Why It's Not Too Early to Start Talking About Relegation or the Title Race

The Premier League season is just three games old, but a number of teams have already seen their expectations shift


Last year, Arsenal played Chelsea on the second weekend of the season. The transfer window had recently closed, both teams were just a game into new eras under first-year managers. Whatever happened in that match would be a write-off, a confrontation between two nascent forces, two teams that wouldn’t be true versions of themselves for at least another month, if not more.

The way the game played out seemed to prove the point: The two sides took a combined 39 shots, and the two halves of the match had nothing more than a stadium and a day of the week in common. Over the first 45 minutes, Arsenal took 12 shots to Chelsea’s nine -- a slim margin -- but the Gunners created five Big Chances to the hosts’ one. Here’s Opta’s definition for Big Chances, by the way:

A situation where a player should reasonably be expected to score, usually in a one on one scenario or from very close range when the ball has a clear path to goal and there is low to moderate pressure on the shooter. Penalties are always considered big chances.

After halftime, Chelsea took 15 shots to Arsenal’s three -- a dominant second 45 that secured a 3-2 win. Two unsettled teams produced a match with no clear patterns or conclusion. Except, that match ending up really mattering. Chelsea finished the season in third, securing a Champions League spot, while Arsenal finished in fifth, just two points back of their opponents from the second week of the season. After two games, Chelsea had six points and Arsenal zero. Over the final 36 games, Unai Emery’s team took 70 points, while Maurizio Sarri’s side claimed just 66. Arsenal were the better team for 95 percent of the season, but it didn’t matter, in large part because of that loss to Chelsea.

We’re only about eight percent done with the current Premier League campaign, but every year there seem to be a couple of early-season results that have an outsize impact on the way the table looks come May. So, let’s take a look at how each team’s betting-market projections have changed after three weeks. Who’s gained the most, who’s stayed the same, and who’s started to slip down the table?

The Big Leapers (plus-3)

-Burnley: 39 points to 42 points
-Norwich: 36 to 39
-Sheffield United: 33 to 36

Three teams that came into the season with a legitimate chance of being relegated, three teams that have already lessened the likelihood of their own relegations as much as they could ever have been reasonably expected to at this point in the season!

Despite matches against Arsenal and Wolverhampton, Burnley have eked out a positive expected-goal differential (plus-0.4) through three games that have also produced four points. Norwich thumped Newcastle, 3-1, a week ago, and although they lost both of their other matches, they held their own on the road against Liverpool and at home to Chelsea. Even if he’s the Nordic Messi, Teemu Pukki won’t keep scoring at this rate -- four goals on 1.76 xG -- but north of 4 shots per 90 minutes is a hell of a clip so far. Sheffield, meanwhile, have increased their expected-point total by the largest percentage (11) of any team so far. With three goals scored and conceded, their underlying numbers (4.3 xG, 2.8 xG) are a bit better, and their only loss so far required one of their opponents to revive the ancient practice of alchemy:

Chris Wilder might be my favorite new manager in the league. The fans applauded the team’s effort in the 2-1 loss to Leicester, but here’s what he said to the BBC after being asked why he wasn’t happy with the performance:

"I'm not going to applaud, credit or reward the players because they run around. Whatever the supporters do after is up to them but for me it's a normal, natural thing to do and if they don't they won't play.

"I'm not here to appease fans. I walked through the door three-and-a-half years ago and set our standard, if you keep giving the ball away you allow good players to gain momentum.

"We tried to gain momentum but we turned the ball over so cheaply it was frightening. We didn't play quick enough. It's disappointing and bewildering to see that happen for a team like us who are really good with the ball."

Speaking to BBC Match of the Day, the clearly frustrated manager added: "I'm not pleased with effort because that is just a bog standard thing for me. Cheers for effort doesn't tick a box for me.

"I didn't think we deserved anything. If we keep giving the ball away to good players we are going to create bad situations. In both boxes they were better than us. There is a lot for us to work on."

Love this guy.

The “Hey, Good for You” Guys (plus-2)

-City: 93 to 95
-Liverpool: 86 to 88
-Arsenal: 68 to 70
-Leicester: 55 to 57
-Crystal Palace: 42 to 44
-Brighton: 38 to 40

Last season, 12 different spots in the table were separated by two points or fewer. Point is: these improvements, coupled with run-of-the-mil, expectation-meeting performances for the rest of the season, might already be enough to make a difference in the end-of-the-year standings. 

City so thoroughly dominated the one game that they didn’t win, so the markets didn’t seem to care about two dropped points at home. Their xG differential is plus-6.1; no one else is above 3.5. Liverpool, of course, are the only undefeated team remaining in the league. They averaged about 2.55 points per game last season, and this year they’re at 3.0 -- you can do the math there. Arsenal just got turned into roadkill by Liverpool, but they’re also ahead of last year’s points-gathering pace (1.84 up to 2.00). (Also: I bet a friend lunch that Arsenal would lose by at least three goals to Liverpool. He doesn’t want lunch; he wants a mention in the newsletter. So: shoutout to Dan.) Leicester haven’t performed particularly well in any match yet (2.3 xG, 2.5 xGA), but five points from a schedule that’s already included matches against Wolves and Chelsea is enough to turn them into the favorite for seventh place. And here’s where Palace’s point total picked up that extra heft:

As you, a reader of this newsletter, may know: I thought the markets were too low on Brighton to start the year. They were projected to be in the relegation zone, but with a talented roster and an accomplished, unique manager in Graham Potter, they seemed more like a mid-table team to me. This past weekend’s loss to Southampton is a write-off -- Florin Andone picked up a red card less than 30 minutes in -- but in the games against West Ham (draw) and Watford (win), Brighton conceded just 19 shots and and edged the possession battle in both. Last season, they conceded 15.3 shots per game and averaged just 44.1 percent of the ball.


