Artifact has been out for about a month now and I've been playing it a lot. By now I have a pretty good grasp on the game's mechanics, I can easily tell when mistakes are made and I've reached a pretty good rank (67 out of 75). Although because of the way Valve implemented the ranking system I can't tell if this number means anything. Either way, the point is that I'm fairly experienced at the game at this point.

The main reason for writing this article is that Artifact is a game with a lot of sources of randomness and this makes it the perfect environment to exercise the mindset I went over in this article about luck. And so that's what this article will talk about!

If you already play Artifact you can skip to the Luck section.

RNG sources

Artifact has about 5 types of RNG sources that I can think about right now: hero placement and creep placement, arrows, item shop, and the effects of some specific heroes or cards. Let's go over them one by one.

Hero and creep placement

Every deck has to have 5 heroes and you get to choose the order in which they are deployed. Since the game has 3 lanes, the game will start deploying the first three and then the last 2 over the next 2 turns.

In the image above, we have a deck with the following heroes: Venomancer, Necrophos, Zeus, Ogre Magi and Tinker. The first 3 will be deployed first, and this is what the board state could look like right after the game starts:

As you can see, the first 3 heroes (Veno, Necro and Zeus) are deployed against 3 enemy heroes. The catch is that creep deployment is randomized, and so in the picture above you can see that the first lane has two creeps on our side, the second has none, and the third has one. If creep placement is randomized, the chances that our heroes will face the opponents heroes are also randomized. In the picture above we got lucky and our heroes faced none of the enmy heroes. But check this deployment out:

In this one, our Necro is facing the enemy Bristleback mid and immediately dying. And Zeus is facing Viper right and dying next turn.

This source of RNG is the first one that you have to learn to deal with, and the general way of by placing stronger heroes first (as the first 3 deploys) so that they can trade better. Often times this won't be possible (like in a blue deck because blue generally has weaker heroes), but more often than not it is. And this also has to be weighted against the fact that losing a hero early is hardly ever a huge advantage to the enemy, so sometimes you'll want to place weaker heroes first because you have better cards for that hero's color in your deck, even though there's a chance it will die immediately.

For instance, blue has a lot of cards that control the board state (ally/enemy placement and arrows), so if you have lots of early game cards that do that, placing blue heroes first isn't that big of a problem. In the deployment above, Zeus is facing Viper, but in our hand we have a card called Cunning Plan:

Which means that we can go from this:

To this:

The point being, hero and creep placement at the start is random, but there are many ways around it. Finally, as rounds go on, you can choose to place next heroes on specific lanes, but you also can't choose their specific positions, nor the positions of your creeps. So for instance, the next round of the game above looks like this:

And so now we have to choose where to place the Ogre Magi. If we place him mid he will 100% of the times face Bristleback and die immediately. The only scenario where that doesn't happen is if the enemy places Magnus mid also and then there's a 50% chance that Ogre will face Magnus, but relying on this would be a big mistake, especially given that we don't know if the enemy will place Magnus there.

Placing Ogre right has a 50% chance that he will face Viper, given that we also have a creep going there and either the creep or Ogre would be placed in front of Viper. And finally, placing the Ogre left has a 33% chance it will be in front of Ursa and die immediately. Additionally, we might want to consider where the enemy will place Magnus, which will probably also be left, given that the first lane is generally the strongest one that you want to contest and win. If the enemy does that we're also good, because even though our Ogre might die in 2 turns if facing Magnus, it's better than all the other options we have.

And so in this case what actually happened is that we got unlucky and our Ogre faced Ursa, and on top of that he played a spell that gives him cleave and so he will just kill everything that lane:

This is the exact type of scenario that happens when you have sources of RNG like that in the game. We chose the best option out of what we had and it still didn't work in our favor!

