What the first Premier League meeting of a season tells us about the second

Chelsea and Fulham meet for the second time in 23 days on Friday night. On Saturday Manchester United and Crystal Palace will play for the second time in 18 days, and a day later Tottenham and Manchester City will also play for the second time in 18 days. Next week Manchester United and Leeds will play twice in five days.

Every pair of Premier League teams meet twice in a season. What does the first meeting tell us about the second? And is it different if they are close together?

I looked back over the last 27 completed seasons, 1995-96 to 2021-22.

The shortest gap between games was in 2009-10 when Arsenal and Bolton played twice in four days. Arsenal won 2-0 away then 4-2 at home.

So the second result was the same as the first. This often happens. The standout feature of games played in quick succession, though, was that in the second meetings there were an unusually high number of draws.

Southampton and Liverpool played twice in 11 days in 2001-02. Southampton won 2-0 at home then drew 1-1 away. The same season Chelsea and Bolton played twice in 21 days. Chelsea won 5-1 at home then drew 2-2 away.

I studied all pairs of games completed within a month – by which I mean that if, for example, the first game was played on February 3 the second game was played on or before March 3.

Broadly speaking the relationship between first and second games was similar to what it was when they were further apart. There was only one difference worth mentioning: no matter what the result of the first game, in the second games there were an unusually high number of stalemates. Thirty-three per cent of those games were drawn.

Familiarity in football probably does not breed contempt but it might breed a conspicuous number of ties.

Across the 27 seasons pairs of games were separated by as little as a few days up to as many as ten months. On average the gap was four months. More generally, what did the first meeting tell us about the second?

Overall 46 per cent of games were home wins, 25 per cent were draws and 29 per cent were away wins. Those are our reference points.

The proportion of home wins in return meetings was 40 per cent for teams who had lost away in the first meeting, 50 per cent for teams who had drawn away in the first meeting and 55 per cent for teams who had won away in the first meeting.

The proportion of away wins in return meetings was 36 per cent for teams who had won at home in the first meeting, 24 per cent for teams who had drawn at home in the first meeting and 20 per cent for teams who had lost at home in the first meeting.

The proportion of draws was less variable. It was always between 24 and 26 per cent.

The better teams did in the first game the better they did in the second. The chance of winning the second meeting differed by up to 16 per cent depending on whether teams had won or lost the first meeting.

There is a common sense explanation for this. More often than not, the team who won the first game will have been more skilful than the team who lost. There would have been some fluke results, scores that made a nonsense of the run of play or performances that were unrecognisable from those that had come before or would come after. More often, though, games would have gone according to form.

True, we can assess form more fully in other ways. Nonetheless, the timing or result of the first meeting between teams can still tell us something – perhaps more than we had imagined – about what the result might be in the second meeting.

Why Arsenal and Liverpool are new versions of Leicester

What do you think is the next number in this sequence: 31, 27, 34, 35, 34? If you said something in the low to middle 30s your answer is perfectly understandable, and completely wrong. The right answer is 50.

The numbers are points won by Arsenal in half seasons since Mikel Arteta became manager.

What do you think is the next number in this sequence: 30, 43, 33, 35, 40, 51, 46, 55, 44, 34, 35, 41, 51?  It is longer and more varied than the first. Any answer from the middle 30s to the middle 50s might seem plausible. All of them are wrong. The right answer is 29.

Those are points won by Liverpool in half seasons since Jurgen Klopp became manager.

Before this season started I did not hear anyone predict that after 19 games Arsenal would be top and Liverpool ninth. Now I can hear explanations for both turnarounds. Some of them might be partly or even wholly correct. The reality, though, is that both developments were highly unusual.

Arsenal won 34 points in the second half of last season. I compared points won by teams in the second half of one season and the first half of the next. I did this for 26 pairs of seasons – from 1995-96 and 1996-97 to 2020-21 and 2021-22. No team with 34 points in the second half of one season gained more than 37 in the first half of the next. Two rose from 35 to 44 and one from 36 to 45. Only three teams from anywhere in the table improved by Arsenal’s 16 points or more.

Leicester transformed from relegation escapees in 2014-15 to champions in 2015-16. By how many points did they improve from the second half of 2014-15 to the first half of 2015-16? Eleven.

Liverpool, though, have experienced a similar decline before. In the second half of 2013-14 they won 48 points then in the first half of 2014-15 they won 28 points, a drop of 20. Liverpool deteriorated by 22 points between the second half of last season and the first half of this season. Only one team in my study lost that many points. Leicester dropped by 22 points from the second half of their title season to the first half of the next season.

Sign up to emails from Racing Post Sport and get all the latest news and tips

Today's top sports betting stories

Follow us on Twitter @racingpostsport