There are a lot of good games out there with interesting and playable skill systems, and it is rare that I’ll decide against a game based on that alone (though there are exceptions). I find the Advantages and Disadvantages to be much more telling, whether their called that, Merits & Flaws, Edges & Hindrances or Lord knows what else. In many games, a lot of the real meat of things is hidden among these.
It’s a model that makes a lot of sense: if you put your crunchiest rules in your ads/disads section then you reduce the overall crunch of the game in play, since players are effectively selecting which rules they’ll play with, and the ones which aren’t chosen can be set aside. Done consciously, this can be a fantastic model. For example, if all of your advanced firearms rules are actually put into your firearms ads/disads , then your gunbunny can use the advanced rules while everyone else can get along without worrying about them.
Unfortunately, this is rarely done consciously, at least not across the board. There are exceptions – usually in the form of fighting styles – but most often the ads/disads list tends to look pretty similar from game to game, usually with the difference emerging in levels of detail. Sometimes there will be embarrassments, like the advantage everyone must take to be viable in play, but it’s usually not broken. It’s just boring.
The trick comes when you start trying to map these advantages and disadvantages to fiction, which is to say to the kinds of characters that players are thinking of. There’s a pretty profound disconnect between what ads/disads are attempting to model and how they model it. The simple difference is this: If a character in fiction has some sort of advantage that’s worth noting, then whatever it is it’s turned up to 11. If one guy in the group is particularly tough, he’s TOUGH. If one’s a tracker then he can tell you what you ate by looking at your footprints.  If you’ve got money, you’re rich enough to build a batcave. If he has visions of the future, that’s a BIG FREAKING DEAL.
In fiction, these things are a 24 ounce porterhouse, but most RPG systems send you to the salad bar with a tiny plate. It just doesn’t work out the way you’d hope.
One of the best tricks I ever saw in a game was in Adventure!, with the inclusion of 6 point merits. Most merits in the system were rated like stats, from 1-5, but if you bought one up to 5, you could buy the 6th point, at which point the merit would transform into what you had always imagined it to be. Wealth became unimaginable wealth. A man of mystery became a true enigma. You flipped a switch and turned on the awesome.
On some level, I’m not sure why to bother with any ads/disads except those 6 pointers, so to speak. I mean, sure, in some games it might be interesting to distinguish between one character having a middle class income and another having an upper middle class income, but I think we’re all smart enough to make that distinction when appropriate. But for high adventure, high octane stuff? The difference between level 1 and level 2 toughness just doesn’t cut it – I want to be tough as nails or not bother.
Now, as interesting as it would be to see a game designed from the ground up with this sort of thinking, that’s simply not practical for most folks out there. We have our existing games, and we want to make them work, and the good news is that this is pretty easy to do with any game that has advantages and disadvantages.
First off, look though the list for character-defining advantages. A lot of systems have these, and they’re easy to spot because they tend to be the most expensive options, and they very clearly define the character in some way, like making him non-human. Look at how much that costs, and set that as your ballpark for about how much an advantage should cost.
Once you’ve done that, start building bundles. To make a bundle, you start with an and you take all the advantages that seem to match that idea and combine them together that totals up to about your goal number of points. 
For Example: For a hypothetical game, I’m looking at making an approximately 20 point bundle for “The Tough Guy”. The game has a Toughness advantage (3 points per level), a High Pain Tolerance advantage (5 points) and a Quick healing advantage (4 or 8 Points). I make a bundle out of Level 4 toughness, Level 1 Quick Healing, and High Pain Tolerance. Total cost is 21 points, which is close enough to my ballpark. I now write that up as a 21 point advantage with all the mechanical benefits of the advantages I used to make it.
Depending on how the game handles disadvantages, you could even fold those into your bundles. If you can do this, then it’s fantastic – if a bit more abusable – and it’s a great way to make sure that the “Ninja” advantage is badass, but comes with built in enemies. It means that if the disads are strongly tied to the central concept, then they’re more likely to be remembered, and they’re more likely to matter when they come up.
Once you start on these, they tend to be very easy to pull together quickly if your players express an interest in one concept or another. And once you have enough of them under your belt, you can pretty much remove the entire ads/disads table from player sight, except for purposes of building new bundles.
Now, the last thing to consider here is pricing. Up to this point it’s all been a bit of sleight of hand, changing nothing about the underlying game. But the bundles are going to be pricey, and in some games that could mean really gimping a character on skills or in other areas, which kind of sucks. The simplest way to address this is to offer discount pricing on bundles or to offer more points, but while both of those solutions are simple, they are just treating symptoms of the problem. The bottom line is this: if you’re doing bundles, it’s because you want players to buy them because you like the fact that it puts a strong concept front and center. If that is the case, then just give each player one. Dock them a few points if you feel you need to, but let them otherwise just spend as they see fit. If they buy a second bundle, then that’s great, they’ll do so with full awareness of the tradeoffs, but without feeling shoehorned into it.
Anyway, this is something that will add a bit to your prep for your next game, but can be used with anything from Savage Worlds to nWoD to Cortex to make characters whose concepts are worn a little bit more on their sleeve (assuming you think that’s a good thing).
1 – Yeah, there’s some bleed between ads/disads and skills depending on the system. Honestly, the same logic applies to skills, but that is more the domain of competence porn.
2 – The one other exception is when we’re talking about a character’s goal. If it is a character’s goal to become super-rich or incredibly well connected or anything else like that, then it might be appropriate for them to have some stunted version of the full advantage because that is guaranteed to change over the course of play (or at least it should be). In fact, having a goal that has this kind of concrete game-manifestation can work out very well because there’s no confusion or miscommunication regarding what the end-goal looks like.
3 – Fans of Eden’s Buffy the Vampire Slayer may recognize that this is basically how that game handles a lot of supernatural powers.
4 – Realistically I’d probably round it off to 20 points, but that’s just because I like round numbers.
5 – And here we hit the rub. This is not going to be hugely popular with your min-maxers because they will have put a lot of thought into the specific combinations they want to build. It would be easy to dismiss those concerns as mere twinkery, but don’t be too hasty. Use this energy to your advantage, and get the min-maxers help in building bundles. He may look at your “Tough Guy” bundle and point out a way you could make a guy much tougher for those 20 points, and if so, then make the change. After all, the goal is not to penalize players, it’s to make these concepts pop much more strongly.
6 – Yes, you should be considerate if someone buys a bundle that overlaps with another bundle in such a way that some points are wasted. Just redefine one of the bundles to remove the overlap, and carry on. This will save you infinitely more headaches than giving back points. If there’s a lot of overlap, consider that a red flag of a muddy concept or a Ninja/Special Forces problem.
7 – The Ninja/Special Forces problem is something that comes up in many games with player-created attributes. If you are playing a game like Over the Edge and you have one character with “Ninja, 4d” and another with “Special Forces, 4d” then you are goign to have precious little mechanical differentiation between them. They’ll roll the same dice for most of the same sort of activities, so the difference is almost entirely one of color. And in that case, the color becomes INCREDIBLY important to keeping the characters from becoming dull. (and yes, it’s a footnote on a footnote. Sue me.)