I was wondering what forces could make the car, while going down the ramp, move away from the surface of the ramp in the x-direction (if x was parallel to the ground)? I was thinking it could be the center of gravity of the car in the x-direction. In other words, the CG is too high above the car. Would this be a reason why my car does this?
You mean your car jumps off the ramp part way down? Yes, if I understand your suggestion, you're saying that your car is quite tall, and so when your car is quite vertical on the ramp, the force of gravity on the center of mass is tending to rotate your car about your front wheels, so the rear wheels lift off the ramp. If that's your problem, you need to lower the CM of your car, or make the ramp less steep. I suppose it could also be telling you that you're getting too much friction from your front wheels - if the front wheels don't rotate cleanly then you'll get a bigger rotational force on your car.
Another question, (physics noob), in my opinion, the inertia of the wheel shouldn't affect the exit velocity of the car at the bottom of the ramp by too much. So I think it would be better to use smaller wheels, which increase accuracy, rather than using the 2 3/8 in wheels. So what would be the advantages of using the 2 3/8 in wheels overusing the smaller like 1.5 in wheels?
The wheels are a small fraction of the car's mass, so you're right that the inertia of the wheels isn't very relevant. For accuracy, I assume you mean less distance per wheel turn, so it's easier to set your brake. I think that's true - but there might also be accuracy disadvantages associated with smaller wheels. I'd guess you could build a good car with either set of wheels, and that wheel choice will not be a dominant contributor to your final score. But that's not much more than a guess.
Also, I tried doing some math, and the max time is 1.81 sec (starting at the top of the ramp). Starting .75 m above would cause the time to only decrease by around .281 sec (this is in the ideal world). The effect will probably be more prominent in the real world so let's just say .5 sec. How will making the CG to the front of the car differ the accuracy of the car? Would this only negatively affect skidding (which is positive)?
You're assuming 100% energy transfer to the car's KE, no frictional losses, and timing the car over the 8m timing segment. As you say, this is ideal, and the real world will be slower and have friction, but probably wont slow things by more than a factor of 2 for a reasonable car. So I think you have a good working number there - suppose moving the CM costs you 0.5s, which is 0.5 points of total score. Can you build a car where the score you get for multiple runs is consistent to better than 0.5 points? If not, don't worry about the 0.5 points.