On driving games and the challenge of challenge

Driving games are, as a game genre, possibly the most exposed to providing an erroneous level of challenge to the player. When I think back on the most boring gaming experiences in my life, nearly all of them have been with driving games. The stereotypical game design of a driving game is pretty much designed for boredom rather than challenge, at least with regards to the game reward structure. Let me elaborate:

Most racing games feature a challenge structure which attempts to strike the optimal challenge flow through a slow increase in difficulty, this provided through a series of more and more challenging races. The increased difficulty has up until now mostly been provided through a combination of more challenging race tracks, faster cars, and more skilled AI drivers.

This, combined with the standard reward structure of racing games, places the player in a rather unfortunate situation:

If the player wants access to faster cars, better components, and the possibility to repair his rides, then he is forced by the game design to find a way to generate some income. 

Now, it is my impression that the game designers try to balance the game structure so that the player will receive just enough credits to get by on, if he follows the normal career path established by the game designers. In other words, the player will win prize money that will only just cover his financial needs, with regards to purchasing better/different cars. These purchases are most often necessary in order to progress in the game structure, and so the rewards can actually be considered mandatory.

However, this is where I see the glaring difference between good design intentions and actual player experiences.

For these reasons, many if not most players will be tempted to compete in races that are not challenging enough, since this will allow them to generate some in-game income more easily than when driving the tougher races with a more appropriate challenge rating for providing the player with a joyous, fun play experience (See the flow theory, by Mihály Csíkszentmihályi). In effect, due to the game design and reward structure, we're luring the player into cheating himself of a good play experience, and as a consequence, he will enjoy our game less. 

How can we remedy this?

As I see it, given that the reward system is the main reason why players are tempted to repeatedly compete in races with a challenge rating that's basically too low for them to have much fun, we should consider revising or completely altering the reward structure of racing games. Instead of basing rewards upon credits, we should try to hone the system so that the game rewards behaviour consistent with the player having the optimal play experience. 

So, in other words, we should give the player the highest, most usable reward for having the most fun, so that he or she is not tempted to trade a fun play experience for a more lucrative one.

Obviously, more research would be necessary in order to clearly determine the 'optimal fun zone' for driving games, but it is a relatively safe bet to say that a close race experience which nonetheless results in the player winning the race (barely), would constitute a very good and fun play experience when compared to a tedious race where the player easily wins the race without much challenge.

Thus, the most obvious answer would be to reward the player for having very close races, and this could be done in several ways: Giving a significant credit bonus if the race has been a close one, eranting additional rewards/unlocks for a narrow victory, and establishing some sort of multiplier for a very close second and third place.

So how do we determine whether a race has been close (and fun)?

I would propose an algorithm which examines not only the more obvious solutions such as the time difference between the player and his competitors, but also tracks the number of overtakes, crashes, and paint trading incidents, all in order to provide an AI evaluation of whether the race has been close. However, as we determined earlier on, the players will find the easy way to exploit the system, which is to not give it their all in the races, so we need a method for detecting this as well. Seeing that the player will most likely provide us with a very large data set over the course of the game, it shouldn't be too difficult to create an addendum to the algorithm which tracks player input behaviour in various race circumstances, in order to determine whether the player is lifting off through a corner in order to let the competition gain on him.

Furthermore, one could entertain the idea that a less skillful player might actually be awarded a better car than the more competent one, in order to ensure that the play experience conforms to the flow theory. But this is probably a relatively radical solution, and steps must be taken to ensure that the competent player does not feel cheated or begins to hold back in races in order to be awarded a better car. Perhaps we could do this behind the scenes, as with automatic difficulty adjustments in other genres? 

To sum up:

Players are tempted to grind in driving games, in order to get access to better cars and parts. This robs them of a potentially much more fun experience. We, as game designers, could presumably benefit greatly from attempting to revise the current standard reward structure in driving games, so that it entices the player to participate in races that will suit his or her skill level most. 

Comments, suggestions and different opinions more than welcome...