对你的决策进行可信度加权

我发现，和高度可信、愿意表达深思熟虑的分歧的人交流，一直都能深化我对问题的理解，改进我的决策质量。这通常能让我做出更好的决策，并给予我令人兴奋的学习机会。我建议你也这么做。

为了做好这一点，你必须避免以下常见失误：（1）不合逻辑地高估自己的可信度；（2）不区分可信度不同的人。

在与其他人产生分歧时，应该首先看看你们能不能就决策原则达成一致。在进行这样的讨论时，你们应该分析不同原则背后理由的优劣。如果能就决策原则达成一致，你们就把原则应用于眼前的问题，得出一个人人赞同的结论。如果对决策原则有分歧，你们可以通过比较彼此的可信度来努力解决分歧。我将在“工作原则”中更详细地解释具体做法。

这种有原则的、可信度加权的决策方式很有吸引力，效果比一般的决策方式好得多。例如，假设我们用这种办法来选总统。就一个好总统应该具备什么样的条件、谁在做出上述判断时最为可信这两个问题，看看我们能提出哪些原则，这应该很有意思。我们最终的讨论结果将是实行一人一票还是别的方式？如果是别的方式，应该怎么做？这肯定能带来非常不同的选举结果。在下一次选举的时候，我们可以在正常选举的同时进行这样的演练，以观察二者的差别。

尽管可信度加权的决策听起来有些复杂，但你很可能经常这么做，比如你问自己“我应该听什么人的意见”时。但几乎可以肯定的是，如果你更有意识地这么做，效果会好很多。

I have found triangulating with highly believable people who are willing to have thoughtful disagreements has never failed to enhance my learning and sharpen the quality of my decision making. It typically leads me to make better decisions than I could have otherwise and it typically provides me with thrilling learning. I urge you to do it.

To do it well, be sure to avoid the common perils of: 1) valuing your own believability more than is logical and 2) not distinguishing between who is more or less credible.

In case of a disagreement with others, start by seeing if you can agree on the principles that should be used to make that decision. This discussion should include exploring the merits of the reasoning behind the different principles. If you agree on them, apply them to the case at hand and you’ll arrive at a conclusion everyone agrees on. If you disagree on the principles, try to work through your disagreement based on your respective believabilities. I will explain how we do this in more detail in when I get to my Work Principles.

This sort of principled and believability-weighted decision making is fascinating and leads to much different and much better decision making than is typical. For example, imagine if we used this approach to choose the president. It would be fascinating to see which principles we would come up with both for determining what makes a good president as well as for deciding who is most believable in making such determinations. Would we wind up with something like one person one vote, or something different? And if different, in what ways? It certainly would lead to very different outcomes. During the next election, let’s do this in parallel with our ordinary electoral process so we can see the difference.

While believability-weighted decision making can sound complicated, chances are you do it all the time— pretty much whenever you ask yourself, “Who should I listen to?” But it’s almost certainly true that you’d do it a lot better if you gave more thought to it