Adding simple statistics can add heft to your arguments, written or spoken.  However, when qualified they can serve as little but a distraction.

Take the following sentence from a paper by a college professor with a PhD, whose name shall remain anonymous for his/her protection.

Some estimates are that as many as 20-85% of people experience more or less anxiety when then need to speak in public.

This is a classically wishy-washy (to use a technical term) sentence, full of qualifications:

Some estimates – better to skip the word “some” here – do the other estimates contradict your point?

as many as – followed by a range; “as many as” should cite the upper limit of a range.

20-85% – that not exactly nailing a narrow range, is it?  Each major party candidate for president will get somewhere between 20-85% of the vote, also.

more or less anxiety – We can agree that 100% of people experience more or less anxiety.  But more or less than whom?

Edit the statistical fluff and ambiguous phrases from that earlier sentence and you get this: Estimates are that as many as 85% of people are nervous about speaking in public.

If you are one of that anxious 85% you should work with us; we are 100% certain that we can help you become less nervous.