Who Will Live And Who Will Die On 'Game Of Thrones'? A Group Of German Students Created An Algorithm To Try And Predict It

Sansa Jonsa Game of Thrones ships
Sophie Turner as Sansa Stark in "Game of Thrones" Season 8. Series fans have long imagined a romance between the two characters, though there's little evidence to support it in the HBO. Helen Sloan / HBO

As the final season premiere of Game of Thrones approaches on April 14, a group of German students have created an algorithm to try and predict who will live and who will die on the hit HBO series.

The students study at the Technical University of Munich and utilized two different algorithms to come up with the formula behind their predictions. According to Mashable, the models were created based on survival-rate algorithms used in the medical field. The students also combined information from both the television show and the book series written by George R.R. Martin to determine the Potential Likelihood of Death (PLOD).

"The Song of Ice and Fire series is known for killing many characters, including prominent ones. While only fewest characters die of an old age, the majority meets a violent end. Does the death come at random or does it come to only those selected ones who exhibit similar features? The features can be certain age, heritage and gender. It can also be the fascinating ability of likely-to-die characters to make similar, mostly wrong, fateful decisions in their lives," the website created by the students reads.

The algorithm also takes things like popularity, house, gender, age, alliances and number of relatives into consideration in calculating a character's PLOD.

It isn't the first time students taking the class at the Technical University of Munich have embarked on such a project. According to Phys.com, a group of students in 2016 used a similar technique to predict the resurrection of Jon Snow.

"While the task of predicting survival chances for Game of Thrones characters relies on data taken from the world of fantasy, the exact same artificial intelligence techniques are used in the real world and are having a powerful impact on our everyday lives," Dr. Guy Yachdav, the class's lead mentor, told Phys.com.

According to the group's website, Daenerys Targaryen and Tyrion Lannister are most likely to survive the final season. Daenerys has a less than 1 percent chance of dying, while Tyrion follows at 2.5 percent. Rounding out the top five are Varys (3.2 percent), Samwell Tarly (3.3 percent) and Jaime Lannister (4 percent.)

On the opposite end, Bronn and Gregor Clegane are the most likely casualties of season eight ranking at 93.5 percent and 80.3 percent. Sansa Stark (73.3 percent), Bran Stark (57.8) percent and Sandor Clegane (47.5) are the others in the top five.

The student's work drew praise from the Chair of TUM's Informatics faculty, Professor Burkhard Rost, who told Phys.com, "The combination of passion and teaching is a brilliant way to create new tools that matter. In our course at TUM, we found a fun way to teach students how to use this technology and prepare them to build the next big thing once they graduate."

As for who lives or dies on Game of Thrones, fans will have to tune in to see. Season eight premieres April 14 at 8 p.m. on HBO for a six-episode run.

Uncommon Knowledge

Newsweek is committed to challenging conventional wisdom and finding connections in the search for common ground.

Newsweek is committed to challenging conventional wisdom and finding connections in the search for common ground.

About the writer



To read how Newsweek uses AI as a newsroom tool, Click here.

Newsweek cover
  • Newsweek magazine delivered to your door
  • Newsweek Voices: Diverse audio opinions
  • Enjoy ad-free browsing on Newsweek.com
  • Comment on articles
  • Newsweek app updates on-the-go
Newsweek cover
  • Newsweek Voices: Diverse audio opinions
  • Enjoy ad-free browsing on Newsweek.com
  • Comment on articles
  • Newsweek app updates on-the-go