Monday 22 September 2014

Infinity - How to win games and influence outcomes

There was an interesting article written by Kris Sherriff over at Bell of Lost Souls about not doing stuff "just because you can". The gist of it, as far as Infinity is concerned, was that "opposed rolls are bad" and that spending orders to give your opponent a chance to kill your own models seems like something you should avoid.

Kris stated that "One of the keys to Infinity is maximizing your Normal Rolls whilst forcing your opponent in to as many Opposed Rolls as possible."

I left an initial comment on the article and then spent a lot of time really pondering his statements. Why are Normal rolls good and can we extrapolate that to other areas of Infinity? Can there be such a thing as a 'key' to playing Infinity well? After a lot of thought (and several re-writes of this article) I've come to believe that you can boil down good Infinity play to three factors (or keys) - order efficiency/effectiveness, reducing variance and mitigating failure.

Whilst the old adage of "it's not your list, it's you" is still banded about, the current tournament missions mean that it's simply not true any more - suitable army composition is an important factor. As such, some of the things discussed below are about list building and unit selection as well as in-game tactics.

Note: This article has now been subtly amended to take the changes from N3 into account. Please do let me know if I've missed anything.

I don't have anything in the way of relevant pics, so enjoy some pretty concept art - all rights belong to Corvus Belli

Order Efficiency and Effectiveness

"Efficiency is doing things right; effectiveness is doing the right things." - Peter Drucker

Order efficiency and effectiveness can be summed up with the following two points:
  • Actions you take should be successful as close to 100% of the time as possible. 
  • Actions you take should be furthering the game state in your favour - killing your opponent's models and completing objectives for example.
Efficiency - Never waste an order and try to spend the minimum amount of resources just moving around. If your army is slow then consider playing second to allow your opponent to come closer or use Airborne, Infiltration or Mechanised troops to gain ground. Try to avoid situations where you need to waste the second part of an order or where you need to cancel Impetuous movement.

High quality models - An order spent on a TAG is worth more than one on a lowly foot soldier. The TAG moves faster, shoots more accurately, causes more damage and is more resilient to attacks. This is why 'Rambo' tactics (the act of taking a single powerful model and spending all your orders on it) is a common one; you can spend all your orders on a single efficient character.

High quantity armies - Having more orders to spend each turn means that you can afford to be more inefficient. If you have twice as many orders to spend as your opponent, it doesn't matter if it takes twice as many to complete your tasks. Obviously this is at odds with the point above and quality versus quantity is an interesting balancing game. Typically a couple of high quality models are good to have around and you should always strive to have at least a full combat group (10 models) at 300 points.

Effectiveness - Have a reason for every order you spend and try not to take any unnecessary actions. Whenever you place a model consider why you are putting it one place and not another and even why you are moving that model in particular. Orders are a finite resource so every model you activate has an opportunity cost - the loss of an alternative use for that order. The smaller your order pool, the less frivolous actions you can afford to take.

Deploy with foresight - Good deployment can make or break a game. Make sure that your troops can't easily be picked off, aren't overly clustered and they they cover your deployment zone against airborne troops, Impersonators and similar threats. Try and deploy in such a way that you can minimise the amount of re-positioning needed later in the game such as when trying to reach objectives and be wary of placing troops on or in tall buildings where their manoeuvrability will be restricted later on.

Play the long game - Don't just think about what you'd like to do with this order but what you'd like to do with the rest of the turn and even the rest of the game. Take your time and don't feel pressured to make snap decisions at key moments of the game. It's okay to spend a few minutes thinking about a long string of orders before playing them out.

Make a plan - At the start of each turn, formulate an idea of things that you must accomplish. This could be killing a key model (a sniper, a TAG or an obvious Lieutenant perhaps), capturing a table zone, completing an objective, setting up a strong ARO position and so on. Make sure you do this first. By all means take a crack at any targets of opportunity but don't let yourself get distracted from actually winning the game.

