It is actually about right. Returning a tennis ball is not an emergency situation, you are watching and calculating trajectories and are placed and prepared and swinging well before the ball is anywhere near you.
Starting a sprint is also an event thoroughly expected and planned for i.e. you know exactly what to do when the gun goes off. Those reaction times average about 0.15 of a second for Olympic sprinters.
A car emergency is completely different. It is unexpected and could be any number of situations with a number of possible responses.
So you first have to
(a) Recognise that there is an emergency event
(b) Process the situation. Is it a dog or a kid, is that car parked or moving, do I have room to swerve or will braking stop me in time, is the road surface stable on the edge, is that a tree etc.
(c) Choose a reaction from the multitude of possible responses.
I used to race and was involved in training drivers. There is a big difference in reaction time on the skid pan to an expected event to an unexpected one, and those published reaction times fit with what we use to see.
Monash Uni has a paper on it that is an interesting read and explains a lot.
http://www.monash.edu.au/miri/resear...ther/hfr12.pdf
The data starts on page 26, but the whole thing is a worthwhile read.
Try this experiment for reaction time in a totally expected and simple event
www.brainmac.co.uk/rulerdrop.htm