Here are my thoughts about topics surrounding Artifact, long post ahead, even trying to keep each separate point brief.
---
On the business of games:
First, about Richard Garfield. For me, that name means nothing. Great games in the past are bad by modern standards and often succeeded from luck. Diablo by modern standards is a bad game. DOTA 1 by modern standards is a bad game. And MTG (A/B/Unlimited) by modern standards is a terrible game. MTG had the great help of many designers that came after and cleaned up the mess. Even today, I think it is still not that great a game, it's just (a lot) better than the crap out there.
Second, about monetization. In the modern era, games need to capitalize on users that continue to play. A person that plays 1000s+ of hours and only pays $50 is a bad deal for the developer and represents a lot of uncaptured value. This is why Diablo receives so few updates - the bottom line doesn't justify it.
I think the correct model that people were moving towards is, let's call it "ceiling pay to win". Which is, there's a cap on investment to be competitive, similar to getting all the playables in Hearthstone. This is also based on the concept (not yet fully accepted by the public) that yes, if you like a game and play it obsessively, it is legitimate for the developer to ask you pay $150 a year or whatever (no different from a WoW subscription.) I think the only way for games to stay balanced and fun (and avoid powercreep problems) is by having rotations. The main problem with rotations is that it does not give a permanent reward to investment in the game, but it's a problem that can be solved by a clever designer.
---
I think the next thing to fall is the idea that constructed is an integral part of CCGs, but people haven't realized it yet. This is based on the weakening of the notion that the cards even have to be "fully collectible" - as probably making them not tradeable was better. This is a nuanced point - the cards being tradeable is better for constructed as it allows a cheaper price for players to enjoy the game through experimenting with tier 2-3-4 decks, but there are also costs in that you cannot normalize the luck of packs (they all must have the same EV) which means you can frustrate players paying $200 for 100 packs and not having any tier 1 playable decks, and also you do not capture the theoretical value of these trades.
Already there is surprisingly wide agreement that Artifact constructed probably blows, and I think that realization will eventually cause a major shift in the psyche towards developing games that are "limited first". Traditionally for CCGs, the support of constructed formats have mostly been to justify card sales, not because it's a format that people demand in particular. In theory, limited is inherently more fun as there are more diverse gameplay situations, and more corner cases where unusual lines of play can be correct. The unfun-ness of constructed is also exacerbated by netdecking, as players quickly close in on the best decks which leaves the meta stale. Overall, paying $100 or whatever to play constructed is not really a good value proposition because of the repetitive gameplay states.
I think phantom draft is a good litmus test for the public's tolerance for paying for limited, like an arcade, as the variable is not confounded by the user's desire to play constructed. You pay $1 for a draft and may get some money back, so it's more like $0.60 or so. It has never seriously been tried in a CCG to put limited first and extract the majority of money that way. Once a better business model is figured out for card games (and believed in), you will see much better games come out that are limited first, but this will take time. Think "the League of Legends of card games".
===
On the gameplay:
First, a longwinded discussion about the mechanics of card games. The reason why games like MTG are fundamentally bad is because they have a problem of "non-games". This is a much bigger deal than most people think. It is not just the 20% of the time that you are completely screwed/flooded and I am not, but it is also the closer positions where eg. I am able to play 15 mana of stuff, and you are only able to play 11 mana of stuff. In those closer positions, which occur very frequently, the cards need to have higher variance to compensate.
So for example, I can play a 4/4 flier you can't deal with, I can play a board clear and get a 3 for 1, I can play a planeswalker that threatens to take over the entire game. If we look at a distribution of the difference between two (of the same) binomial distributions [ie. difference in amount of mana of stuff we can play on turn N], it is quite fat tailed, and this itself warps the design of the game. It also warps the design of the game in that consistent decks (ie. curve out by 4) are inherently stronger, and so to compensate the swing by the control player needs to be even more severe.