The flip-side is forcing your opponent into a position where they are forced to use their orders inefficiently and on ineffective strategies.
  • Actions your opponent takes should be unsuccessful as close to 100% of the time as possible.
  • Actions your opponent takes should not advance their game state - they should not capture objectives or kill your models. 
Forcing your opponent to re-position their troops, cautious move around AROs, cancel impetuous orders and destroy mines are all examples of ways to make your opponent's turn less efficient.

I have to mention Dodging at this point as it has received a huge buff in N3. Dodging can now almost always be performed (the exception is being attacked from the rear, outside of Zone of Control and without SS2 or a 360 Visor) and will generally suffer only a -3 penalty at worst unless it is done by a Remote, TAG or Motorbike.

Dodging is now a very realistic answer to Surprise Shots, Guided/Speculative fire, Direct Template Weapons and other situations that would otherwise be very unfavourable. Don't be afraid to spend AROs on Dodging instead of taking low odds on shooting back if it wastes your opponent's orders and preserves your models. Remember though that you almost always have a ~5% chance of doing damage with an attack via a critical hit!

Of course none of this is about hard and fast rules as much as it is about having the right mind-set. There are many times where there is no clearly correct play. For instance do you Move-Move to deny your opponent cover before opening fire? Do you shoot on 10s or dodge on 13s in ARO? Should you use a normal or a template weapon in ARO? These are all questions that are difficult to answer and will depend entirely on the particular circumstance that you are in.

So, how do we make our orders more efficient and how do we get close to that magical 100%? Read on...

Reducing Variance

The fact or quality of being different, divergent, or inconsistent.

Variance is roughly analogous to randomness and is something that you want to avoid as much as possible in Infinity. Whilst this is of course impossible to do entirely due to the nature of it being a dice based game, you should still strive to make the odds favourable. One thing to keep in mind is that, unlike most games, if your models are hard to hit then they are also more likely to hit your opponent's models.

Naturally you will also want to increase the variance for your opponent in whatever ways you can and it is important to force engagements on your terms whenever possible. Of course the active player will generally have the advantage but you should strive to make every action as risky as possible for your opponent.

I highly recommend playing around with the Infinity Dice Calculator made by Jonathan Polley where perhaps the most enlightening thing you may find is how often nothing happens. Two Fusiliers (BS 12, Combi Rifles, ARM 1) shooting at each other in cover will achieve nothing almost 40% of the time. Just tweak the numbers to see how different stats, ranges and weapons can affect the percentages.

Use cover effectively - Try and stay in cover yourself whilst denying cover to your opponent. The bonus it gives is huge so you should assume that any model you leave out of cover is going to die next turn. In the example of the two Fusiliers the attacker goes from 48% chance of causing a wound to 71% just by denying cover to the defender.

Roll more dice - Higher Burst weapons are more likely to win face-to-face rolls, force lots of armour saves and cause critical hits. Double Action and Explosive ammo is similar but not as good as just having more Burst in the active turn because it doesn't help you win face-to-face rolls. In ARO it's great though.

Suppressive Fire: The suppressive fire rules are changed substantially from N2. Now, rather than a corridor down which you can fire full Burst, there is simply a modified weapon profile with a 24" range, Burst 3 and a -3 penalty to the active model's attacks.

Suppression fire is an excellent tool for defending your deployment zone or for locking down vital table areas and it is fairly order efficient because it persists between turns and can be set up with a Coordinated Order. Although be aware that you don't get to shoot when you activate this state so try not to waste orders by overusing it on out of the way troops.

It can also be utilised with heavy weapons (such as HMGs) when you expect to be assaulted at short range and you wish to negate your range modifiers. This can make units like TAGs or HMG armed HI very difficult to engage when in the middle of the table.

When confronted with Suppressive Fire that you are having difficulty breaking try and attack the troublesome trooper from outside 24" (where they have to break their Suppressive Fire in order to react), from out of sight with grenades or with disposable troops armed with template weapons. Once the state is cancelled you can engage them normally if they are still alive.

Stack modifiers - Try and use weapon ranges to your advantage and stack this with cover, camouflage, surprise shot and similar effects. If you have a Camo/ODD equipped model then forcing a fight at long range can make it impossible for your opponent to hit. In a similar vein, don't let your opponent fight on their terms - deny them any beneficial modifiers whenever possible.

