How to Evaluate Artifact Cards - Neon - August 29, 2018

Screen Shot 2018-08-29 at 12.59.13 AM.png

PAX is so close I can taste it. Just a couple of days and we are going to get a whole pile of new and wonderful cards to pore over, analyze, and talk about. This is extremely exciting, but it raises a question: how can we tell what is actually good?

Obviously, there is no way to know for sure. We are going to need to actually play with the cards to know where they stand. Still, it is fun to theorize and speculate, so I thought I might suggest some guidelines to help think about what makes cards good or bad. Let’s get to it!

How do they interact with Creeps?

We may not know a ton about Artifact, but we do know there will be a lot of Creeps. You start the game with Creeps on board, you get new creeps every turn of the game, and there are even cards that generate additional Creeps. They may not be flashy, but these little 2/4s are going to be incredibly important, especially in the early turns of the game before larger creatures come online. Just think about how they impact gold income through the early game. If you are effective at clearing out your opponent’s Creeps, you will be able to buy items faster than your opponent, which might allow you to begin to snowball one of the lanes to generate more gold, and so on. For this reason, heroes, spells and other cards should really be evaluated in how they match up against creeps. Let’s look at a couple examples.

Screen Shot 2018-08-29 at 1.14.42 AM.png

Chen is a 4/9, meaning that he one-shots creeps. Any creeps that spawn in front of him will be easily handled, clearing the way for attacking the enemy tower. Good stuff!


Rix only has 3 attack. While the difference between 3 and 4 attack may not seem like a lot (it is just 1, I did the math) this will have a massive impact on how the game plays out. On turn 1 if there is a Creep in front of Rix they will attack each other and both survive. On the following turn Rix will be forced to hit this Creep again, meaning that 2 damage is wasted. In addition, we could imagine another Creep appearing nearby, and now Rix could be taking a lot more damage. If another Creep spawns in front of Rix on the following turn, there is a good chance he might die just to creep damage. As you can see, things can get awkward fast, all because Rix is missing 1 point of attack. Obviously other stuff can happen, but the natural play pattern of Rix will play out far worse than the natural play pattern of a hero like Chen. This isn’t to say Rix is bad - he has other features that are quite attractive - but missing that 4th point of attack means he is extremely reliant on buff effects.

Screen Shot 2018-08-29 at 1.16.01 AM.png

(Rix is naturally a 3/7, but there are no great screenshots of that. Not really important here, but for the record)

Screen Shot 2018-08-29 at 1.17.17 AM.png

Zeus is actually a fascinating exception. He has the same base stats as Rix, which suggests that he would have the same issues. While that is true, Zeus's passive is very important. This lets him “ping” enemy Creeps, which turns the math in his favor. As long as you are able to cast 1 blue spell per turn he “virtually” has 4 attack, meaning he is an effective Creep killer.


Lycan is another interesting case. He himself obviously has 4 attack, which is important, but his passive ability gives neighboring units +2 attack. This means that your neighboring Creeps will be able to 1-shot your opponent’s Creeps. This could be a huge deal in the early game, especially if Lycan happens to land on a crowded lane.

Screen Shot 2018-08-29 at 1.19.42 AM.png
Screen Shot 2018-08-29 at 1.21.48 AM.png

This type of analysis doesn’t just apply to heroes. Disciple of Nevermore has a big impact on “combat math”. By giving your Creeps +2 attack it means your Creeps can now kill your opponent’s Creeps, but the -2 armor now leaves your creeps vulnerable to a counterattack. In case you weren’t aware, “negative armor” is part of Artifact, and it means your unit takes extra damage from damage sources. Ultimately, when you compare Disciple of Nevermore to Creeps, you will see that it is extremely fragile, and does not really improve your Creeps’ ability to fight your opponent’s Creeps. While Disciple of Nevermore is certainly good for pushing damage, it is not so good at managing Creep combat, making it a situational card.

Interacting with Heroes

It is difficult to overstate how important heroes are. Heroes engage in combat. Heroes let you cast spells and creatures. Heroes carry items. Any card that lets you interact directly with your opponent’s heroes has the potential to be incredibly powerful. We don’t have many examples, but let’s take at one of the few we do have.

