I was pointed to this by a facebook post and found it interesting enough that I had to share it.
While it mostly references board games, a lot of the points could easily be applied to RPGs.
A list of some of the best advice for anyone creating a game in any form. This was originally written by your friends over at Board Game Designers Forum (http://www.bgdf.com, all original credit will go to them) and we think people would do very well to follow these few "Best Practice" rules closely. Please add any additions to the thread about your own personal experiences and best practice principles so that we can build an industry wide important resource together for both budding and professional Game Designers.
Game Design Principles
Do not add a rule to take care of an unusual situation. In almost every case, the game can be subtly changed to prevent the situation from occurring. Each added rule, no matter how uncommonly it's required, adds complexity that makes your game harder to learn and, potentially, to play.
Your scoring design is your game design (J. Degann) Although designers tend to emphasize creation of innovative or clever mechanics, the true motivator of player decisions will ultimately be the game's scoring system rewards. The interesting decisions that the mechanics promote will only be interesting to the player if the scoring system encourages them to be.
Avoid false strategy(S. Appelcline) Avoid situations wherein players are required to make decisions that have no significant impact on the outcome of the game. Players will generally assume that games only present choices that are consequential and worth contemplating carefully, so present only decisions that matter. This principle will help shorten a game's playtime while simultaneously helping to keep players more engaged for the duration.
Balance with incentives and costs, not with restrictions It's common during playtesting to find several imbalances in a game. Perhaps when a certain situation comes up, a player is unduly rewarded or punished, or perhaps a player finds one particular strategic path that is considerably more successful than alternatives. It is possible to add rules that prevent unbalanced situations from arising. It is, however, much more satisfying to encourage players to behave in the way you want by modifying the rewards that you dangle in front of them, and by modifying the costs (both resource and opportunity costs) of the things you want to enable or prevent. A classic example is the "hand limit." Many games restrict the number of cards that players are allowed to hold, presumably because having too many cards gives players some overwhelming benefit. But a game's rules can discourage large hands in other ways — by driving up costs to acquire more and more cards, or providing some reward for a small hand. At the level of rules simplicity, stating "you can't have more than X cards" may be simpler and more effective in the long run, but as a principle, more interesting gameplay results from encouraging the "right" behavior organically (through so-called "natural limits" of your incentive and cost structures) rather than forcing that behavior by explicit rule.
Pay attention to the rules people forget Introduce some new people to the game, but don't give them the instructions. Instead, explain the game and start playing. Is there a rule that they frequently forget? Do you find yourself saying "no, you can't do that because of this..." or "don't forget about..." If so, consider either eliminating that rule or re-framing it so that players don't forget it. This is a particularly important principle if you plan on ever giving a demo at a trade show. Nothing will send a buyer walking away like saying "no, you can't do that because of this rule you didn't remember."
Simplify When you want to add rules, refrain from doing so. Even if it's so clever you just can't stand it. Often, a later design will prove to be a happy home for that neat idea you thought you couldn't live without in your current design.
Removal is OK One of the hardest tasks for a game designer (and me particularly) is knowing when to remove parts of a game design. Not because I do not notice the need to remove, but the issue of wanting to keep my rules/mechanics unchanged. BUT REALLY, removing things from a design is OK! You can always add it back later, if needed.
Theme and art are very important All else being equal, a game with an appealing theme and art will sell better and get played more. If the theme and art don't engender interest, your game will have a much harder time finding and retaining an audience. (eg: "I don't like space games")
Understand how mathematics and probabilities affect game play Spend some time learning and refining how your game reacts to probabilities: linear, bell curves and other probability distributions. You need to know the difference between a mechanism that could come up '1' twenty times in a row (die) and one that cannot (deck). Rolling a d20 is different from 2d10, or 2d8+2, or drawing from a deck of cards numbered from 1 to 20, or players simply "choosing a number". Each method offers different results and therefore different game play.
Try to make your mechanics reflect the theme What are players doing in the game? Driving a herd of cattle? Tending to mayoral duties? Commanding an army? Mechanics should reflect and be associated with the game's theme. There isn't a lot of this out there in the gaming world, but when it happens, it makes designers proud and gamers happy. When you can incorporate a mechanic that reflects something you would actually be doing if you were the actual person/thing you are portraying in a game, by all means do so. As a corollary, use the theme to suggest interesting or novel mechanics; base the mechanic on the kind of decisions a person in that situation might face, and interesting choices may suggest themselves to you.
Design Process Principles
Never assume players will make smart decisions. While you need not ensure that a player who makes poor decisions has a chance to win, you do need to make certain that your game doesn't break down when one or more players is playing badly. It's fair to assume that players will play to try to win (as otherwise, most any game ends up "broken," in a sense), but don't assume they'll be good at it.
Failures are successes too No matter how horrible a game design might be, chance are, you have learned something new! So learn from those bad or horrible designs! Dont kick yourself for it, use it to make the next design better!
Half of building should be breaking Games are unique structures that are subject to unusual stresses when used. Before releasing a game into the wild, try to break the current system by finding a systematic exploitation of the rules that assures either a guaranteed win, or that the victory conditions never arrive. This is game breaking, and a game must be able to be unbreakable by any one player. Playtesting with people who are "rules lawyers" or are expert at exploiting loopholes in the rules can especially help with this process.
Include a "fair" setup Many games feature a player-controlled setup phase (e.g. Settlers of Catan), which can be a good way to introduce variability into each playing as well as to promote strategic development of a player's position, and differentiation of the players' strategies. However, asking players to make setup decisions before the game starts also adds a learning curve to the game. First-time players just want to get a feel for how the different mechanics of the game work; they will not yet have a basis for knowing what a "good" and "bad" initial setup strategy will be. It is a good idea to include a "fair" setup, one that has been tested and that gives each player position equal opportunity to do well in the game.
Graphics are meaningful When working on the graphics for a game (even for the earliest of the prototypes), remember that in addition to adding flavor to the game, and reinforcing the game theme, graphics can really make or break a game. If used intelligently, icons, colors, etc. will help the players understand how the game works, and make it more fun, allowing the players to focus on playing the game. Using graphics just as an adornment will most likely result in a harder to play game.
Know for whom you're designing the game If designing for a family audience, you may want your theme to cater to a "lowest common denominator" within such a group. Gamers tend to have more specified interests and tastes amd ss such, the more unique themes will tend to appeal to more specialised niches of established Gamers.
I've looked at many of these points before, but it's nice to see other people thinking on the same wavelength.