在一般的公司里，很多决策或是以独断专行方式，由高层领导自上而下做出，或是以民主协商方式，由每个人分享各自观点，最终实施得到最多支持的观点。这两个决策系统都有缺陷。这是因为最佳决策应该是在创意择优中，做可信度加权，其中能力强的人努力解决彼此分歧，他们都各自独立思考了事实是什么，应采取什么行动。
与能力较弱的决策者相比，要对那些能力更强的决策者的观点赋予更大的权重。这就是我们所谓的
“
可信度加权的决策
”
。那么，怎样确定谁在哪些方面能力更强呢？最具有可信度的观点来自：（
1
）多次成功地解决了相关问题的人；（
2
）能够有逻辑地解释结论背后因果关系的人。当可信度加权能够正确实施并保持下去，那就形成了最公平、最有效的决策系统。它不仅能产生最佳结果，也能保持步调一致，即便是对决策持不同意见的人，也会接受它。
为了做到这一点，可信度加权的标准必须是客观且得到每个人信任的。在桥水，每个人观点的可信度都被记录在案并接受系统性评估，使用类似棒球卡和集点器的工具，及时记录和评估其工作经历和表现。开会时，我们经常通过集点器
App
来对很多事项投票表决，不仅仅呈现等权重平均的结果，也有可信度加权的结果（还有每人投票情况）。
一般而言，如果基于等权重平均和可信度加权的投票结果吻合，我们就认为问题已经得到解决，可以继续讨论别的话题。如果两类投票结果不一致，我们就再次努力解决。如果解决不了，我们会采纳可信度加权投票的结果。视具体的决策类型不同，在有些情况下，某个
“
责任人
”
可能会推翻可信度加权的投票结果，在其他情况下，也会反过来。但不管怎样，在所有出现分歧的情况下，可信度加权的投票结果都会得到高度重视。即便责任人能推翻可信度加权投票的结论，其也有义务在推翻之前尽量把分歧解决掉。在桥水工作的
40
年中，我从来没有做过与可信度加权结论相反的决策，因为我觉得，那样做是傲慢的，而且与创意择优的精神实质相悖，尽管在此过程中我会为自己认为是最佳的方案竭尽全力去辩护。
我举一个例子说明实际工作中这个过程是怎么完成的。
2012
年春，欧债危机不断升级，我们的研究团队运用可信度加权的决策程序，解决了关于下一步趋势判断的分歧。当时，意大利、爱尔兰、希腊、葡萄牙，尤其是西班牙政府的借债和还本付息金额都已远远超过了其偿付能力。我们知道欧洲央行要么会史无前例地购买政府债券，要么令欧债危机自行陷入深渊从而可能导致国家违约和欧元区解体。德国强烈反对欧洲央行出手救助。显然，这些国家以及整个欧元区的经济命运将取决于欧洲央行行长马里奥
·
德拉吉领导下的欧洲央行的下一步措施。但是德拉吉究竟会怎么做呢？
这个过程就像针对国际象棋棋盘开展分析，预测不同棋手的棋子移动方向以及影响，我们每个人都从不同角度对形势走向提出自己的看法。经过漫长的讨论后，我们仍无法达成一致：大概有一半人认为欧洲央行会印更多钞票来购买债券，而另一半人则认为不会，因为与德国闹翻会给欧元区带来更大威胁。尽管这些理性、公开的交流是必不可少的，但同样重要的是，要有双方都认可的方式来解决分歧，达成最优决策。因此，我们运用可信度加权的分析系统来打破僵局。我们采用了集点器工具，它能帮助我们分析不同观点的特点，找出分歧的源头，按照可信度进行梳理找出解决办法。大家不同特质的可信度权重不一样，比如对某个领域的技能、创造力、综合分析能力等。这些“点“是根据一系列混合评分而定（包括同事的打分和各种测试）。通过对这些特质的分析，并判断哪些与当前的问题更加密切相关，我们可以得出最佳决策。
在上述案例中，我们进行了可信度加权的投票，所选取的特质是专业技能和综合分析能力。集点器清楚地显示了这样的结果：那些可信度更高的观点认为，德拉吉将违背德国的意愿，印钞购买债券。我们就据此做了相关决策。几天后，欧洲的政策制定者公布了一个大规模、无限量购买政府债券的计划，可见我们的判断是对的。尽管可信度加权给出的答案未必总会是最佳答案，但我们发现，与任凭老板一个人裁决，或一人一票公投式表决相比，用这个办法更可能做出正确的判断。
不管你是否运用这种技术和结构化的程序来做可信度加权，最重要的问题在于你正确理解了可信度加权的概念。很简单，在做决策时，看看你自己和你的团队，考虑谁最有可能是对的。我敢保证，如果你这样做了，你肯定比不这样做更容易做出上佳决策。
In typical organizations, most decisions are made either autocratically, by a top-down leader, or democratically, where everyone shares their opinions and those opinions that have the most support are implemented. Both systems produce inferior decision making. That’s because the best decisions are made by an idea meritocracy with believability-weighted decision making, in which the most capable people work through their disagreements with other capable people who have thought independently about what is true and what to do about it.
