In early March of last season, when West Bromwich Albion were flying, I did a statistical analysis of the side which reached two conclusions: 1) the Baggies had developed an effective counter-attack at home and 2) they were riding their luck, because an abnormally high percentage of their shots on target were going in, and their goal totals would probably drop.
Naturally number 1) was largely ignored and I was pilloried for number 2), because that’s what football fans do, and that’s fine by me. As it turned out, though, number 2) was correct. By the end of the season, their percentage of goals from shots on target had dropped from 42.1% to 34.5%, their conversion rate from 13.2% to 10.8%. Traditionalists will ascribe it to Tony Pulis’ famous 40-points malaise; analytics types will say the odds just pointed that way. But the results were the same.
So now it’s late October in a new season, West Brom are crawling instead of flying, and there are more numbers to crunch. It’s early in the season, but already some significant trends have emerged. NOTE: unless otherwise mentioned, these stats are all from before Saturday’s match against Man City. You’ll see why at the end of the article.
We start in the natural place, set-piece goals. In Pulis’ first full season, the Baggies scored 12 set-piece goals, a moderate number. Last year they had a remarkable 17 by March, and tailed off to finish with 20, two behind leaders Chelsea. So far this year West Brom have three set-piece goals, which would put them around 12-13, right about where they were the year before.
Now to possession, a fairly consistent stat for Pulis sides. Two years ago it was at 42.2%, with an expected split of 44.3% at home and 40.1% away. But something interesting happened last year: not only did overall possession drop to 40.7%, but the split between home and away dropped to 41.3%/40.1%. The split was even tighter in early March.
At that time, we saw that the drop in possession at home was accompanied by an increase in long passes, but a significant reduction in long aerial balls. This suggested that the side was using the long outlet ball for counter-attacks, and in fact West Brom had scored three counter-attacking goals at home, compared to zero the year before.
Since that approach was so effective the previous year, you’d expect it to continue. The team is currently at 40.7% possession, same as last year, and possession at home is actually lower than possession away, 39.5% to 41.7%. Plus, long ball percentage at home is even higher than last year, from 23.4% in early March to 23.8% now. So far so good.
Accordingly, if the tactics were the same as last year, aerial duels by strikers would be low again. But they’re way up. Last March Salomón Rondón was averaging 9.1 aerial duels per 90 minutes at home, and finished the season with 10.5, which suggests the counter-attack was already slipping at the end of last year. This year before last weekend, he was averaging a sky-high 15.8 at home. The figure dropped after the game against Man City, where he had only three – although Jay Rodriguez, now another long ball outlet, had nine.
In other words, the extra long balls aren’t going as outlet balls on a counter-attack, they’re going to the heads of strikers. A big reason for the drop in outlet balls has been the play of Matt Phillips, spotty even when he’s been fit. He brought pace to the counter-attack, and without him there’s no clear target. He hasn’t started since the sixth game of the season. Yet against Watford the counter-attack still came through, on a long ball from Grzegorz Krychowiak to the feet of Rondón. It suggests that that approach can still succeed.
But right now it looks like the aerial bombardment hampers the attack, and not just at home. Even including the Man City game, Rondón is at 13.4 aerial duels per 90 minutes overall, and Hal Robson-Kanu is at 13.3. Last year they finished at 10 and 7.4 respectively.
One way or another, the basic attacking numbers have been sobering. Up to this weekend, West Brom’s conversion rate was at 8.8%, way down from the March high of 13.2%, and almost identical with the 8.9% of Pulis’ first year. Before Saturday, West Brom were averaging only 8.9 shots per game, compared to 10.5 at the end of last year and 10.2 the year before. Shots on target were at 2.6 per game, compared to 3.2 last year and 2.8 the previous year. It all points to a team that have dropped back to where they were the season before, or even lower.
And now here’s the pay-off. West Brom’s first nine matches, on which most of these stats are based, were against Bournemouth, Burnley, Stoke, Brighton, West Ham, Arsenal, Watford, Leicester, and Southampton. Looking at this list when the season started, you’d have said only Arsenal were out of their reach – and Arsenal are the most consistently vulnerable of the top six. The team’s stats have been compiled against ordinary opposition, and if Watford are significantly improved this year, it only goes to show that West Brom are not.
The City game offered some hope. The Baggies didn’t flinch against a superior opponent, and were even willing to press the big boys where possible. Maybe that’ll presage a revival. Last week mailboxer Rich, Cambridge sent in an excellent letter describing tactical ways out of West Brom’s doldrums. For Baggies fans and any other interested parties, I highly recommend it. And I’ll add: cut down the aerial balls to the strikers, and think fast counter-attacks at home.