You could run the Spatial Join GP tool. Choose for
- target features: streets;
- join features: points;
- Merge rule:
Mean for the value field you have in cars point feature class.
This is how it is done:
After running the tool, you will have a new line feature class which will contain the information on how many cars are located on points as well as the mean value. The sample run result:
Things do get more complicated if there are multiple line features between the points (in the example I've run, there is only a single feature between the points. You would need either to merge those lines into single ones and then split at points where you have cars (to be able to get the values from cars properly distributed).
If you want to check whether a car point is on either end of a street segment and you won't / can't use Python scripting, you could use the Feature To Vertices GP tool to get both ends of every street segment and then run Find Identical GP tool to find those that are located on those end points. All of this would involve multiple steps.
You might look at ModelBuilder to see if building models is something for you, otherwise learn some primitive arcpy.