People don't evaluate your facts on their own merits. They evaluate them through a filter of whether you understand their side of the problem. If that filter reads no, your facts get discounted, no matter how accurate, because the listener treats you as arguing from outside the problem rather than inside it.
Two people can present the same data and get opposite receptions. One is heard as someone who grasps the full tradeoff, so their facts land as informed judgment. The other is heard as someone who found a number to support a conclusion, so the same fact lands as a tactic. The data didn't change. The credibility filter did.
You build that credibility by stating the other person's position better than they stated it, including its actual strongest version, before you present a single fact. This isn't agreeing with them. It's proving you did the work to understand their constraint. Once you've done that out loud, the filter flips, and your facts get evaluated instead of deflected.
The order can't be skipped. Facts first, understanding second, and the understanding reads as a concession after losing ground. Understanding first, facts second, and the facts read as the next step from someone who already gets it. Same words, different sequence, different outcome. This is why strong data loses to weak data. It's sequencing, not persuasion.
The IBC course calls this steelmanning: stating the other side's position more fairly and completely than they could themselves, before you present your own case, so your facts are heard by someone who already trusts that you understand the problem.