Arrows

Arrows are another source of RNG in Artifact which dictate where the unit will attack in case there's nothing in front of it. There's a 50% chance the unit will attack the tower, and a 25% chance it will attack a neighbor to either side. In the image below, for instance, the enemy creep routed to our Necrophos instead of dealing damage to the tower:

In this image we have 3 enemy heroes routing to a creep in a very unlucky manner and overkilling it by a lot instead of dealing more damage to the tower:

And the examples are endless. This is also something that you can play around by having cards that deal damage and remove enemy creeps/heroes altogether before the combat phase, or by using cards that manipulate arrows, like Compel:

Either way, it's another source of RNG that exists and is fundamental to the way the game works.

Item shop

Every deck has to also have an item deck. These are at least 9 item cards that can be bought with gold in the game. Gold is acquired by killing creeps (1 gold) or enemy heroes (5 gold). Between each round you will be shown the item shop and it will you give you 3 options to buy from:

On the left you have the secret shop, which as far as I can tell gives you the option to buy a random item from the pool of all items in the game. The secret shop will occasionally show items that are really good to have if you have the gold to buy them, but that you wouldn't necessarily put in your item deck for many reasons. You can also spend 1 gold to hold that item so that it shows up on the next turn in case you don't have the gold to buy it right now.

In the middle you have your item deck. These are the items you choose yourself. They are also shown in random order each round, so a strategy that people developed is to place 6 very cheap but useful items in the deck, and then 3 really good items that they actually want. This means that they can buy the cheap items more easily and also find the 3 really good ones with more consistency. For instance, this is an item deck I just drafted:

The 3 really good items here are the bottom ones, while the top ones are the cheap filler ones. Depending on your heroes and cards you might not want to do this. For instance, if you have lots of heroes that can trade well and get kills you can go for more expensive item decks. You can also do that if you have multiple Paydays or Iron Fog Goldmines:

Finally, in the right you have consumables, which are health flasks, a potion that draws a card, or town portal scrolls. These are also shown randomly. One thing about this game is that heroes can get "stuck" in lanes that are dead. For instance, if you placed 2 heroes in one lane and won it, but now you want to place them in other lanes so that they can do work there, you either have to find a way to kill them or use a town portal scroll. Because town portal scrolls are randomly spawned in the consumables section of the item shop, you may not get one when you need it. This is another source of randomness that can be played around by always buying portals and never using them needlessly, but it's still a source of randomness that can fuck you over.

Heroes and cards

And finally, the last main source of randomness is in how some heroes or cards work themselves. The best example of this is probably my favorite hero, Ogre Magi, who has this passive ability:

When you get multiple multicasts with this or win because you multicasted the right spell at the right time it feels really good, and when you lose because the enemy did that it feels really bad. But it's still something that can be played around in many ways. To play multicasted spells the enemy has to have mana, and generally if he's playing high impact spells those are costing lots of mana so he has to use one either Aghanim's Sanctum or Satyr Magician to replenish it:

Both of which can be removed from the game. One being an improvement which can be destroyed, and another being a creep which can be killed. And you can also just kill Ogre Magi himself which makes the passive not work.

Another source of randomness that people don't seem to like is Bounty Hunter's Jinada:

Bounty Hunter is a 7/7 hero, if Jinada procs it makes him 11/7, which is enough to kill most heroes in the game in one shot. The way to counter play this is by simply always assuming that it will proc and trying to play around that. Being a 7 health hero makes him kinda squishy too so while he is a big threat it can be dealt with pretty well.

There are also spells that have some randomness to them, like Chain Frost or Eclipse:

With Chain Frost you can choose the starting enemy from where the bouncing will start so you can generally math it out before using it if it's likely to kill everything you need or not. And with Eclipse you can also calculate if it's worth it or not. If you have an enemy you want to kill that has 6 HP, he has two creeps with 4 HP by his side and your Eclipse has 5 charges, it's very unlikely that you won't kill the 6 HP guy. There's a chance the spell might hit both creeps twice and the other one only once, but that's less likely to happen than it hitting either creep only once.

Finally, there are cards that have affect things randomly like:

Lost in Time locks 3 random cards in the opponent's hand for 3 rounds, which is a really good effect. You can't control which cards get locked, but this card is best used when opponents have few cards available to play. Generally when people are doing really well in the game they're playing all their cards every turn and being left with few cards in their hands, meaning that no mana in any lane goes wasted. And some people even go so far as to make "spend mana effectively" one of the primary goals in their way of playing the game, which is a fair way to approach things. Lost in Time though punishes players that are doing that too well, which is a great way disable key plays that the enemy might have on this or the next turns (given that 6+ mana turns are when the game starts getting dangerous).

