The need to explain inspired me to write a blog post! So that's productive.
TL;DR:
New players tend to focus on the rules and their personal strategy, and don't pay much attention to their opponents. Therefore, their opponents' actions that affect them directly need to be easy to understand, or they won't react to them. So a useful design heuristic when trying to make games friendly to new players is that the mechanics that govern interactions between players should be simple (whereas non-interactive mechanics can be as complex as you like since they are devoting most of their attention to those anyway).
Re-posted:
Interactions should be simple
I remember my feeling of frustration when I first played the Game of Thrones LCG. My opponent and I used preconstructed decks that he had prepared, and I found myself stopping the game each time he played a card to pick it up and read it. Throughout the game, I kept finding myself tripped up by effects on my opponent’s cards that I had forgotten about or hadn’t realized all the implications. Just keeping track of my own cards was difficult enough – reacting properly to my opponent’s strategy was more than I could handle.
Attention is a limited resource that players tend to spend on themselves, especially when they are new to a game. Veteran players have dozens of heuristics built up to automate much of their strategy, but new players have to balance an often overwhelming set of rules with the question of what they should be doing. Keeping track of their opponents’ moves on top of that is too much, and is often ignored. If you want players, especially new players, to pay attention and react to what their opponents are doing, then the mechanics governing the relevant interactions need to be as simple as possible.
Most strategy card games (like the Game of Thrones LCG) do a terrible job of this. The problem with card-based games is that reading and understanding the text of each new card takes a surprising amount of effort. It is also difficult to read cards upside down, so picking up or reorienting the card adds further time and effort. Once players have played the game many times they can recognize cards on sight without having to read any text, but until that point, players may not be willing to dedicate attention to their opponents’ cards when their own cards already provide enough of a challenge.
An example of a game that handles interactions well is Galaxy Trucker, in which players build spaceships out of tiles and evaluate how well they built their ships against a deck of encounter cards. Player interaction is mostly indirect, with some encounter cards comparing different ships to determine which ship has the most or least of a certain attribute. In each case, it is easy to estimate how your own ship compares against the competition by looking for the relevant tiles on the other players’ ships. All the more complex mechanics in the game are non-interactive; it doesn’t matter how another player deals with intruders or meteors because it doesn’t affect how you handle them.
The trading game Sidereal Confluence has very strong player interactions, since the game requires players to trade with each other due to the weaknesses of each faction. Practically anything can be traded – resources, (binding) promises, and even sectors of a player’s economy, represented by “converter” cards that transmute resources. However, new players only pay attention to the raw resources that other players have, not to more complex options such as converters. When converters are traded, it is usually at the behest of the player that owns the converter, not the player receiving it.
One way to help players interact with each other is to find a way to highlight actions that matter to other players. Games with a strong spatial element are naturally good at this, because proximity to an opponent catches a player’s attention. In Catan, if somebody is building a road that threatens to cut you off, you notice because the road approaches your own road. In Carcassonne, if somebody is trying to get into your city, you notice because they place their meeples and tiles aggressively close to yours. Because players can ignore their opponents’ actions outside the area that they care about, they can devote enough attention to those actions to respond. In games where threats (and opportunities) do not exist at a concrete location, it may be necessary to find other ways to indicate which of an opponent’s actions are relevant to other players and which can be ignored.
With experience, players internalize enough of a game’s rules and heuristics to spend more of their attention on their opponents. But if they are frustrated by a game – as I was with my Game of Thrones experience – they may not play long enough to get to that point. So if you find that your players are ignoring each other or complaining about being overwhelmed, a good first step is to simplify their interactions.
Here are the ones I can put into words/think of off the top of my head (some of these are pretty common advice, and I don't claim to have come up with all of them)
+ Each choice should benefit a player in two distinct ways
+ Random events should be foreshadowed (preferably in proportion to their impact)
+ In multiplayer games, if one player targets another with a negative effect, the victim should benefit in some way (a silver lining)
+ In multiplayer games, if one player targets another with a negative effect, the perpetrator should gain some unique benefit
+ Avoid in-game text as much as possible
+ Randomization should determine available choices, not the outcomes of a choice
+ Use at most one method of randomization per game (dice, cards, etc.)
+ Break up long procedures with choices
+ Avoid symmetrical resources
+ Replace math with tables
+ Minimize or eliminate in-game shuffling
+ Players should get stronger in some way over the course of a game (progression)
Interesting - could you start a new thread to elaborate? This is similar to the heuristic of choices providing multiple benefits, but I would describe those benefits as orthogonal, not diametrically opposing (and it is more about benefits than motivations).
Also interesting to discuss (in another thread). I can see some of the utility, but would like to hear more about your views on how to do this (and examples, if you have any)