It has been said that creativity, and art, come from restrictions, from what can't be done, that it's the obstacles that create solutions. Certainly a game which allowed you to do anything would be pretty quick and not bundles of fun. If you accept this, it might not be surprising to you that a way to breathe some extra life, playability and replayability into your very favorite computer games is to throw in a just couple more restrictions on what you can do when playing them - extra rules that the manufacturer didn't include. For example what units you can build when, whether you can attack first, etc. If you like Classic Gaming, retrocomputing or abandonware you may already be improving your gameplay with sets of restrictions.
I'm not talking about just any extra rules or restrictions: choose shrewdly, and one or two extra rules can either let you skip the most boring or irritating or predictable parts of the game, or let you skip known game design problems and produce a more enjoyable game experience. If you're bored with a game, changing the rules just enough to make your decisions more difficult, and therefore more interesting might be just the ticket to extended game play with a favourite game.
Call it game liberation - you don't have to stick with the manufacturer's idea of how the game should be played, and if you're a very experienced player, you may well be able to figure out a better way without much trouble. You've probably played the game longer than it took to design and publish it.
After all there's a delicate balance or two that has to be maintained to keep a game replayable, with continued interest. For one thing, it's no fun if the very experienced player can quickly tell whether they will win or lose, before the game's barely begun - that was my experience with the first published version of the ancient game of Empire, a 4X ancestor - but note that that sort of predictability is something that doesn't happen with Canfield or Freecell, say. I say, it's avoidable. Also it's no fun if there's a hole in the game design that makes victory too easy or straightforward, time after time. But game designers have to release a game before they've played it for years, so they can't find all the things that might make the game boring five or six years into playing - or even one or two years in.
You can keep your favorite game playable by allowing yourself to modify it just a bit, not by getting ahold of the source code but by placing simple restrictions on what you can do and not do - just your own personal restrictions that are over and above what the program doesn't allow. Carefully chosen, such restrictions can transform a game, and keep it alive for years longer.
Sometimes such modifications are obvious - using weapon X makes the game too easy, so I'll just never use that weapon. Other restrictions will be more subtle - for example, in the original MOO, not allowing yourself to build any small scouts (other than the two you get at the start). This prevents you from parking scouts on every star in reach at the beginning, thus seizing so much territory so easily that you'll be too likely to win, and will know that fact too soon to take much interest in the game.
I do find that there are three themes in the restrictions I make: 1) play balance 2) avoiding predictability and 2) avoiding tediousness micromanagement. I actually think these are the three foundations of good game design. However, I'm willing to admit that 1) and 2) do overlap somewhat.
As a bit of an aside, someday I'll publish an article discussing ways to improve game designs to avoid predictability and maintain play balance that aren't being used right now, perhaps under the title "Competence-reward vs predictability in game design". and analyze the tension between those two. Note that Freecell manages pretty well to reward competence without being all that predictable, even in outcome, and that it does so without dazzling AI.
I'll grant that these sorts of restrictions are more necessary with solitaire games where you play against computer opponents, but sometimes human players adopt such restrictions for online play against each other, such as in "no air" games of Starcraft.
Also, note that restrictions are usually applied at the second highest level of difficulty a game allows, since the highest level is often A) all but impossible under the best of circumstances and/or B) alters the game out of recognition - for example so that only strategies so risky that are just simple die rolls or really odd strategies can possibly win.
Someday, manufacturers will build in a continuum of difficulty levels, not just three or four kludge ones. Ironically, older games sometimes did this. Then kludges like restrictions would be less necessary.
As an example of what I mean by a set of restrictions for a game, look at Restrictions for Master of Orion, where I've detailed both a particular set of restrictions for an abandonware game and detailed the precise reasons why each of these restrictions improve game play.
As a quick example from a more recent, although slightly obscure game - Baseball Mogul - within a week of getting Baseball Mogul, I had to introduce a restriction, already, otherwise even its highest difficulty level was too easy. Namely, I forbade myself from selling players during the few days after the season and before free agency applied. Otherwise, other teams would pay extraordinarily steep prices, in trades, for these players who had no time on the meter, and usually do so simply for the privilege of being stiffed by them in contract talks. (Whereupon I could rehire them as free agents, if I wished, later in the week.) That's just a hole in the AI, allowing me to make "Dutch Book" (at least figuratively) against my computer opponents, but there you go. Maybe the privilege of being the only team able to talk contracts with a player for a short time is overvalued by MLB teams now in fact, but not by that much. (Now I consider carefully whether to trade off players in their last year, and even then I get too much back in trades.)
Looking at game restrictions that have slowly been evolved by the most experienced players of a given game can be very educational for game designers and wannabe game designers, as these additional rules added into the game are usually shrewd choices made for reasons that the experienced player can usually articulate, and the game has usually been thoroughly tested by those players both with and without those rules.
With more modern games that allow greater player choice, positive additions to the game, say in the starting conditions, can also be made. This would be more analagous to the common rule change to Monopoly in which taxes and penalty money are accumulated in the center of the board and awarded to whoever lands on the Free Parking space. This rule change is only the latest modification by players over some decades to the original, more boringly educational game which was designed and later patented (1904) by a woman named Lizzie J. Magie who wished to promulgate the theories of the famous populist American economist Henry George in an entertaining way. (According to the PBS-TV show “History Detectives”.)
It was only after decades of modification by players who made their own sets and added their own restrictions and additions to the game's rules that a much later and much revised version of what was originally been called "The Landlord's Game" was published by Parker Brothers in 1939. (Only quite early versions of Monopoly cite all relevant patent numbers on the board; which obscures this fact, however.) By that time, what had begun as a criticism of contemporary capitalism could be mistaken for a celebration of it - but more obviously, it had become a highly entertaining game which was highly replayable. Obviously, modifications by users can work great success.