Before the holidays we started introducing core concepts from our book, Decision Strategy, that was awarded the 5th best management book globally of the year. This is part 5, so if you have missed the earlier parts, click the links below:
Part 5: If confidence is good, then overconfidence must be…
”All you need in life is ignorance and confidence, then success is sure”
“The game is over”, said Mohammed al Douri, Iraq’s ambassador to the UN in April 2003. After only three weeks of fighting US forces occupied Baghdad, while Saddam Hussein was caught in a cellar on the outskirts of Tikrit later in the year. But while the invasion turned out to be a great success, the Iraqi liberation war was anything but. In Washington Saddam’s dictatorial regime was expected to crumble as soon as America set foot in the country, but Shiites did not rise, the Sunnis fought fiercely, the Iraqi guerrilla war surprised the unprepared forces and there were no weapons of mass destruction. How could the US be so wrong in a time when the intelligence community was better equipped than ever?
It all started with a few neoconservatives – among which the most prominent was Defense Secretary Paul Wolfowitz – who had long been convinced that expelling Saddam Hussein would pave the way for a grand reorganization of the Middle East that would move it away from tyranny and anti-Americanism and toward modernity and democracy. But maybe the most prevalent reason to attack Iraq was, that it would be a courageous use of American power, which mixed raw strength with idealism after the 1990s years of retreat and the terrorist attack on September 11. A combination of ambiguous intelligence and a strong belief that one could easily ‘smoke the bad guys out’, underpinned rosy scenarios, and poor planning – for example the original game plan called for 500,000 soldiers but was cut down to 160,000 for 3 months in the final approach – and only 8 years later the last soldiers were extracted.
The reason we often enter hopeless wars lie in our tendency of overconfidence. Although a more realistic assessment of the situation and the alternatives could lead to more peaceful solutions, our view is obscured by positive illusions, wishful thinking and overconfidence. Historical data show that each side before the start of the war, are convinced that they have more than 50 percent chance of winning!
Our tendency to overestimate our skills, the accuracy of our decisions or the value of our ideas is not limited to war. It is human and found across time, cultures and circumstances. Even the best laid business plans are often ruined by the annoying interference by reality. There is an abundance of examples of overconfidence in professional contexts such as stock market bubbles, the number of new entrepreneurs setting up shop despite 90% default rates and the many high profile acquisitions despite high risk of failure.
But could overconfidence not be a good thing? Is it not our self-conceit that makes us rise to the challenge and pushes us to perform better than our perceived limits? The short answer is no. With the exception of some limited benefits in innovation processes, overconfidence is one of the biggest obstacles to good decision making.
Although self-conceit can be fatal to decision-making processes, it is almost ironic how easy it is to correct. Churchill once said that it is important to remember that no matter how certain you feel about victory, there would be no war if the other person did not also think he had a good chance. If only the Americans had remembered to ask themselves a few perspective changing “what-if” questions, they could have avoided some of the biggest challenges. For example, why did Saddam Hussein dare risk a war? With the loss of the Gulf War in 1991 and a widening gap between Iraq and US forces, the outcome was given for the invasion – but the war was not.
When discussing overconfidence there are three different scenarios:(1) overestimation of our actual performance, (2) overplacement of our performance against others and (3) overprecision, that is exaggerated confidence in the accuracy of our beliefs:
Overestimation – the reason for project being late and over budget
You are preparing slides for a last minute presentation. It is the night before, but you do not worry much, because you are an experienced speaker. Although it is a new topic, you got slides from the former lecturer and you usually get good feedback. When you take the stage, you realize that your lecture is completely off target. You were not prepared for the participants’ prior knowledge of the subject, you have not quite understood all the points in the slides that you got and 45 agonizing minutes seem like a lifetime.
We tend to believe that we are better across a wide range of domains than we actually are, that we have more control over situations than we actually have and that we can plan out things quite detailed. This planning fallacy is reason for projects coming in late and over budget.
This is also yet another reason that bonuses can be difficult to use as a motivator. You may receive a bonus of USD 10,000, but you have expected 5,000 or 15,000. The easiest way to exceed our expectations is to reduce them and we have therefore developed defensive pessimism: While we will start out optimistic in a year or a project, we shield ourselves from any disappointment towards the end of the year or project through pessimistic assessments about ourselves and our possibilities, because it feels extra hard when our inflated belief meets reality.
Overplacement – the reason we enter hopeless projects
93% of motorists believe that they are better than average and 25% of students consider themselves in top 1%. Besides making performance appraisals difficult to agree on between managers and employees, it means that entrepreneurs will go into markets where their objective possibilities for success are limited or we will continue a hopeless lawsuit at high costs.
Interestingly, recent research also provides for our tendency to believe that we are performing worse than others when it comes to very difficult tasks. Both extremes are problematic. While overplacement may lead us to throw money at bad projects, then underplacement may prevent us from pursuing great projects – whether in business or in life.
Overplacement can create some nasty surprises, when we suddenly face reality. For example highly intelligent people joining ivy league schools often drop out, when they finally meet real competition. We simply tend to focus on ourselves instead of comparing ourselves with the particular group we belong – or we think of our own team as more competent than average. That would be okay except the other talented teams may impact on our own teams opportunities. In the business world this leads to lack of understanding of the market and competitors.
Overprecision – never ask experts to predict the future
Throughout history experts’ overly precise estimates have proven to be wrong again and again. For example, the neoclassical economist, Irving Fisher, who became famous for shortly before the Wall Street crash in 1929 saying that stock prices seemed to to have reached a permanently high level. Or Harry Morris Warner, one of the founders of Warner Bros, who in 1927 rhetorically asked who on earth wanted to hear actors talk? Or Thomas Watson, founder of IBM, who in 1943 predicted that there might be a world market for five computers.
Of course everyone can make mistakes in areas where they have limited knowledge, but the question is whether expertise buys us more precision? Studies show that we in areas with prior knowledge or expertise closer to the correct answer. The problem is that we as experts are becoming more cocksure about our ability to predict the future and therefore defines narrow confidence intervals (i.e. how sure we are of our estimate), so we still miss the target. The problem is that our one lone estimate has a tendency to make mistakes.
The problem is further cemented by human interaction patterns. For example few voters will choose a politician with the slogan: “I think it’s way, but I’m not sure.” We feel that confident people are more persuasive, competent and we reward them with influential positions. When the confidence and capability is positively correlated, it makes sense. But people quickly learn that to succeed, they must adopt a ‘fake it till you make it’ approach.
6S Model – bring in the devils advocate
Overconfidence is not the hardest bias to work with. Often we just need to consider consider worst-case scenarios, adverse information or adjust our project estimates with standard percentage to minimize the effect. Here are a few examples from the 6S parameters of Strategy, Structure, Steps, Systems, Skills and Style:
In Strategy you might towards the end of the process deploy Gary Klein’s “premortem” technique, where you imagine that the project has failed and you are analyzing the reasons behind. It is a clever way to invite in the devils advocate without strong opposition and is great at uncovering the biases and challenges that may threaten your strategy or project.
In Systems overoptimistic organizations often overestimate their abilities and underestimate how long something takes. Although you should always start with detailed plans to increase realism, then make sure to build in flexibility around those very precise estimates and consider whether the decisions in the system should be “opt in”, where you must actively choose a direction rather than an “opt out”, where you can just let system 1 press the accept button without reflecting on the consequences
This was part 5 of our book, Decision Strategy. Next week we will look at the power of loss aversion and the crazy lengths we will go to avoid it – stay tuned!.