Survivorship Bias: Don't follow your dreams off a cliff
Phil Whitby | Aug 1st, 2019 | 5 min read
Taylor Swift telling you to follow your dreams is like a lottery winner telling you, ‘Liquidize your assets, buy Powerball tickets, it works!'
Those were the days!
Do you ever reflect wistfully about the music of your youth, classic cars, or films and conclude “They just don’t make them like that anymore”? When you look at buildings in a beautiful city like Florence, Athens or Paris do you wonder about the delights of bygone eras, how productive and inspired they were to produce gothic masterpieces like the Notre Dame cathedral or Westminster? If so, it is very likely that you are suffering from survivorship bias- a tendency to focus on the successful outcomes of history to the neglect of the myriad failures. A head for numbers
The metal helmet was introduced relatively recently in the first world war, but the reports from the field after the first months of usage were worrying: the number of head injuries was increasing.
This led to high level discussions about the possibility that the helmet was in fact a danger to the soldiers and should be recalled, until an astute individual realised that shrapnel would previously have killed a soldier who would now be surviving but injured. Soldiers who were uninjured and therefore saved by the helmet would not be recorded, and those that died would simply be listed as dead.
A similar misunderstanding occurred in world war two, when returning bombers would land with bullet holes and damage in perceived vulnerable areas of the plane and would therefore be reinforced in those areas before the next flight.
The similar error here is that the data for the destroyed planes was not available, and therefore the areas for critical destructive damage were not observed: the parts damaged on the planes that returned were tolerable and the bombers could remain effective. So (somewhat counterintuitively) the areas of the returning bombers that remained unscathed should be the areas considered for protection.
Turn on, tune-in, drop-out (and get rich)
Hugh Moore was a co-founder of American Water Supply Company, and a Harvard drop-out. He became extremely rich by helping popularise the paper cup as a water drinking container, shortly after the emergence of the germ theory of infection made drinking out of a publicly shared cup unacceptable. Other famous billionaire drop-outs include Steve Jobs, Bill Gates, Larry Ellison, MIchael Dell…and the ultimate college drop-out: Kanye West.
What can we conclude from the fact that all these successful individuals were drop-outs? Presumably applying to a University is strategically beneficial only for the purposes of leaving at a crucial juncture. In reality Steve Jobs et al. are similar to the surviving bombers or the injured soldiers of the war, they are not representative of the whole cohort of drop-outs, and it only takes a moments reflection to realise that the majority of drop-outs attempting to emulate their heroes will fail. The broader reality is that 94% of the most successful business owners are in fact graduates.
Children are a significant investment (group)
Imagine prompting a class of primary school children to pick a stock from a list. Those that chose a stock that increased in value over the week would be retained in your elite club of investors and the ‘losers’ excluded. Over the weeks you would eventually be left with the ‘elite’ investors who only picked the finest stocks, and possibly dreaming of riches beyond your wildest dreams.
Why hasn't this happened, and why doesn’t it seem like a good idea? The reason - that we intuitively grasp - is that the children will all be using an essentially random approach, and that the ‘survivors’ will only be lucky and nothing more.
This is also the reason why the financially excessive ‘retention bonus’ model that investment banks use is completely bogus. If the necessity for talent retention is so strong, why would the bonus for investment portfolio performance be awarded to different employees year after year?
Daniel Kahnemann demonstrated that maintaining the same portfolio over the year rather than buying and selling lead to better overall gains in general, and therefore any ‘talent’ was barely detectable.
Eyes on the survivor or eye of the tiger?
Our elite school of child investors will not pass the Dragon’s Den test, but what other situations are we analysing when we do not consider those that do not survive?
As an avid football enthusiast, I am often reading about the success of particular football academies that have produced clusters of talented youngsters who go on to sign a bumper professional contract with a footballing giant. A famous example is Barcelona FC’s ‘La Masia’, an academy that produced a significant portion of the great Barcelona team of this century, but now barely supplies one or two players to the team in recent years.
Many column inches have been written about the system’s innovation and failures, but if a handful of players make it through couldn't this just as easily be attributed to chance? The high throughput of young players (40 boys to a room according to Carlo Puyol) and the tiny margins mean that any tiny change in luck could be amplified massively. After all 1% of boys who enter a football academy make it to a professional contract, and only 180 out of the 1.5 million players of organised youth football make it to the premier league. How many young children are taking the advice of the 99% of boys who didn’t make it rather than the ‘follow your dreams’ 1 percenters?
This is not an article written with the express intention of dissuading people to follow their dreams. Heightened awareness of survivorship bias allows you to plan for contingencies and worst-case scenarios. Similar to the illusion of control discussed in my previous article, the point is to optimise your chances of success in general, rather than the success-at-all-costs approach that popular culture prescribes.