 8-10 vehicles at 1250 points with very little cut and paste (or conversely very optimal lists with the same amount in a competative setting). Upping it to 2000 points makes it even worse.

When I play people who have very little antitank (new players at local shop) it doesn't end well.

Of course if you live in an area where everyone plays gaunt swarm nids, you can ignore this all.

It would be nice, but I'm not sure it's possible to do a test for other archetypes. Besides being the most common archetype, mech MSU armies are very straightforward. There's not a whole lot of complex strategy involved, the goal is simply to maximize shooting efficiency in list construction and deliver a crippling blow in the first two turns. Obviously there's still some endgame strategy, but the game is often decided (or at least heavily in favor of one player) by the outcome of the first exchange of fire. This makes it a good candidate for analysis, since the deciding factors are so often "how many penetrating hits can I generate per turn" and "how much shooting can I take and still return fire effectively".

Other strategies are just a lot harder to quantify with a simple test. The game takes a lot longer, there's a lot more use of movement/terrain/etc, and generally a lot more strategic decisions to be made besides picking which priority targets die first. These factors are a lot harder to include into a simple "count your damage output" type of test, so the Deathwing equivalent to this test would be a lot less informative. If you've got a good idea, feel free to write it, but I don't really know where to start.

Overall I like the idea and it will likely give a few people some food for thought and should prove very helpful to any players wishing to try out some tournaments if they are mostly used to playing in friendly gaming groups.

I do have one problem with it though, that it lacks some realism due to no consideration for terrain.

A list with a lot of transports on a 6' by 4' with 25% terrain is going to be limited in manoeuvrability, this helps us as it will often mean they need to drive vehicles down certain pathways if they wish to reach you quickly and without taking difficult terrain tests (trying difficult terrain on that many vehicles for a turn or two its likely to immobilise a few). This should be considered during deployment as you should try to force a funnelling effect as much as you can. Against a short board edge and some terrain works well.

I have experience with this on both sides playing against guard and with my sisters army (10 rhinos at 2000) and would say finding a clear path that’s unlikely to end up cluttered on a board with a decent amount of terrain at that point level can be very difficult if your opponent knows what they are doing.

Though massed mech is definitely a major concern most Tau players seem to not take seriously enough, I think comparing the armies in this way can make the melta vet list look even scarier than needs be. As they need to come to you to be effective making terrain a big factor, so it’s not just a case of how many vehicles can you down a turn after all if every chimera was driving down one pathway and you immobilised the few at the front, then what are those vets going to do? (forgive the highly exaggerated example, you get the idea)

It's also aimed at people who just play "friendly" games. As I said, this is hardly an optimized tournament list, and it's very reasonable to expect your opponent in a "friendly" game to bring something like this.



Three things:

1) The point of the test is to give a better understanding of just what "enough anti-tank" means, since a lot of people don't seem to intuitively understand the sheer number of tanks 5th edition armies can put on the table. It's not intended as a substitute for proper list construction and playtesting, so adding extra detail like terrain would just add clutter without really adding anything to the main point.

2) Sure, I ignore terrain, but I also ignore other things, like the fact that some of your units will probably be killed before they get a chance to shoot. If you average it all out, it's probably comes out about even, especially since the list is a poorly optimized one. If it turns out that the net result of all the factors I ignored is 1-2 Chimeras less than the test suggests, well, a properly optimized list could very easily have 1-2 more Chimeras than mine.

3) Even if it's a worst-case analysis, you have to consider worst-case scenarios in designing your army. If you've built your list on the assumption that terrain factors will remove or delay some vehicle threats, what are you going to do when the terrain doesn't favor you? Unless your dice are above average that game, you're probably going to get overwhelmed and lose. You might not be able to cover the worst-case scenario (after all, there are hordes/marines/etc to worry about), but you need to be aware that a game like this is a reasonable thing to expect, and that you are taking a very real risk by not having enough anti-tank firepower to deal with it.

I think for other lists, the "test" would be pretty similar - just getting a ballpark of how many of the main thing in the army you can shoot down. Like, can you shoot down 120 Ork Boyz with KFF in 2 turns (or whatever the right numbers are), or can you drop 60 Blood Angels if 40 deep strike on turn 1 and the remainder deep strike on turn 2. These sorts of things.

Though, honestly, I think your first test (the "Does this scare you?" test) is similarly informative, and the more in depth test is just a suggestion on how you might go about realistically answering the first question. So, maybe if people are interested in expanding this, and if it doesn't fall into the category of Army Listing, we could collaborate to put together a gauntlet of these lists to run any given army through.

