When to use evidence
If you say the sky is green, and your opponent says the sky is blue, even if you backup your claim with what seems like sound reasoning, it might still not be believed by the judge. Citing a qualified author at the right moment, or referencing a study when rebutting a point by your opponents can often make the difference between winning and losing a debate.
This doesn’t mean that you have to use a piece of evidence in every argument you make. But as long as what you say is a piece of common knowledge, or your point is backed up by sound logic, evidence is not always necessary. For example if I claim that “Russia and America often disagree about international issues,” that is something that anyone who reads a newspaper would know, so I don’t need evidence. And based on that I use analysis to make predictions about how Russia and America would behave in the security council.
But often facts on their own or or someone’s personal opinion is not enough to win a debate and could be interpreted in a variety of different ways. For example if I argue that, based on my analysis above, the UN Security Council can never get anything done because Russia and China would not agree. Then my opponent could argue that actually they agree on a lot of things. So being able to tell the judge about a piece of evidence from a qualified source could help persuade them. For example, evidence of a time when Russia and America disagreed and it has caused problems. Or a report from an expert on the UN on how the security council works in practice.
What is good evidence
The evidence you use should be: Fairly recent; from a qualified or relevant source; and sound in its reasoning.
Date: If the research you are citing is ten or twenty years old, many things will have changed so it won’t be very accurate. This varies a little depending on the topic. An article written last year about the Syrian civil war will probably be very out of date, but one on American infrastructure investment is probably still valuable.
Credible sources: Avoid websites like wikipedia or Baidu Baike as these encyclopedias can be edited by just about anyone, whether they are qualified or not. So although they are often accurate it is not guaranteed. Good sources are often newspapers or magazines, academic publications or information from organisations that do research into relevant issues.
Relevant sources: Stephen Hawking is a great physicist, but his opinions on history, politics or economics aren’t necessarily going to be better than an average person’s. The same is true for publications. Many have specialist areas of expertise, but won’t be as good in other areas.
Reasoning: Just because a source is generally good doesn’t mean it is right about everything. Even smart people and credible publications make mistakes. The Cato institute believes “the UN is bad because it limits individual nations economic sovereignty,” is better than “The Cato institute believes the UN is bad because it's secretly plotting world domination.” It is always more convincing to explain the reasons behind a source’s conclusion than to simply quote it.
Using evidence to win debates