Manchester City Have Five Wins From Five, But History Is Not On Their Side

Manchester City head into the international break at the top of the Premier League and three points clear. 

New coach Enzo Maresca extended his perfect winning start to the season to five games, and the fifth win was not straightforward. City twice squandered the lead against Sunderland, and Brian Brobbey outshone Erling Haaland with a hat-trick, before Haaland settled a 5-3 win in the 81st minute. 

The timing helped. Defending champions Arsenal lost 3-0 at Brighton the day before, which ended their own perfect start and handed City outright first place. 

That leaves one question for anyone looking at the outright market: how often does a five-from-five start actually end with the trophy? 

What history tells us

City are the 15th side in Premier League history to win their opening five league matches. That group runs from Newcastle in 1994/95, through Chelsea's record nine straight wins in 2005/06, to Liverpool in 2025/26.^[1] 

Of the 14 previous sides, only four went on to lift the title. That is a conversion rate of 29%. The two Chelsea sides of 2005/06 and 2009/10 did it, as did Liverpool in 2019/20 and City themselves in 2023/24.

A perfect start is a much better guide to a top four finish. Twelve of the 14 finished in the top four, and runner-up was the most common outcome, with five sides finishing second. The only two to fall outside the top four were Kevin Keegan's Newcastle, who finished sixth in 1994/95, and last season's Liverpool.

Liverpool are the warning from last season. City's 15 points matches Liverpool's haul from the same stage last season, but the Reds eventually struggled to a fifth-place finish.^[2]

City's own record with perfect start

This is City's fourth perfect start in the Premier League era, and the previous three ended very differently. The 2015/16 side won its first five and finished fourth. The 2016/17 side, in Pep Guardiola's first season, won its first six and finished third. Only the 2023/24 side turned a perfect start into the title.

The first season angle matters here. Maresca is the fourth manager to win his first five league games in charge of a side. Carlo Ancelotti went on to win the league with Chelsea in 2009/10. Guardiola's City in 2016/17 and Maurizio Sarri's Chelsea in 2018/19 both finished third. 

Maresca took on arguably the most daunting task in the sport when he replaced Guardiola over the summer, so a third-place finish would not be a disaster. But it would not win anyone a title bet. 

The underlying numbers

The underlying numbers are the reason for caution. 

SportsBoom's model has City on 9.2 expected points from their five games. That is 5.8 fewer than the 15 they have actually taken. No side in the division has overperformed its chance creation and concession by more.

Game state explains part of the gap. City have spent 52.5% of their minutes in front, more than any other team. Leading sides tend to concede territory and shots, which drags down their xPts.

It does not explain the Sunderland game. City led that match and still conceded three at home. City could not establish the control they had shown in the Manchester derby, and Sunderland kept finding gaps in the home defence. 

Chelsea in 2005/06 and Liverpool in 2019/20 had dominant underlying numbers to back up their perfect starts. On current evidence, this City side does not. 

What the market says

Bookmakers have not been swayed by the table. Arsenal remains title favourites at 11/10 best, while City are 6/4 best.^[3]

Liverpool are third favourites at 20/1 best, and Chelsea are 28/1 best. 

City's best price of 6/4 implies a title chance of about 40%. That is well above the 29% historical conversion rate for perfect starts.

The models are close to the market. Opta's prediction model gives Arsenal a 43.52% chance of the title and City 40.96%. It projects Arsenal to finish on 76.06 points and City on 75.64. That is a big move from the start of the season. Heading into the campaign, Opta gave City just a 19.6% chance of winning the title in Maresca's first season. 

Prediction markets are slightly more cautious. Combined Kalshi and Polymarket pricing had City at 38.2% and Arsenal at 44.8% as of 21 September.^[4]