Screen Shot 2018-08-29 at 1.27.20 AM.png

At first this actually looks like a lousy deal. 6 mana? You need to discard a card? At random?! Horrible, right? Well, not so fast. Imagine you use this against the only hero in a lane on a key turn. You might invalidate your opponent’s whole turn! Also, imagine they play a card like Divine Purpose on their Hero.

Screen Shot 2018-08-29 at 1.29.42 AM.png

If you can respond to this with a Coup de Grace you have totally countered this powerful buff effect in addition to answering the hero itself. Seems really good! We have seen very little in the way of “hard” removal for heroes so far, but any that we do get that is reasonably costed is likely to see a lot of play. There is more than one way to interact with a hero though.

Screen Shot 2018-08-29 at 1.31.26 AM.png

Disarming a unit stops it from attacking for a turn, though it can still take damage. A strategically deployed disarm effect could easily set up a kill on your opponent’s hero over a couple of turns. Imagine if your opponent’s strategy involves dumping buffs onto one of their heroes. Disable effects might really trip them up.

Screen Shot 2018-08-29 at 1.33.22 AM.png

I have been very curious about the meaning of “silence”. We know that silence is a mechanic in Artifact from the gameplay videos, but we don’t know what it means. Does it cleanse modifications off a unit? Does it disable the “color function” of a hero? Stop activated abilities? Having an answer to these questions will help us evaluate the impact of cards like Truth to Power.


As with many things in life, Artifact decks are more than the sum of their parts. Cards that combine well with other cards to have an increased impact are extremely desirable. Take a simple example like Dimensional Portal.

Screen Shot 2018-08-29 at 1.35.33 AM.png

Getting 3 Creeps for 4 seems like a decent rate on its own, but what if you pair it with another card that scales with the number of creatures you have in play like this?

Screen Shot 2018-08-29 at 1.36.38 AM.png

Or this?

Screen Shot 2018-08-29 at 1.37.48 AM.png

Both Act of Defiance and Assault Ladders scale with the number of creatures in play, so a spell that throws down a lot of units will be extra impactful. Cards with strong synergy potential may be more impactful than they superficially appear. These are pretty obvious examples. A more subtle example would be the synergy between Conflagration and…. Conflagration.

Screen Shot 2018-08-29 at 1.39.01 AM.png

You might think that I am joking, but I am actually totally serious. This card seems powerful as a base, but if you are able to get 2 copies in a single lane you will be able to fry every Creep in a lane before your opponent even has a chance to act (in addition to softening up your opponent’s heroes)! Cards like Conflagration are said to “stack well” or are “good in multiples”, which can be a very attractive feature. All this is to say that synergy is good, even if that synergy is with more copies of the same card.

There are other cards that could be said to have “diminishing returns” meaning that the second and third copy tend to be worse than the first. We don’t have too many cards like this yet, but situational effects like “condemn an improvement” might be a good example. It is obviously useful to condemn powerful improvements sometimes, but what if your opponent just doesn’t have any improvements in their deck? There is a chance that cards like this could be a dud, especially when you draw multiple copies.

Hate Cards/Counter Cards

Synergies are great when you are left to your own devices, but what if you actually need to interact with your opponent? They are trying to execute their game plan, so shouldn’t we try to stop them? We should! In fact, there are already a bunch of ways that we can interact with our opponents to disrupt what they are doing and break up their synergies! One of the most obvious in my opinion is a card like Conflagration, which is very effective at countering “go wide” strategies that use cards like Dimensional Portal to flood the board with small units. Another example might be a hero like Pugna.

Screen Shot 2018-08-29 at 1.45.12 AM.png

You can’t see it here, but Pugna has an active ability where he can condemn a random enemy improvement every 3 turns. While this ability will not be great in every game, this is an extremely effective counter to expensive improvement cards.

It is hard to point out great examples of counter cards without more information about the most powerful and popular deck archetypes, but keep your eyes peeled for cards that effectively hate on potential strategies. Pugna is only interesting as a hate card if improvements are seeing play, which isn't necessarily always going to be the case. Part of the reason that Pugna is a good example is that the "risk" is fairly low. His stats are pretty solid for a hero, so assuming his signature card is decent, the cost for including improvement hate is extremely low. Speaking of "solid stats", let's talk about the vanilla test.

