Drafto - Core
I'm still working on Drafto.
Playtesting over the last couple of weeks has honed the game into something everyone seems eager to keep playing. I think the rules are more or less locked down, now - it's just the individual objectives that require tweaking. Once this is done, I'm going to write everything up and release it under a Creative Commons non-commercial license, allowing people to freely reproduce and modify the game (as long as you credit me, and don't sell it).
It's good. I look forward to sharing it with you.
Way back in the...two weeks ago...I discussed a set of pillars proposed by Paul Sottosanti that he saw as important mechanics drafting games should possess. Some of these I agreed with, and others I felt were more optional.
Here is my counter-set, which for lack of a better descriptor I'm going to refer to as the core of drafting games - central mechanics which I feel are important enough that all good drafting games should possess them.
Some cards should be clearly more valuable than others. This is to do with the size of the decision set. If all cards are of roughly equal value, then the player has a number of choices to consider equal to the size of the pack - which is problematic unless packs are very small. Conversely, if cards have clearly different valuations (up to a point), the player can ignore all of the weaker cards present and have a much smaller set of "true" decisions to choose from. This is what occurs in Magic. Players don't start with any external direction, so the first pick of the first pack is ostensibly a choice from 14 unique cards. However, each pack is stratified into 10 common cards, 3 uncommon cards, and 1 rare card - the higher rarity usually correlates with card power, and this allows an inexperienced drafter to ignore most of the 10 commons and face a much easier choice of between 4-6 cards. (Veteran drafters ignore rarity altogether and will be able to perform a similar pruning operation on the decision set based entirely on learnt card evaluation / experience.) As it's important for players to understand and enjoy your game before they become proficient in it, I feel this overt strength signalling is critical. That, or tiny packs :)
Cards should have different value to different players - and each player should be able to approximate the valuation functions of his or her opponents. This was on the original list of pillars (albeit in a simpler form), and it's a property I strongly agree with - in my opinion, this is the whole point of drafting games. If every player values all cards the same at any given point in the game, then you lose all the subtlety and nuance of the genre which you would otherwise get for free. My addition is to specify that each player should have the capability to be at least partially aware of the valuation functions of the other players (by "valuation function", I mean the criteria that player is using to judge how valuable a particular card is to them if they were to pick it). This will usually be a consequence of the weaker condition, as cards valuable to one player will be missing from packs they pass, allowing that player's valuation function to be inferred by their absence (this is known as cutting a particular subset of cards). However, I felt it was worth stating explicitly, as other mechanisms could be introduced to strengthen this property. For example, consider the strategy known as wheeling or tabling (as in "to wheel" or "to table" something). Imagine if, given a pack, you're presented with a choice between two cards that you'd ideally like to draft both of. If all players value all cards equally, then whichever one you don't take, the next player will - so you take the one with the highest immediate value (which isn't a real decision at all, because it was strictly better than the alternative). If all players don't value all cards equally, then the correct choice suddenly becomes less obvious - whichever card you don't take won't automatically be taken by the next player (or indeed any of the other players), making the decision process involve some non-obvious function of how valuable the guaranteed card would be to you now and how likely it is to make it all the way around the table so you can pick it in the future (based on your beliefs about the valuation functions of the other players). Another strategy this enables is hate-drafting, which is when a player takes a card that is not particularly valuable for them, but is something that they believe is of significant value to an opponent. As the players are in direct competition, this can often (but not always) be a correct action - the accuracy of the player's beliefs about the valuation functions of their opponents is a key component of this, which makes deducing these a skill intensive part of the emergent gameplay.
Some cards should have unknown value. Let's be honest - expecting a new player to immediately grasp the intricacies of forming beliefs about other players' valuation functions is a bit of a tall order. While these are cool parts of drafting that reward invested players, you still need enough of a game to be present for players who are unaware of this, else they won't stick around long enough to find out. For these players, drafting becomes more of a puzzle than a game, as their opponents are irrelevant. To introduce meaningful decisions into this puzzle, the player must be presented with several attractive choices at each step, and it must not be obvious which (if any) is the best of these choices. To use Magic as an example (again) - assume a player is completely accurate at assessing how powerful a card is, apart from its cost. The cost of a card affects how easy it is to play with other cards that have been drafted - for example, if you've drafted mostly green and white cards, picking a card with a heavy black requirement is unlikely to be a good move, even if the card itself is very potent. The kicker is that this doesn't just depend on the cards you've already drafted, it also depends on the cards you have yet to draft. The true value of the card won't be known until after you've finished drafting! In a more general sense, if you know that when you pick a card it's worth a fixed amount, the game devolves into calculating this fixed amount for all of the cards in the current pack, and picking a maximum. If the value of a card is (partially) unknown, then the decision incorporates all the other information you've been exposed to so far in the game - i.e. "will I get enough other cards to make this card work based on what's already been passed to me / what I've passed"? This doesn't need to apply to all cards in the game - this mechanic adds complexity, so if you want a more accessible game it's worth using this sparingly - but it is a good way of adding clearly signalled interesting decisions to a game.
I think that's it. Everything else I could think of seemed to be overly specific, optional or adjustable (e.g. visible objectives, starting asymmetry, amount of complexity).
It's worth noting that Drafto v1.0 satisfied all of these (but was still terrible), as does Drafto v2.0 (but is much better). So too do Magic and 7 Wonders, though in these cases I don't think they do a particularly good job of clearly signalling card value (point 1) without a reasonable amount of prior knowledge about the resulting gameplay - this doesn't mean I think they're bad games, it's just that the drafting portions of them perhaps don't work as well as they could do because they have this other, more complex game attached.
Anyway, let me know what you think about the above reduction - feedback is important, and appreciated :)