Like this list, it wouldn't be super-optimized or personalized, but just rudimentary, broad strokes of what is possible type lists. Lists like:

1) High model count Blood Angels DoA.
2) Dark Eldar Raider Spam (test here is similar - how many Raiders can you take down on turn 1)
3) Grey Knights footslogging Psybolt/Psyfleman Spam
4) Space Wolves with max Long Fangs and Razorbacks full of Grey Hunters
5) Grey Knights Terminator-Only (probably a better test than Deathwing, honestly)
6) Tyranid Gaunt Swarm
7) Tyranid Godzilla List
8) Whatever the hot new Necron thing becomes

Though I'm sure this list can be trimmed. It really seems like what would be necessary is the worst case scenario of:

Armor (the IG list above - I'm sure anything that can handle it can handle the 3 Land Raider scenario also)
Horde (either Orks or Gaunts)
MCs (Tyranid Godzilla)
Firepower (Raider/Dark Lance spam)
Assault Rush (Either DoA, or possibly kill two birds with one stone and let the Gaunt army count here too. Rhino Rush is probably more akin to the IG list.)


EDIT:

Also, re: what Peregrine was saying about a worst-case test. It's true, it'll be difficult, if not impossible, to write a list that consistently tables Mech Vets, Gaunt Swarms, Raider Spam, DoA, etc. It's more of a goal to aspire to, than to expect to actually achieve. But even if a list isn't stellar at any one of these, it still gives you a feel for the strengths and weaknesses of the army, and hopefully ensures that you at least have some game against all comers. I'd prefer a list that wins 60% against Mech Vets and 60% vs Gaunt Swarms over a list that wins 90% against Mech Vets and 20% against Gaunt Swarms.

I think that the best testing gauntlet should focus on two things: simplicity, and lists that most people underestimate. The reason I posted this is because people often don't intuitively understand "enough anti-tank" the same way they understand what other lists are capable of. From the way they talk about things like sniper drones being useful anti-tank, it seems like a lot of players have an intuitive understanding of "mech army" that resembles a battleforce with a couple tanks added, not 15+ identical tanks hitting you by the second turn. Any other additions should have a similar goal: demonstrate what "enough" really means as clearly as possible, and leave the fine-tuning for later.

With that goal in mind, the ones I think would be useful:

* Mech MSU (this article)
* Green Tide horde with army-wide cover saves
* Jump pack BA (with FNP)
* GK Paladins (with FNP and optimized for wound allocation)

That covers all of the major archetypes pretty well: mech MSU, assault hordes, fast FNP elite infantry, and defense-optimized death stars. The second two are especially important since a lot of people don't seem to realize how much of a negative effect FNP (especially with a 2+ save and wound allocation) has on the effectiveness of the "volume of fire" approach to removing MEQs and how important plasma/melta weapons are when GK and/or BA are popular.

Raider spam/Tyranid MC spam/etc are just mech MSU armies (MCs are effectively tanks) and can be covered by the IG test, the Paladin list is a worst-case scenario that covers all other terminator lists, and the mixed SW/Necrons/etc lists aren't the kind of one-dimensional spam list that really tests whether your definition of "enough" is appropriate.

From what I can see in general, there's a healthy metagame, though it's roughly rock-paper-scissors. Tank Spam > Hordes > BA DeepStrike > Tank Spam, or something to that effect.

I would suspect we would see lists that have a very good game against one, and very poor games against the others, or lists with good games against most, and poor games against some. (I'd take that second one.)

My ideal would be one that +60% win vs All Flavors of Marine, and Tank Spam, and an unfavorable matchup, though no worse than 20%, vs hordes.

I do like this test, though. Fair? Not exactly. Useful? Most assuredly. Terrain, as mentioned, creates exponential problems for tank spam lists. Just as IRL, ruining the first tank in a convoy can hamstring the entire rest of the line. But it's not easy to make a standard test that takes terrain into account. So it's still a relatively accurate test. Especially the 'simplified' version.

From my own experience, the GK player either saw elsewhere or had the idea to rhino/razorback spam with purifier troops, psycannons everywhere. And Psyfleman dreads in the back. At 1750, 8+ things with an AV, and it wrecked me because I'd never had to deal with mech before then. Now, my list could at least give some licks back and not be completely crushed, and I'd expect similar against this sort of list here.

I guess my thinking on Raider Spam, for example, is that it's not as much about your ability to bring down 15 Raiders as it is about your ability to weather 30 Dark Lances starting on turn 1. They seem to bring up a threat that Mech Vets don't - the one-two punch of immediate Dark Lance craziness, with a follow-up of assault units that nothing in the Tau arsenal can handle (Wyches). I personally have never seen a Raider Spam list without two squads of Wyches.

And about MCs, yeah, they're tank-like, but they tend to feel different to me, personally. Like, against tanks it's easy to get Fusion Blaster happy, and then the MCs come out and you realize why Fusion Blasters are not Rail Guns. Though the flip-side is, Plasma Rifles count for way more against MCs than they do against tanks. - But maybe I'm arguing with myself here. Maybe once you have the Rail Guns to deal with IG and the Plasma Rifles to deal with Blood Angels, the MC match-up just falls into place.