Another example would be in League of Legends, as a level 1 character you need 20 hits to die, but in a level 18 fight you can get 1 or 2 shotted. The (very high) increase in variance is to give the player behind a better chance. However, what this does is make the game less fun, similar to how open face chinese might be less fun if the deck had 4 jokers, as the meaningfulness of your decisions are reduced, and it's largely about who draws more jokers. This isn't to say that RNG is bad (it's not), but just that games like "slay the spire" or "dominion" (or open face chinese, surprisingly) execute on RNG elements a lot better than games like MTG which are fundamentally handicapped.
Where I am getting at is, Artifact [and HS, etc], as a game where cards are gated by mana (ie. some % of the time, you are not allowed to play this card), is still subject to these effects, although admittedly a lot less so than MTG. Again, this warps the design of the game by making the game bomb dependent, as you may often be missing early plays.
Why this matters is the same reason HS blows today: whereas before, playing a 4/5 on turn 4 was a good play, and the decision of whether to kill that 2/1 or go face is interesting, now you have people vomiting huge boards and making insane swing plays every turn. Artifact reminds me of modern HS in that it seeks to do the same. The problem is we've tried this, and HS lost a lot of gameplay appeal from this problem as whales started to quit. The reason they cited in reddit posts, is because the average meaningfulness of decisions was greatly reduced, which means that the learning loop of being able to learn from their mistakes was reduced, which is a core component of fun for longterm players [due to the relationship between learning and dopamine].
---
So what does this mean for Artifact? I have a few problems with Artifact:
First, it's very hard to read the board and the gamestate. A *single lane* can frequently have 8+ minions (one time I saw 20), and the equipment and abilities are confusing as they are not broken down into primitives but rather expect you to know the myriad of icons. This is not just confusing for viewers, it actually removes effective strategy as there is high computational overhead for players to understand the gamestate. This is like chess with 16 different types of pieces: its a lot slower and ****tier to play, and the quality of play is severely diminished.
In eg. Hearthstone, unit abilities are broken into primitives like Divine Shield which take a lot of complexity away. Even when the ability is eg. a death rattle (ie. text ability), the icon clues you in about what that means. This makes it very hard for a casual audience / to grow the game. I personally am willing to read through 200 cards to know what I'm watching, but other people are not. A good analogy would be trying to watch live MTG broadcasts when you don't know the cards. It sucks. In a modern design, games should be similar to Hearthstone in that you can effectively parse the content. Though, Artifact *is* based on DOTA :eyeroll:
Second, Artifact's initiative system is terrible. Very specifically, the fact that you cannot play spells in lanes where your hero has died creates tremendous variance in the game, and warps the game towards bombs. I think right now people are like "hurr durr, there is so much strategy created! passing and letting your guy die is strategerery!!!111oneoneeleven", but time will prove that this is actually terrible gameplay. You shouldn't have a hand of 10 cards and be passing over and over in your game. You shouldn't be locked out of playing a card. This reminds me of classic Pokemon TCG, where the best deck when the game came out is a deck that killed your starting pokemon on turn 1, since you weren't technically allowed to replace it in time, so therefore you lose.
Third, the tri-lane mechanic is a bust. Best 2 of 3 is a decent mechanic, but the reality of the situation is that lanes are often forfeited. This creates a solitaire effect, where eg. I am attacking for 32 on an empty lane into a 38 tower, and you are attacking for 21 into an 80 tower or whatever. With the high numbers involved (low granularity), this does not feel like a race, it feels like a coinflip - you just so happened to be lucky to deal 38 instead of 37 on that turn, so you won. One subtle reason that should be pointed out as to why lanes are forfeited, is due to the snowball-ey nature of combat and the ****ty initiative system. If for example blue has weak heroes, then when you have a 7/11 axe hero out, I don't want to play my bull**** hero into that lane where it will promptly get 1 shotted and I don't retain my ability to counter play in that lane, even if I have better spells that want to play there.
Fourth, draft is very high variance due to the dependence on strong heroes. This is much more variance than just "opening a bomb in MTG (limited)". If for example you open Zeus, you now get 3 flamestrikes too.
===
Parting words, I expect that this game will get propped up for a long time due to the money investment by Valve, and their willingness to just buy the entire pro/streaming scene. Probably I will end up playing it. Though, I really wish there was a good competitive card game that I can invest in. If someone doesn't make one within 2 years, I may have to start making one.