Sixth Sense L2 and 360 Visors - Don't give your opponent the chance to make Normal rolls or use Surprise Shot against you. 360 Visors and Sixth Sense can both increase variance for your opponent by forcing them to go face to face or by negating some of your penalties.

Smoke - Nothing can go wrong if no dice are being rolled. Just smoke up a corridor or objective and pass on through without worrying about costly fire-fights. Most smoke throwing troops can put a smoke grenade where needed on 15s or lower which is pretty reliable.

Of course in any game of chance things can (and will) go awry, which leads us neatly onto part three...

Mitigating Failure

"Expect the best, plan for the worst and prepare to be surprised." - Denis Waitley

Sometimes things will go wrong; no matter what you do, your opponent will pass an unlikely series of armour rolls or get a lucky critical hit. But what are the best ways to overcome a bit of bad luck?

Have lots of Orders - The more orders you have to complete a task, the more you can withstand a string of bad rolls. I'm sure we've all had our WIP 14/15 specialists repeatedly fail objective rolls and wished that we could have one more attempt. 

Use attacking and specialist models with multiple wounds - Losing an important model to a critical hit (or just a lucky shot) can be extremely disheartening but models with multiple wounds (or NWI / Dogged) can take the hit and keep on fighting.

Don't rely on one model - Self-explanatory I think. Don't build a list that is entirely reliant on a single model (or even a single link) to accomplish everything. Whilst in some games this will win big and you'll be rewarded in others you might get completely blown out. A consistent army is a good army.

Have a backup plan - Before committing to an attack think ahead about what you will do if the attack fails or even if your model is killed. Also consider how many orders you can devote to turfing out a particularly stubborn defender. If your plan requires killing a model in a single order so you have enough left to go off and complete objectives then it might be time for a rethink. Of course sometimes the reward is worth the risk.

Have a Doctor/Engineer - Yes, I know that healing/fixing your guys is inefficient - especially on turn two onwards. However, if there is a model that your plans absolutely depend on then having a way of patching them up can be very beneficial. If you are serious about using them then servant bots are a must-have too. Command Tokens can make these rolls much more reliable but you are likely to find them in short supply.


Fireteams - Fireteams get their own special mention here as they embody everything I've been talking about so far. They are order efficient, have low variance and can mitigate failure.

  • You can move between two and five models as a single order so you are able to put a significant presence wherever you need it.
  • They roll more dice and may have bonuses to Discover and BS rolls depending on the size of the team. They are more likely to pass discover checks, win face to face rolls and cause wounds.
  • If one or more members die then another can swiftly take over without loss of momentum, you don't have to spend orders bringing a new model up into an attacking position.
  • Sixth Sense L2 (four members) means that they can't be surprised by attacks from the rear and don't take penalties from Surprise Shot or Speculative/Guided weaponry - as well as other minor benefits. Reducing these penalties will greatly increase their survivability 

One thing in particular that I still struggle with regularly is placing all members of a Fireteam correctly. It's difficult enough to place a single model in exactly the right spot sometimes let alone four or five at once. Things to consider are whether your opponent will be able to make normal AROs against non-leader models, whether you are too vulnerable to templates, if you are set up for good AROs, if you are in cover and if you are in range of the current (and possibly future) link leader. 

A Fireteam is possibly the most powerful tool in Infinity but also one of the easiest to misuse.

Command Tokens - Command Tokens are a bit of an unknown at the moment, although do take a look at my strategy guide for them, and I'm sure many devious tactics for their use will come up. These are two stand-out uses that I've seen:

  • Coordinated Orders: Coordinated Orders have been substantially improved in N3 as four models may now be activated by spending a single Order plus a Command Token. This effectively gives you three Orders per Token - an excellent rate of return. This can also be used to put four models into the Suppressive Fire state at once.
  • Strategic Use: If taking the second turn, you may spend a Token to reduce your opponent's Order Pool by two for the first turn, or prevent them from using more than a single Command Token that turn. This isn't as efficient as spending them all on Coordinated Orders, but is still highly recommended to slow your adversary's advance and stop early rushes.

