Disclaimer: Wondering why so many early reviews have more hours than is possible on release? That's because a lot of copies were given out privately, probably via Discord, so that they could be playtesters for months and queue into players of the demo. Apparently the game has a Patreon per another review, so perhaps these are pre-paid players? Also, I was a demo player for a few weeks or so, playing heavily during that time. And I'm more than a little biased, since Grail heavily resembles a game I've been developing on my own over the last few years but lack the polish to release on my own, which is why I've sunk a lot of time into this game (~150 hours in the demo.) --- Grail is a game trying to appeal to several audiences at the same time, audiences with very different motivations, and while I think it fails at this for some, its still the first successful "deckbuilder autobattler" I'm aware of. Here's the basic rules: * Start with 4 energy and at the end of each turn you take involving a card draw, gain 1 energy. * Most cards cost energy, shown by the pips on the bottom of the card. On your turn you draw a card, put your energy into the pips, and then play the card if the pips are full (spending them) or discarding the card if they're not. If your deck is empty, you instead spend 2 turns shuffling your discards back into the deck. You do not control the order in which your cards are drawn or played unless you have gone out of your way to rig draw order via "stickers" that force a specific order. * Each card can have one sticker. These do things like make a card critical (x2 to most card effects) more often, play twice, return once per shuffle to the bottom of your deck, and so on. * Most cards can freely be sold at any time outside of battle, and you even gain gold for doing so. This heavily rewards running a thinner focused deck, assuming its strong enough to mitigate the shuffle penalty. There are incentives to build a wider deck, but they require relatively more effort and usually not for a stronger overall payoff barring certain highrolls. * Status effects consist of Sleep, Burn, Shock, Poison, and Bleed. Sleep makes the target not attempt to draw for their turn, though they still gain energy. Burn makes the target take damage when they play/discard cards from their deck - importantly, shuffle turns evade this. Shock makes the target take damage when their opponent creates energy, including end of turn energy and individually for effects generating multiple energy at once. Poison makes the target take damage when they shuffle, and tends to come in larger quantities due to being harder to trigger. Bleed makes the target take damage when they take damage from other sources, including the above statuses. Those are the basics, though there's a metagame to discover that I don't want to fully spoil. But speaking of spoilers, let me get into the issues I have with Grail, which aren't deal breakers for me but are very annoying: * There are hidden events for the player to discover with hidden conditions that the game won't tell you about. But don't worry, most of the invested playerbase this early already knows them all due to Discord discussions, datamining, third-party wiki, etc! Because I respect parity in PvP I haven't looked these up myself, which means I frequently get to queue into players building around these hidden event payoffs despite not knowing many of them myself and despite being an experienced top player by winrate during the demo! What I can say is that the events I know about involve having one or more specific cards in your deck to give the event a chance to show up, or in some cases the event hiding within another event - like the Legacy event hiding within the Fencer location, which the tooltip won't tell you about even if you're holding a Vertical Cut or Vital Strike to trigger it. * Wondering why so many players are well ahead of you on the leaderboard? Turns out ranking points mostly carry over from season to season, with the exception of Deity point climbing, so anyone playing the demo character or who was a playtester pre-release has a huge head start on you, the new paying customer. Wondering why they're wearing cosmetics that you currently don't see in the shop? While the developer has sent mixed messages on this, they advertised during the demo period that demo cosmetics would be exclusive. It seems very strange to me to inflict this sort of FOMO (see also that cosmetics are behind a gacha that will give dupes, though that currency can slowly manually buy cosmetics you don't have). UPDATE: Post-launch the developer has made buying characters cost 100 instead of 200 trophies and sent out a boost to demo players who might have spent their starting currency on cosmetics, so the grind to play one of the more complex characters is a lot shorter than it was on launch day. * While there is an "endless" and "continue run" as options, these seem to be meant more for a focus on what passes for a story and "good/bad ending" sort of thing. If you're wondering what the unexplained in-game mechanic called Grace that itself is often hidden behind events is, know that its related to this stuff and that if you don't care about whatever lore exists, you can freely ignore all of this. Click Finish Run on each win, and ignore any events offering you Grace other than making a mental note of which events trigger a Grace interaction. * The in-game compendium only shows cards the player has seen or drafted themselves, since apparently this is meant to be a huge achievement chase (win the game with fully upgraded copies of every card in the game!) As a result, the learning curve is screwed up since you can't reference cards you're unfamiliar with, and you won't even be sure of the base values of enemy cards since a mechanic called Augmenting exists. Augmenting boosts the base stat(s) of a card and looks otherwise identical to the base card, and can be in randomized amounts, seen both in specific shops and if the player themselves finds ways to grant permanent stat boosts during events, with traits, or with battle effects. * Matchmaking is done in a way to make the player feel better about themselves rather than in a way with competitive balance in mind. You can't face opponents that outrank you, but you will frequently face opponents below your rank and can face premade weak bot boards at lower ranks. Yes, this means the "play once, face another player on defense once" system commonly used in auto battlers isn't present here and new player boards are over-represented. In a scenario where both players die simultaneously, which is possible to aim for thanks to escalating fatigue damage, the active player always wins. All this put together means players will average over 50% winrate regardless of rank, though this game involves skill such that results will still vary significantly based on skill level. * There are third-party tools that give a significant edge to their user over a player entering the game blind. I don't engage with Discord as a platform, but many of the invested players including the playtesters who have a head start are likely there, and I've been told that a lot of obscure mechanics are discussed there. Too bad Discord is a private platform, its server content not found by a search engine. There's also a wiki trying to document all the cards and hidden events, and a website that pulls from the game via API a bunch of data from players. If someone wants to look up top players and see what their play strategies are for netdecking or anti-meta purposes, they can! For an auto battler where the genre appeal for me is meta discovery, I greatly dislike that tools like this exist to accelerate the process, and I hate that players unaware of these tools have to play at a disadvantage.