Defending against Frostbites in Prismata


I haven’t written a true strategy article in a little while, but today I’m going to post a few words on one of the most controversial Prismata units—Frostbite. Though recently nerfed (its buildtime increased from 1 to 2 in the last big set of balance changes), Frostbite remains a formidable unit that must be dealt with very carefully.

The current Frostbite. Looks cute, but it’s a nasty, evil little thing.

Frostbite is certainly a unit worth studying, as it appears in more games than many other units due to Frost Brooder and Endotherm Kit spawning copies of it.

Speaking of Frost Brooder… you may recall that it was temporarily removed in our last balance patch (because we wanted to modify it in a way that required some small tweaks to the Prismata engine and bots). Well, we’re happy to announce the new version:

asdf

The new Frost Brooder! It will go live this week. Its supply is still 4.

The new Frost Brooder is actually 1 gold cheaper than its previous incarnation—its cost has been decreased from 5RR to 4RR. However, it comes with a huge caveat: the Frostbites that it spawns have Lifespan 3, meaning that you only have 3 chances to use them! This makes it much more difficult to store up a huge pile of Frostbites, and presents a ton of really interesting strategic questions, since your Frostbites are now “use-em-or-lose-em”. We had a ton of fun play-testing this one and we think you’ll enjoy it.

Now, onto the main topic of the article (a part of this has appeared on Reddit before, but the post has been somewhat expanded):

How to Defend Against Frostbites

First, a disclaimer: this isn’t really a “strategy article”. It’s more of a “tactics article”. I’m not going to discuss Frostbite’s strengths and weaknesses, or how it’s much weaker against Infusion Grids than it is against Walls, or whether it’s better to fight against Frostbites by going breach-proof or spamming Xeno Guardians. Instead, I’m going to focus on one specific question:

Suppose you have a bunch of Walls, Steelsplitters, Drones, and Tarsiers. Your opponent has a similar army, plus a huge pile of Frostbites. How much defense should you hold back?

The reason I want to address this one specific question is that I see weaker players making the same mistake over and over again: overdefending against Frostbites. Moreover, misunderstanding Frostbite defense actually causes players to overestimate the value of Frostbites.

Here’s one of many quotes that I’ve seen floating around on reddit:

What is the complexity behind buying or using Frostbites? Almost none… you basically just stack them until you can breach.

I think this quote echoes a common sentiment among many Prismata players. But I think it also betrays a misunderstanding about how Frostbites should be played (and played against). There is something crucial about Frostbites (and Prismata game theory, in general) that many players haven’t come around to learning yet.

This is a misunderstanding of a concept that’s quite complex and nuanced, but I think it lies at the core of why A LOT of sub tier-X players…

  • Don’t like Frostbites
  • Overestimate their value
  • Misplay while defending against them
  • Believe that Frostbites do not create hard decisions for the attacking player

If those opinions sound like yours, then rejoice! You are about to get much better at defending against Frostbites.

 

A Proverb

I will start with a “proverb” that is true in many strategy games, and definitely true in a lot of Prismata situations:

An optimal defensive play should make the attacker’s decision hard.

Frostbite defense/attack situations are a perfect example of this proverb in action.

Consider a game where your opponent has a bunch of Frostbites, and you have the choice of either:

(1) Holding back a lot of Drones/Steelsplitters to defend fully, so that no breach is possible, OR:
(2) Working/attacking with some of those units instead to gain gold/attack from them, but allow your opponent to blow 5 Frostbites to breach and kill a Tarsier.

Most beginners make the mistake of always choosing choice (1) and holding back too many drones. Instinctively, they try their hardest to prevent a breach at all costs (which is a good beginner heuristic, but starts to break down once more advanced units are added).

But suppose you go with choice (2). Let’s look at the pros and cons. You’ve made yourself vulnerable to a breach, and given your opponent the option of giving up 10RRRRR worth of Frostbites to kill a Tarsier, plus maybe 1 extra HP worth of defenders (instead of attacking for N, killing N-2 HP worth of defenders and having 2HP absorbed by a wall, the opponent is now killing N-1 HP worth of defenders + 1 Tarsier). In summary:

If your opponent chooses not to breach:

  • They keep the Frostbites for a future turn.
  • They kill [N-2] HP worth of defenders.

If your opponent chooses to use the Frostbites and breach:

  • They lose 5 Frostbites (value 10RRRRR).
  • They kill [N-1] HP worth of defenders (1 more than above).
  • They kill a Tarsier.

