?I used to do drugs. I still do, but I used to, too.? ? Mitch Hedberg
Dope Wars (or Drug Wars) is a text-based computer game that?s been around for about thirty years, rewritten countless times for every platform from DOS to graphing calculators to iPhones. You play an independent drug dealer in a big city (usually New York) who travels around buying and selling various illegal drugs. The market prices vary widely from day to day and from one neighborhood to the next, so you can make large trading profits if you watch them closely, but you have to keep an eye out for the police, muggers and other business risks.
It wouldn?t have been around this long if it wasn?t fun and addictive ? in fact, it incorporates a lot of the elements from my ?What Makes Games Addictive?? post ? but there are some basic improvements to the gameplay that (as far as I know) have never been made. Instead, every major redesign has tried to change it into something that it?s not. Zynga?s short-lived version tried to turn it into a typically scammy ?social? Zynga game, a kind of R-rated Farmville. No surprise that didn?t work. Others have tried to change the subject to something more innocuous than drugs. But none of the ones that I?ve seen have addressed the core gameplay.
If you?ve never played the game, the rest of this post won?t make much sense to you, so you should take a break and go try it out. Amazingly there doesn?t seem to be a decent web-based version anymore, but if you have an Android phone, this and this are both free and pretty good. For iOS, where the censors won?t allow the original merchandise, there?s this one (free) with prescription drugs, and this one ($1) with zombies.
OK, had a chance to try it out? Here are some of the problems I see, and how they could be fixed:
1. Too much randomness. Some randomness is always good, but too much gets frustrating. It?s realistic that muggings and police stings happen on a fairly random basis, but should you really have to wait for the right roll of the dice to buy a new coat with more pockets? Why not have a ?shop? that sells equipment in a fixed location, just like the bank and the loan shark?
2. Too much spread in prices for different drugs. As your bankroll grows, it quickly makes sense to deal only in heroin and cocaine, which means that the prices and ?events? for all the other drugs are irrelevant. I get that you can carry a lot more dollar value in cocaine than in Quaaludes, but this is a case where realism should be sacrificed for better gameplay.
3. Too easy to compound money. The bank is completely safe and pays extremely high interest rates, so very often the optimal strategy is to leave most of your money there, especially when you reach the point where you can only carry a small % of your net worth in merchandise anyway. The interest rate should be reduced or eliminated, and your account should occasionally be subject to ?freezing? and confiscation. Meanwhile, there should be more ways to reinvest your money in the business. You can only have so many ?pockets? I guess, but after that why not buy a car, or a van? Or recruit employees?
I think there?s a real opportunity for someone to make a slick, professional, updated version of Dope Wars for iOS and/or Android ? one that tries to improve the core gameplay in ways like these, rather than slathering extra graphics and social features onto a game where they just get in the way. I?d pay for an app like that, and I bet a lot of others would too, and it wouldn?t be very hard to build.
And there are lessons here for anyone trying to implement ?gamification? to increase engagement with their own app; the first thing you have to do is understand which game elements are engaging and why, and the best way to do that is to look at the simplest, most stripped-down ?games? ? like Dope Wars or the math game I described in that previous post ? where those elements are most exposed. ?Gamifying? often seems to mean just bubbly childlike graphics, arbitrary ?achievements,? high scores lists and other ?game-like? elements. This is a cargo cult approach to the process, and it?s not surprising that it doesn?t usually work.
UPDATE: Some follow-up comments here.