Either way, the Big Three are in the list: Mech Vets, Orks, and Paladins. (i.e., the worst-case scenarios of Armor, Horde, and Elite Infantry).

That's why the IG test has those Vendettas. On turn 1, you're going to be facing 9x lascannons and 10x melta guns, plus whatever Hydras/Manticores/etc have the range to hit you without moving. And don't forget how much of a difference AV 10 open-topped makes, the DE alpha strike is paid for with the fact that their second turn is going to be a lot less impressive. If you average the two, you'll probably find that the total damage done is about the same for both armies.

So in the end, it's just another mech MSU list. It has some differences in its execution of the mech MSU strategy, but it's still close enough to approximate it as mech IG for this kind of general "sanity check" type of analysis.



Pretty much. MCs are different, but not so different that you're likely to have a good understanding of mech and FNP MEQ spam, but fail to understand what MCs require.

Firstly, excellent article for those who need to understand the "mech is king" principle. I prefer the simple version to the complex as I have enough experience to do so. For newer players, who don't understand that it can take about 8 str4ap5 shots to kill a marine, the complex method explains exactly how little chance to do damage you have these days.
With regards to people saying about paladins. Can you please go look at the codex again? 300 point HQ and squads that cost even more as troops is not a difficult list to beat. Replace EVERYTHING with G.K. Terminators / dreadknights and it becomes a scary army. I've found that if you can kill 10 Terminators per turn with ap1/2, it's enough for the 1500-2000pt games. Now, my 1500 army (3broad,2iontank, 3helios, 2fish 18FW 5pf, 6XV25) Just about deals with this, and with mech spam fairly well. I would say I achieve the 60% people are liking. Forget raider spam. Anyone foolish enough to take Dark Eldar vs Tau is asking for it (best time to use fire warriors. Str5 fire > Dark Eldar).
Finally, with hordes, I suppose the best way of being objective is to total up the number of unsaved wounds each unit in your army *should* cause in 2 turns (try T3 no save and T4 3+save) and halve the total. This gives you a rough idea of how many things you can kill per turn, what kills large numbers, and what is best to just finish units off.

Well written and thought out articles are excellent reading. Thank you again to ATT's nominated Tank Master; Peregrine.

The point of this exercise (and similar ones) is to correct the misunderstandings a lot of people have about what armies are capable of and what is "enough" firepower to stop them. It's not to figure out the exact chances of beating them, or what opponent is the most difficult, or whatever. That's what playtesting and list development are for, and that's an entirely separate thing.

As for Paladins, the reason I listed them is the conventional wisdom that, thanks to easy cover saves in 5th edition, volume of fire tends to be the best way to stop MEQs and AP1/2 weapons are often overcosted. This shows up in debates about the usefulness of the fireknife setup: why not use TL missile pods if transports are so important? Why not use a BC instead of the plasma? Unfortunately, once FNP and wound allocation are involved, this plan fails horribly. A lot of people just don't understand things like the fact that a single five-model unit of Paladins can take (on average) one hundred missile pod shots before losing a single model. In other words, if you brought nothing but deathrains with flamers, you could shoot at that single Paladin squad for the entire game and kill nothing. Sure, they die a lot faster if you have appropriate weapons, but that's the whole point of the test: to get people to understand just how wrong they are about what is required to bring them down.

Excellent example of how overwhelming mech tactics can be. Fortunately none of my fellow players can afford that many vehicles! (Here in Australia that list would be about $1000-1500.) It is important to realise just what the sheer volume of targets can do for you. To quote Joseph Stalin: "Quantity has a quality all its own".

Got 9 broadsides? Good...you can kill a max of 9 vehicles. (but don't expect more than a few.) Opponent's got more than 9 vehicles? What do you do then?

Thanks Peregrine for posting this test. Well worth scaring the virtual pants off people. Oh by the way, what does "MSU" mean? (yes, I searched)

MSU I believe is stand for multiple small units.

The one thing that this ignores is terrain and board congestion. My guard army features nine chimeras, two hellhounds, and four assorted russes at 1750...the thing is, once you have more than 9-10 vehicles, maneuver becomes nearly impossible, it's just a horde of big models tripping over one another, nearly eliminating the advantage of being able to move faster than on foot. You don't need to kill more than 3-4 vehicles, provided that you choose the targets carefully for maximum disruption (which is not at all the same as automatically targeting the vehicles that pose the greatest abstract threat), to shut down ( =seriously delay) such a "parking lot list."

The idea that you need to pop every can on the board on turn one is flawed, you really only need to kill enough of them... Of course, the definition of "enough" is a bit slippery.