The Vanilla Test

The last concept I will mention is the “Vanilla test” which is a term that comes from Magic the Gathering. A “vanilla” unit is a unit with no text other than its combat stats and its cost. Creeps are an example of a vanilla unit, as they do not have any relevant rules text without some augmentation. The “vanilla test” is the practice of measuring a unit entirely based on just its stats. If the base stats are good, then we know it is a card we want to play (as long as there is no horrible drawback). If the base stats are bad, we know the card needs to give us back a whole lot of power through its abilities. We don’t have quite enough information yet about Creeps in Artifact to really develop a meaningful vanilla test, but we have enough information about heroes that we can construct a rough “Vanilla Hero” test, so let’s take a look.

Screen Shot 2018-08-29 at 1.52.14 AM.png

How do I put this gently….. Ummm… Crystal Maiden’s stats are… dog shit. Like, just god awful. She is barely better than a melee creep. In terms of the “hero vanilla test” she fails. Badly.

Screen Shot 2018-08-29 at 1.52.50 AM.png




Axe, on the other hand, passes with flying colors. His chad-tacular body is top-tier. 7 attack is a lot, 11 health is a lot, and 2 armor is a lot. It has been suggested by Valve that Axe was designed to be the biggest dumbest bruiser, meaning that it is unlikely that anyone will challenge him based on combat stats alone.



Screen Shot 2018-08-29 at 1.55.00 AM.png


If Axe and Crystal Maiden are the extremes, then there are obviously a bunch of heroes in the middle. Sorla Khan’s attack is above average, while her health seems a little low, meaning she is about average on balance. Chen (who was shown above) has high health but lower attack, meaning he once again is about average. Zeus and Rix both appear to be slightly below average. I should briefly mention that I think Armor is really good, so a hero with 1 armor and 4 health is worth a lot more than a hero with 5 health. If we count each point of armor as being worth “1.5 points” we can come up with a simple scale.

Combined Stats Vanilla Test Score

10 or fewer - Poor

11-15 - Average

16 or more - Excellent

This is my guess based on what we have seen so far. These standings aren't just "for fun" - they really matter. A poorly stated hero is less likely to kill the enemy heroes, and is more likely to feed, giving your opponent a gold advantage and a lane advantage. A hero with excellent stats can generate a gold and lane advantage in the early game, which can translate into a stronger mid and late game. Chain feeding into heroes like Axe and Bristleback could lead to some disastrous games.

It should be noted of course that heroes are not just balls of meat. For example, I haven’t even talked about abilities! Take Crystal Maiden. Her ability gives you 2 extra mana every time you cast a spell. This is really powerful, and in the right deck could even lead to some crazy combos! She also has a “signature card”, which could be a lot better than Axe’s “signature card”. This doesn’t even mention the differences in color! Blue spells are likely systematically better than red spells given that red heroes seem pretty overstated compared to blue heroes. All of these things matter when evaluating a hero, but it is important to start with the baseline of “how does this do on the vanilla test”. Crystal Maiden is the clearest example of this in my opinion. She fails spectacularly at the vanilla test, meaning that her passive and her signature card need to be insane to compensate. Both her ability and signature card seem good, but they don’t seem strong enough to compensate for her puny body. Unless you can find a way to synergize or combo off with her passive ability CM is probably trash. 

As I mentioned briefly above, we don’t yet have enough info for constructing a meaningful vanilla test for creatures. As new cards come in over the weekend we will begin to construct our own vanilla test for everything. What is the average size of units at different costs? What is the average stat bonus we get from items of different costs? All this information will be extremely useful for evaluating cards moving forward.

As a closing thought, I should briefly touch on how to mix together your thinking on each of these topics. Some cards might be good on one score, but bad on another. How should we evaluate such cards? Well, that is part of the fun and the mystery! Weigh the various advantages and disadvantages, and then decide where it lands! This is certainly not an exact science.

That will do it for today! Hopefully that is helpful for processing PAX cards as they come at us. I am extremely pumped for these card reveals, and I can’t wait to see everything this weekend. What do you expect to see? What are your predictions? Share your thoughts on the Reddit thread! Also, make sure you don't miss my conversations with Cymen and Anger on the Secret Shop podcast, where we discuss some of our card evaluation, as well as joke about Artifact stuff. Expect lots of great content about Artifact on A+Space next week!