Premier League odds 23/24: What does the data say might happen next season?

sportinglife.com

31 May 2023

Premier League odds 23/24: What does the data say might happen next season?

Underlying data covers a wide range of potential data points, with expected points (xP) the best in assessing just how good a team has been in any given season.

That is because xP is the calculation of the number of points we would expect each team to collect based on chances created and conceded in each game, then each game is added together to get a final total.

For context and more understanding, let's use Brighton 1-2 Crystal Palace from a few years ago, a game that ended with an xG total of 3.03 - 0.27 in the Seagulls favour.

We calculated a home win as having a 91% chance, a draw 7%, and an away win 2%, so Brighton earned 2.8 xP from 0.91 x 3 (three points for a win) + 0.07 x 1 (one point for the draw). Palace earned just 0.13 xP from 0.02 x 3 + 0.07.

This metric is a great way to see which teams were perhaps fortunate or unfortunate in the league table, and there is a fascinating trend that I have identified which could give us a good guess as to what may happen to those teams next season.

Overperforming your xP by 10 or more in a season does not bode well for the next campaign.

Since Infogol started collecting data in 2014, there have been 23 occasions where a team has overperformed their xP tally by 10 or more, and 20 of the 23 (87%) teams saw their league position worsen the following season. In fact, only four (17%) of those sides improved their actual points tally the season following an large overperformance, and only two of 23 (9%) improved their league position.

That is an incredibly high percentage of teams that suffer a serious dip - or regression.

On average, teams who overperform their xP by 10 or more finish 13.4 points and 3.3 positions worse off the following season.

So, who falls into that category this season? Who could be affected as we head into 23/24?

Well, there are three teams; Arsenal (overperformed xP by 13.6), Manchester United (12.7) and Fulham (11.0).

Mikel Arteta's side pushed Manchester City hard this season, but have been a little fortunate to have accumulated so many points. Arsenal have featured twice previously on the list of overperformers since 2014.

In 16/17 they finished fifth having overperformed their xP by a wide margin, and the Gunners fell a place and 12 points the following campaign. More recently, in 18/19, they finished fifth again, though fell further in 19/20, slipping three positions and 14 points worse off.

A repeat of regression could occur given the positive variance the Gunners have seen throughout 22/23, and so the 5/6 about Arsenal finishing outside the top four in the 23/24 season is worth a look.

Obviously they could slip from second to fourth and scupper this bet, but given all of Liverpool, Newcastle and Chelsea will be stronger again next season, and Brighton are expected to continue to be a threat, Champions League football perhaps isn't so cut and dry for the Gunners.

Brighton - 1/4
Tottenham - 1/3
Chelsea - 8/15
Newcastle - 8/13
Manchester United - 4/6
Arsenal - 5/6
Liverpool - 5/6
Manchester City - 8/1

Odds correct at 1845 (29/05/23)

The same can be said for Manchester United, who have outperformed their xP by 10+ a whopping four times in the eight season sample. They have finished lower the season following such positive variance on all four occasions, being an average of eight points and 2.5 positions worse off, so they could face a similar problem of regression to Arsenal.

Both sides will have to buck a worrying trend if they are to replicate their 22/23 finishes.

Finally we have Fulham, who this season have accumulated fewer xP than Leeds, Leicester and Everton. Any kind of regression, which the trend suggests is a near certainty, for the Cottagers would immediately put them in the relegation conversation.

Excluding any large overperforming team who finished the season in sixth or above highlights the concern for Fulham.

An overperforming West Ham in 15/16 finished seventh and dropped to 11th the following campaign. Burnley also finished seventh in 17/18 when seeing plenty of positive variance, but that ran out on them in 18/19 as they sunk to 15th.

If we go back further to 14/15, Swansea had a fortunate season in eighth only to drop to 12th a campaign later, while more recent examples feature Everton (20/21) and Leicester (21/22).

The Toffees overperformed their xP by 10.8 when finishing 10th in 20/21, and dropped like a stone to 16th position, picking up 20 points fewer. Leicester finished an even higher eighth only last season when picking up 10.4 points more than their xP tally suggested they deserved, and they were relegated in 18th.

Fulham could go the same way with regression hitting, and the historical data suggests that their price of 10/3 could prove big.

Luton Town - 2/5
Sheffield United - 10/11
Bournemouth - 2/1
Burnley - 9/4
Nottingham Forest - 5/2
Everton - 10/3
Fulham - 10/3
Brentford - 9/2
Wolves 5/1

Odds 1845 (29/05/23)

So, what about the flip side? Teams who underperform their xP by 10+ in a season, how do they fare?

Well, believe it or not, the sample is much smaller across those eight seasons of teams who have done so and stayed in the Premier League for the following campaign, in fact, there are just three teams who have done so.

All three have improved their points tally the following season, while two of the three improved their league position too.

Crystal Palace underperformed in 17/18 and saw their points tally increase by five but their league position drop a place the following year. Southampton also underperformed in same season and saw their points tally increase by three, gaining one league position.

Brighton were the biggest movers. They underperformed their xP by a whopping 20 in 20/21 when finishing 16th, and the following season they increased their points tally by 10 and their league position by a huge seven places.

The only team who fits this criteria in the current campaign are West Ham, who underperformed by 10.3 xP on their way to finishing 14th.

A positive regression could be on the cards for David Moyes' men, so 5/6 about a top half finish appeals.

Safer gambling

We are committed in our support of safer gambling. Recommended bets are advised to over-18s and we strongly encourage readers to wager only what they can afford to lose.

If you are concerned about your gambling, please call the National Gambling Helpline / GamCare on 0808 8020 133.

Further support and information can be found at begambleaware.org and gamblingtherapy.org.