Supercomputer gives Mo Salah an 11% chance of defending his Golden Boot crown

Erling Haaland will win the Premier League Golden Boot this season ahead of Mohamed Salah.

That's according to a Premier League supercomputer, which has revealed the stars who are most likely to finish the season as the league's top goal scorer just days out from the start of the new campaign.

Last year, Salah romped home to the award, finishing the season with 29 goals, six ahead of Alexander Isak with 23 and seven ahead of Haaland with 22.

But everyone will start on zero again next week when the new season starts, which will see Salah involved in the first game of the campaign when his Liverpool side take on Bournemouth at Anfield on Friday night.

It is then he will be looking to get off to a fast start, but is up against it in a bid to retain his trophy, according to the computer.

In numbers provided by CasinoHawks, Haaland has a 64.9 per cent chance of winning the Golden Boot he has already claimed in two of his three season as a Manchester City player.

That is a mammoth 53.9 per cent higher than Salah at 11 per cent, and nearly six times as likely as the Liverpool man.

That could, in part, be down to the Africa Cup of Nations, which takes place this season. Salah will almost certainly travel to the tournament with Egypt, with the tournament running from December 21 2025 to January 18 2026.

If Egypt go far, Salah could miss at least six league matches for Liverpool - just short of 16 per cent of the season.

Arsenal new boy Viktor Gyokeres is, meanwhile, third favourite, given a 4.9 per cent chance, ahead of Chris Wood (4.4 per cent), Cole Palmer (1.9 per cent) and Ollie Watkins (1.0 per cent).

Each player was also ranked on their expected goals projection, with Haaland given 24.2, Salah 18.2, Gyokeres 16.3 and Wood 16.1.

To calculate the chances, a simulation-based approach was used once all players expected goal projections were modelled. Each player's projected xG was modelled using data from previous seasons.

From then, 100,000 simulations – where each player's total goals were generated using a Poisson distribution based on their xG – was used.