Keep good records. These usually take the form of written observations or tables of results which show the effect of changing one variable on the outcome you are studying.
In real life, scientist investigating natural ecosystems and the populations of organisms within them, rapidly come up against some critical limitations in their work. It might be considered unethical, for example, to kill every oak tree in a forest to see what effect such devastation has upon the squirrel populations. Or it might be impossible to reduce the squirrel population to just two members, just so you can study how the population regrows to its natural carrying capacity.
But in this bio-simulation you have none of these limitations or restrictions. Also, you will not have to wait years until a population grows to its final size! So experiment as much as you want.
Qualitative and Quantitative Results Very often you will want to run the simulation just to see what kind of result you get. You might set the initial size of the Blue population at 100 members just to see what happens and what happens to the Blue and Red numbers. Do they go up, do they go down? This type of experiment is very common in science, it is a qualitative examination - just to see what happens. This is an adventure into the unknown and is often one of the more exciting parts of science, breaking new ground.
On the other hand, once you have tried a growth experiment "just to see what happens" you may then want to settle down and quantify the result. See how much a population goes up (20 units or 200 units?). This usually involves a systematic series of experiments in which you control all the variables, change one by a reasonable amount, record the results and do it all over again with a new value. Although often more boring, this is also an important part of science.