-Bournemouth: 44 to 45
-Newcastle: 37 to 38
-West Ham: 50 to 50
-Southampton: 45 to 45
-Tottenham: 74 to 73
-Chelsea: 67 to 66
-Everton: 56 to 55
-Aston Villa: 40 to 39

These teams, more or less, have done nothing to suggest that they are not who we thought they were. Tottenham’s drop, combined with Arsenal, Liverpool, and City’s improvement, however, suggests that Spurs are much closer to the race for third/fourth than the race for the title. Nabbing a point away to City softens the blow of blowing one at home to Newcastle. The Magpies’ win at Tottenham is almost canceled out by their loss to Norwich a week ago.

As for the others, Chelsea’s one-point dip is probably the most confusing. They’ve got a minus-4 goal differential and a minus-0.4 xG differential through three games. I still haven’t seen enough from Frank Lampard’s team yet -- beyond a vague devotion to various tactical buzzwords, like “tempo” and “pressing.” But I guess two road games against Norwich and United and a home fixture against Leicester is a little bit tougher than it seems. Five points from those three would’ve probably kept them where they were.

You Know and I Both Know Why You’re Here (minus-2)

-Manchester United: 69 to 67

If they convert two penalties, United might have seven or even nine points, and they’d be in one of the earlier sections. The status quo would be maintained, or the expectations would have been jolted up a notch or two. Instead, Paul Pogba and Marcus Rashford both missed ... and United drew away to Wolves and lost at home to Crystal Palace. The Wolves draw and the opening-week Chelsea win were actually an improvement on last year -- they took a single point from the corresponding fixtures in 18-19 -- so maybe that’s why the Palace loss only knocked two points off the docket.

United have the second-best xG differential in the league (plus-3.5) and the third-best actual G differential (plus-3). The issue, though, is how tenuous their process currently seems. According to Stats Perform data, Paul Pogba leads the team in final-third entries with 28 and key passes with seven. He’s their best creator and their best ball-mover -- a heavy load for anyone to carry, and a dynamic that can be destroyed by an injury, a transaction, or a savvy gameplan. Plus, nearly 40 percent of United’s xG this season has come from the penalty spot. The relevant question isn’t who should take spot-kicks; no, it’s whether or not United can keep drawing them at such a high clip. (Hint: they can’t.)

Hear Me Out: What if ... Offense Is the Best Offense? (minus-3)

-Wolverhampton: 52 to 49

The big question for Wolves before the season was “Can you guys actually put the ball into the back of the net?” By xG allowed, they already had a top-four quality defense last season. It’s just that ... they only scored 47 goals in 38 games. To break into the Top Six or beyond, you simply just have to score more goals. Without an all-time great defense, 1.25 goals per game just isn’t enough to accrue the required points total for a place higher in the table.

And well, Wolves are currently undefeated, but they’re yet to win and they’ve scored a grand total of two goals in two games: one on a three-percenter half-volley from Ruben Neves and the other on an injury time penalty against Burnley to salvage a draw at home. They’ve conceded two and allowed two. The best possible outcome from those numbers in three games would be two one-goal wins and a two-goal loss, while the worst is either a win and two losses or three draws. As we’ve seen so far, the floor and the ceiling are never too far apart.

Oh ... Oh No ... Oh God No (minus-6)

-Watford: 45 to 39

Nick Dorrington wrote this about Watford in his season preview for Statsbomb:

Through the first half of the season, Watford ran an average xG difference of 0.19 per match, the sixth best in the division, and superior to the figures of both Arsenal (0.14) and Manchester United (0.11); during the second half of the season, they had an average xG difference of -0.51 per match — the fourth worst, verging on relegation numbers. Their attack was pretty consistent throughout, but their defensive numbers plummeted. Over the final 10 matches of the campaign, they averaged nearly two xG conceded per match.

There is no immediately apparent reason for that defensive drop off. Watford suffered no major injuries during the second half of the season, and neither did coach Javi Gracia rotate especially heavily around the club’s FA Cup run. Watford handed a league-high 10 players 2000 or more minutes of action over the course of the campaign, although they weren’t unique in that — five other clubs also did so. But Watford undeniably altered the way in which they defended as the season went on. They began to defend closer to their own goal.

Seventeen of the 20 Premier League teams have between three and six points right now. City have seven, Liverpool have nine ... and Watford have zero. Poor Elton John; the first three games of this season have been a lot like the final 19 of last year.

Javi Gracia’s team is defending higher up the field again -- third-highest pressing rate in the league (defined by passes per defensive action) -- and despite only one goal so far, the attack has been solid, as they’re eighth in xG (4.3) and have completed more passes within 20 yards of the opposition goal (30) than everyone but Manchester City. However, the defense is struggling once again, as they’ve conceded the fourth-most xG (5.4) and have done so against a soft schedule that’s yet to include a team that finished higher than eighth last season.

Watford have certainly been unlucky -- their G and xG differential have the widest negative gap in the league -- but the formula for relegation usually looks something like “unlucky + inadequate”. The rise of the likes of Norwich, Burnley, and Sheffield certainly doesn’t help, either. Before the season, the markets projected Watford to finish tied for 11th. After only three games, they’re projected to finish tied for 18th.