Golden Ticket is the best item in my opinion, I really like running it whenever I can because it gives you any random item from the pool of items in the game for 9 gold. When this works in your favor it's really really good because you essentially pay 9 gold for a 20+ gold item that has really high impact in the game, and generally you can get this high impact early on, which makes it even more impactful. Obviously people don't like it because it's random, but yea.

Luck

Now, with all that said about sources of randomness in the game, I can get to the meat of the article here. As hopefully I've explained above, the game has various sources of randomness that can be mitigated directly in various ways. For the things that can't be mitigated directly, you can usually set up your board in ways that will mitigate those things indirectly over time. This feels bad and counterintuitive for people in the wrong mindset but it's the way these games go. If the best play in one instance works 70% of the time, you still do it, even if 30% of the time it won't work, because it's the best play all things considered.

Artifact is kinda like Poker in this sense, which falls into the category of high luck and high skill games. There are sources of RNG and you have to know your probabilities and costs to every action to take the statistically correct course of action most of the time. This won't help you win the game 100% of the time, but it will increase your win rate significantly above other people. Sometimes you'll get unlucky, but as long as you can identify the correct plays, over time you should be able to succeed.

With all this in mind, Artifact is the game that is most perfect for me to exercise the mindset that I explained in this article. The mindset is that essentially you filter the world based on a combination of both conscious ideas you have about the world but also what you trained your body to pay attention to over time. In the case of luck/RNG this makes itself very evident in Artifact: people who have trained their bodies to pay attention to the role that luck plays in life, will see the game in terms of luck.

A very simple example of this is people who say that they lost because of some unlucky arrow. This is a clear case of someone who is only paying attention to unlucky events and not how they could have played things better throughout the game. It is the case that you can miss lethal on a lane because an arrow went the wrong way and lose the game, but it's also the case that you could have spent more resources or other lanes so that you wouldn't lose the game in case you couldn't get lethal this turn, right? But instead of thinking about how they could have played things differently, people will focus on the arrow because they want to blame external factors instead of themselves. Two current examples from Artifact's subreddit:

The number of posts and people complaining about this are endless. And as I've explained, these people are simply wrong. Worse than that, they're in a downward spiral that prevents them from improving. Once you start noticing only how luck affects things you will start noticing it more and more and this will prevent you from thinking about how you could have done things differently, which will make it way harder for you to improve. This goes for Artifact and literally anything in life, which is the point I make in the article about luck.

The solution to this is to slowly train yourself to not blame external factors like luck. Whenever you lose a game, think about how you could have played differently instead of how you got unlucky. Catch yourself if you can whenever you blame luck. This is hard to do and takes actual effort. Even a number of Artifact streamers who are pros at the game still do this and occasionally blame RNG for losing a game. It's a really hard thing to overcome consistently. But it's worth it to train yourself to avoid doing it, since it's a very useful skill in general.

On a side note, this is the main role that I think games play in life. Unlike movies, games are active and because of this you can use them to change the patterns in your body or mind. In this case, learning to never blame luck in Artifact will train your body to never blame luck in general, which is a very good thing to do when you're trying to do anything in life, be it making a game, starting a business, getting a girlfriend, and so on.

In the end this is why I like Artifact, the main skill that it exercises is one of learning how to deal with luck, which is a fundamental skill in life that once mastered provides tremendous gain. On another side note, this is also how some types of games that we like now were created. In the old days simulation games were created to help generals train for battles, and the random elements in those games were derived from statistical analysis of what kinds of things would go wrong in real battlefields. This meant that the better generals got at playing the game, the better they got at dealing with all sorts of random unlucky events that can happen in a real battlefield. These morphed over time and became all sorts of game mechanics that games have today.

Armies lose battles when their generals are unable to adapt to the situation. — Legion Commander on maintaining poise

a327ex / blog