Well I think in discussion you basically talk about your experiment. What you did right, wrong or how to improve the experiment. You can also talk about what you had assumed during the experiment to work out your calculations hence the different results in theoretical and empirical results. You really basically just talk about your results

You might want to add things like summarising your hypothesis (Not sure about this one.. you might have to check with someone else)

Since it's projectile, the most likely test is the Pendulum one, so if you want a head start you might want to check up on it

In your discussion;
- Assess reliability
+ if you drew a graph, did the points of the graph adhere well to the LOBF?
+ if you recorded results for more than 1 test run of the experiment, average the results, find the greatest deviation and calculate the percentage of the deviation.
- Assess/justify validity
+Was the methodology correct?
+Were all variables controlled except the independent variable?
+ Did you run enough repeat trials to record consistent results
+ Did you apply appropriate analysis of the results e.g graphical analysis
- Calculate the experiment error using the formula thingymajiggy
- Further investigations
+ improvements that could be made next time or something like that

reliability, validity, accuracy, possible errors in experimental design and how these errors could have affected your reliability/validity/accuracy, ways to improve reliability/validity/accuracy, account for possible disparities from theoretical results...