All things considered, the latter choice probably isn’t a great trade for the opponent. They are probably losing more value by giving up the Frostbites than they’re gaining in extra damage. It might be better to save the Frostbites for a larger breach later.

For this reason, defensive option (2) looks really strong. We get extra attack/gold from the Drones/Steelsplitter, and our opponent is literally going to do the same thing that he or she would have otherwise done: NOT use the Frostbites.

(Side note 1: the 10RRRRR cost I mentioned above is actually a “sunk cost” at this point and isn’t really what’s important; the decision hinges more on the opportunity cost of no longer having those Frostbites in future turns (e.g. the Frostbites could be worthless in future turns if the defender switches to Venge Cannons.) But cost can sometimes be a useful proxy for value.)

 

Defending Optimally

OK, so why not take things a little further? Well, if we work/attack with all of our Drones/Steelsplitters, then maybe the opponent can trade those Frostbites in for a huge breach, killing FIVE Tarsiers. The opponent would almost certainly take us up on this offering, and we’d be way behind.

So what is the optimal number of Drones/Steelsplitters to work with? The answer is… AT LEAST ENOUGH TO MAKE THE OPPONENT’S DECISION DIFFICULT (and possibly more).

Suppose we leave back the perfect number of Drones/Steelsplitters so that the opponent sees the decision of using the Frostbites vs not using them as a 50/50 split. In other words, the opponent considers the two options of [blow the Frostbites now] and [save them for later] as being equally strong.

The thing to realize is that in this scenario, relative to leaving back a ton of Drones/Steelsplitters to guarantee that a breach isn’t possible, we’ve not improved the opponent’s position at all. We’ve given the opponent an additional option, but if that option carries the SAME value as simply not using the Frostbites, we’re no worse off than we would be if we simply defended. However, by working/attacking with more Drones/Steelsplitters, we get extra gold/attack, which adds value to our side. We get free extra value, at no cost to us.

So the point is, the optimal number of Drones/Steelsplitters to work with is at least enough to make the opponent’s decision difficult.

You might have heard this concept by the name of “minimaxing”… we want to minimize the value of the opponent’s best (maximum) play. If we can gain value (by working with Drones) at the cost of increasing the value of one of the opponent’s worse options (blowing the stack of Frostbites), then we should continue doing so until the opponent’s two options are equally good.

(Side note 2: It can sometimes even be correct to work with an extra Drone beyond this to the point where your opponent’s decision leans slightly on the side of blowing the Frostbites, if the extra one gold is worth it.)

In any case, the point is this: if you want to play optimally against an opponent with a huge stack of Frostbites, you should almost always be allowing your opponent a small breach. And more generally, you should always be making your opponent’s decision difficult. If your opponent has an easy decision (e.g. it’s 100% clear that they’ll choose not to use the Frostbites), then you’ve probably overdefended.

 

Still think Frostbites are easy to use?

So, this then brings me back to my original point. I’m going to dump that reddit quote here again for reference:

What is the complexity behind buying or using Frostbites? Almost none… you basically just stack them until you can breach.

From what we’ve learned above, we can conclude the following: if you are playing Frostbite games where it feels like the attacker has it easy and doesn’t have to make difficult decisions, then the defender is probably misplaying.

I would even go as far as to say the following:

In most cases, if an attacker has a large number of Frostbites and the defender is playing optimally, the attacker should have a hard decision regarding the use of those Frostbites almost every single turn.

I say “in most cases”, because of the following exception: the defender might not always have the option of making the attacker’s decision difficult. Some Frostbite games are incredibly sharp: a single breach might result in a dead Lucina or Centurion and a lost game. However, the above rule holds true whenever the marginal value of a breach is a single Wall/Tarsier or similar unit, which is often worth less than the attacker’s stack of Frostbites.

 

The end.

I hope you enjoyed this “tactics article”. Good luck using Frostbite in your upcoming games! To finish things off, here’s a Frostbite skin showcase! All these will be available when arena rewa*ds go liv* on the Pri*m*t*.s**ver.on.Ju*y.2***.******……… [Transmission lost.]

 

 

 


About Elyot Grant

A former gold medalist in national competitions in both mathematics and computer science, Elyot has long refused to enjoy anything except video games. Elyot took more pride in winning the Reddit Starcraft Tournament than he did in earning the Computing Research Association's most prestigious research award in North America. Decried for wasting his talents, Elyot founded Lunarch Studios to pursue his true passion.