Hollywood's award season reaches its extravagant finale at the Oscars on February 9 at the Dolby Theatre.
Experts predict Sam Mendes's 1917 , which was first shown to voters barely two months ago, could outflank all rivals. The degree of accuracy of predictions varies, from one predictive model to another.
A model built by independent statistical consultant Iain Pardoe has proved to be one of the most accurate. Pardoe uses three basic patterns which include using Awards preceding the Oscars as an important data point, the inter-relationship between nominations, and past nominations of lead actors and directors. Between 1938-2018, Iain Pardoe's model has accurately predicted 71% of the Oscars winners.
This year's Oscar favourites include 1917 for Best Picture, Sam Mendes as Best Director, Joaquin Phoenix as Best Actor and Renée Zellweger as Best Actress.
However, the Academy has thrown up surprises on many occasions. For example, in the 2016 Oscars, Moonlight with a 2% chance beat La La Land which had 97%.