AI can tell if a screen star’s best years are behind or in front of them


An actor’s life is often portrayed as a struggle to get to the top, but now scientists claim to have created an AI that can predict success in show business.

Mathematicians at Queen Mary University, London, say they can accurately predict whether an actor’s career has peaked – or if their most successful days still lie ahead.

They discovered that an actors’ most productive year – defined as the year with the largest number of credited jobs – is towards the beginning of their career.

Clear signals preceding and following the ‘annus mirabilis’ (AM) enable them to predict with around 85 per cent accuracy if it has passed or not.

The study also describes how the vast majority of actors and actresses, around 70 per cent, have careers that last for just one year.

Such ‘one-hit wonders’ are the norm rather than the exception as long careers with lots of jobs are rare, according to the findings, suggesting a scarcity of resources in the acting world.

Using data from the Internet Movie Database (IMDB), the researchers studied the careers of more than 1.5 million actors and over 896,000 actresses around the world from 1888, when the first film was made, up to 2016.

They discovered that acting careers are clustered into ‘hot’ and ‘cold’ streaks as actors and actresses don’t tend to work at a steady rate in a business where unemployment rates hover at around 90 per cent.

Experts used the AI software and the patterns it detected to study the careers of former stars of stage and screen. 

It correctly identified that Oliver Reed’s AM was in 1960. 

It also predicted that Amanda Bynes passed her peak in 2006, the year before she was named one of the ’25 Hottest Stars Under 25′ but shortly before she retired.

The research team also discovered ‘huge evidence’ of gender bias in the industry, as most of the patterns observed were different for actors and actresses.

For instance, actors are more likely to find work after a cold streak while the most productive year for an actress is more likely be at the start of their career.

And, when careers last more than one year, it is more common to find actresses with shorter career lengths than actors. 

The researchers were inspired to look into showbiz after previous studies had analysed the career success of scientists and artists which was found to be unpredictable.

They aimed to define, quantify and predict the success of actors and actresses in terms of their ability to maintain a steady flow of jobs.

Study co-author Oliver Williams said: ‘Only a select few will ever be awarded an Oscar or have their hands on the walk of fame, but this is not important to the majority of actors and actresses who simply want to make a living which is probably a better way of quantifying success in such a tough industry.

‘Our results shed light on the underlying social dynamics taking place in show business and raise questions about the fairness of the system.

‘Our predictive model for actors is also far from the randomness that is displayed for scientists and artists.’

The researchers found that the total number of jobs in a career is underpinned by a ‘rich-get-richer’ phenomenon. In other words, the most famous actors get the most jobs.

The researchers said that the result was not unexpected as the more well-known an actor is, the more likely producers will want him or her in their next film, if only for commercial purposes.

But they said that what was interesting was that rich-get-richer effects are well known to develop out of arbitrary and unpredictable random events that get amplified.

So an actor’s success could be down to circumstance rather than their acting ability, known as the ‘network effect’.

The study also shows that actors with long cold streaks who then experience late comebacks are rare but difficult to predict, and as such the destiny of each actor is not entirely determined.

Study co-author Dr Lucas Lacasa said: ‘We think the approach and methods developed in this paper could be of interest to the film industry.

‘For example, they could provide complementary data analytics to IMDb. This does also bring with it a number of open questions.

‘We have assumed that there is nothing anyone can do to change their fortunes, but we have not shown that this has to be the case.

‘Consequently we are interested in finding out how an individual might best improve their chances of future success.’

The researchers hope that their method will contribute to the new science of success, and that refined versions of the prediction model will be even more accurate.

Dr Lacasa added: ‘This research has sparked a lot of attention from unexpected places including from a screenwriter from the film industry who is now developing a movie script partly based on our findings.’

The full findings of the study were published in the journal Nature Communications.


About Author

Leave A Reply