The rest of the applications are more situational, but it's worth being aware of all the options.

Mines - Mines (and minelayers in particular) are one of the most order efficient weapons in the game and deserve an entire article of their own at some point. They have the ability to lock down objectives, cost at least a full order to remove even under ideal circumstances and can be used offensively to devastate massed troops. Flooding the board with mines can be a very effective way of bogging down your opponent and keeping your own troops alive.

One way to fight through mines now is just to dodge through their area of effect. They are obliged to detonate against the first available target and a high PH/ARM model has a good chance of surviving.

Camouflaged Infiltrators and Impersonators - Models with both camouflage and infiltration or Impersonation are great value because they are order efficient (starting part way up the board), can reduce variance through camouflage and surprise shot and may force your opponent to waste orders on discover checks. Some, like the Dasyu, even have extra resilience. Be aware that if you keep a model in hidden deployment for several turns then you may just be wasting orders.

AD Troops: Models with Airborne Deployment get similar benefits to Infiltration, except even more precise. Garudas, Paracommandos, Tiger Soldiers, Ragiks and Gao-Tarsos, for instance, have additional resilience built in as well.

Total Reaction and Neurocinetics - This is the number one way of creating high variance for your opponent. Nobody wants to take 4 dice AROs as the risks are just too high (~19.5% chance of a critical hit) so in most cases your opponent will be forced to go out of their way to deal with them or accept that an entire section of the battlefield has been denied to them. The Sin Eater even has mimetism and ARM 3 to make him more difficult to remove!

Doctors/Engineers - Never heal a model just because you can, especially on turns 2 and 3. If you have to spend an order this turn to (maybe) get an extra order next turn it may not be the best use of your resources There are certainly times when trying to heal a model is the correct option, but it's not as often as you might think. By all means try and patch up your key troops but be ready to leave your grunts in the dirt.

G: Servant Remotes - If you are taking a Doctor or Engineer, then you'll want to attach a servant to them, unless they are in a Link in which case it may not be helpful. Servants move fast, are mimetic and vastly increase the range of your specialist. They also provide an extra model that can make discover rolls or set off mines/koalas.

Multi Spectral Visor L3 - MSV3s are one of the premier tools in the world of order efficiency even despite their nerf in 3rd edition - where they change from being able to shoot at camouflage markers to being able to just automatically discover them instead. A guaranteed success is a huge boon against camouflage heavy armies as a failed roll could signal several wasted orders.

A Challenge

“If you aren't in over your head, how do you know how tall you are?” - T. S. Eliot

Having said all this I guess I ought to put my money where my mouth is. I'd like to build a couple of lists based on the principles above and run them through a gauntlet of missions, so leave a comment with a faction and three ITS missions to complete and I'll pick one person's suggestions to work on. I will build two 300pt army lists with everything I've written in mind, justify the choices and play each of the three missions, doing a little write-up about how they went.

Win or lose, hopefully this will give me a chance to try out a new faction and put theory into practice.


  1. Nice article, especially for a new player like me. Operation: Icestorm here I come! Nomads and Yu Jing (specifically JSA) are the two factions I'm interested in. As for scenarios... I haven't gotten that far.

    1. I'm glad you've found it interesting. You won't be disappointed by Icestorm as the Nomad sculpts are beautiful. :)

  2. Curious as to how you would amend this article now that we've seen N3 reduce the opportunities for normal rolls.

    1. Just so happens that I've been working on an amendment for this article already and it's now been finished and updated. :)

      You are absolutely right that there are almost no Normal rolls left in the game as it is possible to Dodge almost everything. In the reactive turn I expect people will be Dodging a lot now, wasting their opponent's orders and preserving their own models for as long as possible, rather than taking slim chances on shooting.

      Some of the biggest changes from N3 are Sixth Sense 2 and Mines being less important, Command Tokens effectively giving players extra orders and Suppressive Fire becoming a more useful tool.

      Using Command Tokens and Suppressive Fire effectively is going to be avery important skill to learn now.