It is far better to weight the opinions of more capable decision makers more heavily than those of less capable decision makers. This is what we mean by “believability weighting.” So how do you determine who is capable at what? The most believable opinions are those of people who 1) have repeatedly and successfully accomplished the thing in question, and 2) have demonstrated that they can logically explain the cause-effect relationships behind their conclusions. When believability weighting is done correctly and consistently, it is the fairest and the most effective decision-making system. It not only produces the best outcomes but also preserves alignment, since even people who disagree with the decision will be able to get behind it.
But for this to be the case, the criteria for establishing believability must be objective and trusted by everybody. At Bridgewater everyone’s believability is tracked and measured systematically, using tools such as Baseball Cards and the Dot Collector that actively record and weigh their experience and track records. In meetings we regularly take votes about various issues via our Dot Collector app, which displays both the equal-weighted average and the believability-weighted results (along with each person’s vote).
Typically, if both the equal-weighted average and the believability- weighted votes align, we consider the matter resolved and move on. If the two types of votes are at odds, we try again to resolve them and, if we can’t, we go with the believability-weighted vote. Depending on what type of decision it is, in some cases, a single “Responsible Party” (RP) can override a believability-weighted vote; in others, the believability-weighted vote supersedes the RP’s decision. But in all cases believability-weighted votes are taken seriously when there is disagreement. Even in cases in which the RPs can overrule the believability- weighted vote, the onus is on the RP to try to resolve the dispute before overruling it. In my forty years at Bridgewater, I never made a decision contrary to the believability-weighted decision because I felt that to do so was arrogant and counter to the spirit of the idea meritocracy, though I argued like hell for what I thought was best.
To give you an example of what this process looks like in action, during the spring of 2012 our research teams used believability- weighted decision making to resolve a disagreement about what would happen next as the European debt crisis was heating up. At that time, the borrowing and debt-service needs of the governments of Italy, Ireland, Greece, Portugal, and especially Spain had reached levels that far exceeded their abilities to pay. We knew that the European Central Bank would either have to make unprecedented purchases of government bonds or allow the debt crisis to worsen to the point where defaults and the breakup of the Eurozone would probably occur. Germany was adamantly opposed to a bailout. It was clear that the fates of these countries’ economies, and of the Eurozone itself, depended on how well Mario Draghi, the president of the European Central Bank, orchestrated the ECB’s next move. But what would he do?
Like analyzing a chess board to visualize the implications and inclinations of the different moves of the different players, each of us looked at the situation from every angle. After a lot of discussion we remained split: About half of us thought the ECB would print more money to buy the bonds and about half thought they wouldn’t, because breaking with the Germans would threaten the Eurozone even more. While such thoughtful and open exchanges are essential, it’s also critical to have mutually agreed-upon ways of resolving them to arrive at the best decision. So we used our believability-weighting system to break the stalemate.
We did that using our Dot Collector tool, which helps us surface the sources of our disagreements in people’s different thinking characteristics and work our way through them based on their believabilities. People have different believability weightings for different qualities, like expertise in a particular subject, creativity, ability to synthesize, etc. These dots are determined by a mixture of ratings, both from peers and tests of different sorts. By looking at these attributes, and also understanding which thinking qualities are most essential to the situation at hand, we can make the best decisions.
In this case, we took a believability-weighted vote, with the qualities chosen being both subject-matter expertise and ability to synthesize. Using the Dot Collector, it became clear that those with greater believability believed Draghi would defy Germany and print money, so that is what we went with. A few days later, European policymakers announced a sweeping plan to buy unlimited quantities of government bonds, so we got it right. While the believability-weighted answer isn’t always the best answer, we have found that it is more likely to be right than either the boss’s answer or an equal-weighted referendum.
