In Games Against Better Teams, Arsenal are Average; Liverpool are Elite

Arsenal are still a threat, but you cannot doubt that Liverpool have been a lot better

Against the better teams, Arsenal just about edge the underlying numbers.

Against the better teams, Liverpool absolutely blitz the underlying numbers.

Take a look, from a sample approaching 20 games each:

The gulf is literally ten times.

This is perhaps pertinent as Liverpool go to Aston Villa and Man City within four days, even if the previous week’s exertions make this even more difficult than normal. (I’m not expecting six points.)

In the 17 matches so far against the best teams, in terms of actual results, Liverpool took a staggering 40 points out of 51 (against an expected 43 points based on xG); Arsenal, by contrast, took a respectable 32 points (from a mediocre expected 24).

So either Arsenal have got lucky to get just 32 points, or they somehow deserved to beat the xG model in some other way (and set-pieces could be the answer).

The thing in football is that previous performance always means a lot, as a sign of what’s possible and how consistent a team can be; but it can also change. Man City prior to Rodri’s injury looked great; as soon as he was injured, they struggled. But also because it exposed other issues within the squad.

---

I wrote the piece below yesterday on the run-in:

A Funny Story About Arsenal Closing the Gap on Liverpool, and More

---

Judging Liverpool after the physically and emotionally draining Mersey derby is not judging a team in normal circumstances. That game was brutal, and refereed completely unfairly, and out of keeping with how the league is normally refereed. (What next? Will they introduce a new ball at Aston Villa?) And as I noted yesterday, Liverpool’s run will be nine away games from the last 14 in the league by the weekend.

Liverpool’s games against best 17 opponents:

Arsenal’s games against best 17 opponents:

As you can see, almost all Arsenal’s best xG performances have been at home, whereas for Liverpool it’s more mixed. Inter Milan are the only good side Arsenal have gone away to and had more xG than they conceded.

Liverpool away to good teams have been superb on the underlying numbers, without always getting the luck (such as at Forest). But Liverpool’s balance of home vs away xG Difference shows another form of consistency that, so far, have Arsenal have lacked.

I wrote most of this piece before Arsenal, who had ‘won’ the xG in the first League Cup semifinal leg against Newcastle but still lost 2-0, then lost on both metrics and scoreline in the second leg. My prematch hunch for that second leg, as noted on the site when I saw an unchanged XI, was post-City hangovers and hammies for the Gunners.

While Arsenal dominated the first leg, in the second, when they needed to win, they yet again created only around 1xG, which is around their average against better teams.

I looked at the league and Champions League games against the best and better teams for Arsenal and Liverpool, and ended up with 17 games each. (And in general, Liverpool have taken nine more points in the League + Champions League so far, and won more cup games against top sides.)

I picked what were then the top eight in the Premier League, other Big Six teams, and the best of the European opponents.

While I use FBRef’s Opta data, the xG for cup games is harder to find, and can vary, so I haven’t included those as it would be inconsistent data. Arsenal faced Man United (lost on penalties), Newcastle, (lost), and Newcastle again (lost again).

Arsenal had a fairly big xG win at home to Newcastle, but a fairly big xG loss away; Liverpool narrowly lost the xG away at Spurs (due to a player not being sent off and scoring while the player he fouled was off the pitch), and thrashed them in the second leg, on all metrics.

In the 17 toughest games, Arsenal have only created a better xG Difference than +0.9 on one occasion: Manchester Utd at home in December, +1.9. (While Arsenal’s one-off best of +0.9 has also been reach three times by Liverpool.)

Liverpool have reached or surpassed +1.3xG Difference in a game against the better sides some eight times:

1. Spurs, Away +4.4

2. Leverkusen, Home +3.2 *

3. AC Milan, Away +2.5

4. Man City, Home +2.4

5. Lille, Home +2.0

6. Manchester Utd, Home +1.8

7. Notts Forest, Away +1.7

8. Real Madrid, Home +1.3

*This was matched by the 4-0 League Cup win over Spurs, but which is not included.

Liverpool’s worst xG ‘results’ in these games were the -0.2 at Newcastle and the -0.1 away at Arsenal, the two teams who slugged it out in the semis. (The only worse xG result this season has been the -0.4 at Everton, and the Dogs of Yore.)

Arsenal had five xG Differences worse than this, and all were away, in part hindered by red cards.

1. Aston Villa, Away -0.3

2. Atalanta, Away -0.4

3. Brighton, Away -0.6

4. Bournemouth, Away -1.1

5. Manchester City, Away -1.4

But of the 17 games overall, Arsenal only had +0.6 xG, which you can call a win on the underlying numbers, four times; compared to Liverpool’s 13.

While you don’t always have to beat the better teams to win the league, it helps. Liverpool also have far better numbers away from home than Arsenal.

As I’ve been noting, if you win a game by +2.0xG, you’re almost guaranteed a win.

Liverpool’s average against the 17 best sides, as it is against the rest of the sides, is around +1.4. Arsenal’s is +0.14, or ten times lower.

In the rest of this piece I’ll look at what tends to happen in tighter xG games, and then look at Liverpool’s underlying numbers against the lesser sides.

**The second half of this article is for paying TTT Main Hub subscribers only.**