Everyone can see West Bromwich Albion are a different proposition this season. Matt Phillips, James Morrison and Nacer Chadli have been turned loose to provide an attacking verve no Tony Pulis side has ever seen, and the Baggies fans who barracked the manager last year are now singing his name very loudly indeed. Eighth place would match their highest top-flight ranking in 38 years, and even relegation form down the stretch would get them to 50 points, their biggest total ever in a 38-game season.
Traditional Pulis teams have a clear statistical profile. Because there’s been such a big change, we’d expect it to show up in the stats. It turns out some things have changed, but a lot of things really haven’t, and a close look at the numbers leads to some fascinating findings.
Let’s first talk about set-pieces, a Pulis speciality. Any less reliance on the set-piece? On the contrary, West Brom are excelling themselves this season, having already increased their output from 12 to 16 or 17, depending on who’s doing the counting. That’s with more than 30% of the season still to go. They’re doing the same thing but better.
Pulis sides are also known for letting the other team have the ball. Stoke City famously went eons in the Premier League without more than 50% possession in a game. So it’s no surprise that last season, West Brom had the lowest possession in the league at 42.2%. A further note: they were last in the category both home and away: 44.3% and 40.1%.
So do we see higher possession figures this year? N. In fact, their possession is actually lower, at 40.6%, with home and away both last again on 40.8% and 40.3%. Note that the entire drop in possession has come at home; possession away is actually up slightly.
Let’s move on to pass completion percentage, another measure on which Pulis teams have historically scored low. Again, no surprise last season as West Brom were last in the league at only 70.0%. This year? They’ve moved up to 18th, ahead of Leicester and Burnley, but they’re at the exact same 70.0%. Completion percentage is up very slightly at home, less than one percent, which doesn’t seem significant.
So they’re still relying on set-pieces, they don’t have more possession, and their pass completion numbers are static. Perhaps they’re using the famous Pulis long ball less often? Last year, 21.7% of their passes were long. This year, even more at 23.4%. What’s the deal?
The first thing to note is that those pass completion numbers look a bit better now. If you’re making more long passes, and your completion percentage is the same, you’re passing more accurately.
Moreover, as we’ve already seen, there’s been a significant drop in possession at home. That usually means more counter-attacking, and long passes are an integral part of counter-attacks. So here we strike gold. Last year 19.7% of Albion’s passes at home were long, this year 22.5%. A 2.8% increase may not sound like much, but it averages out to roughly nine more long balls per match.
Those nine more long balls aren’t going to the head of Salomón Rondón, either. In fact, at home he’s averaging about one fewer aerial duel per 90 minutes, 9.1 to 10.2. So there may be as many as ten more long balls per home match on the counter-attack, and that’s a huge change. There’s a change in outcome as well. Last year West Brom had exactly zero counter-attacking goals. This year they have three, all scored at home, and there’s still a chunk of the season to play.
So are there any similar style changes away from home? No. As noted above, possession is almost the same, 40.3% to 40.1%. Long balls are down a bit, 23.3% from 23.9%. And in fact they’re going long to Rondón’s head a lot more away this year, 10.2 to 7.5 per 90 minutes.
So we do have a clear style change at home that’s paying dividends. But when we look at the side’s overall shot statistics, we find something surprising. Last year Albion had 10.2 shots per game, this year 10.4 – not a significant increase. And their shots at home are actually down this year, 11.6 to 11.5. Shots on target are up overall, but only from 2.8 to 3.3, and the home increase is no greater than the away increase.
That means about 13 more shots on target for the season so far, only half of which have come at home. West Brom are, amazingly, only 18th in the league for shots on target. And yet they’re tied for eighth in goals, scoring at a dramatically increased rate both overall and at home. They already have 36 goals this year, two more than they had all last season. At home, goals/game have nearly doubled, from 1 to 1.8. Away the jump in goals is much smaller, 0.7 to 0.9. But overall their averages say they’ll score an incredible 19 more goals this season than last.
The answer, as you’ve probably figured out, is that their conversion rate is way up. Last year it was 8.9%, third worst in the league. This year it’s 13.2%, third best. For individual players, such conversion rate swings are common, but for entire teams, quite unusual. No other team has enjoyed close to that big a jump since Suarez and Sturridge ran wild at Liverpool.
So are the new counter-attacking tactics leading to significantly better chances, leading to a higher conversion rate? The stat to look at is expected goals per shot. Unfortunately we don’t have a home/away breakdown, but West Brom’s shots are better this season, averaging a roughly 1.5% increased chance of going in. That’s a decent but not exceptional improvement, and certainly not enough to account for the goal deluge. In fact, even with that boost, their shots this season have been slightly below league average quality.
Sticking with expected goals, we find that if they continue at the same rate, the side will have improved their attack by about five goals over last year. That’s a real achievement, but again well short of the 19 they’re headed towards.
So shot totals haven’t changed dramatically, shots on target are up a bit, shot quality is up a bit, expected goals are up somewhat, and West Brom are tearing the house down. The conclusion is inevitable: as effective as their transformation has been, the Baggies are riding their luck.
This is shown most clearly in shooting percentage, the percentage of your shots on target that go in. It’s a stat that tends to regress strongly to the mean, so outlying numbers will always be suspect. We’ve already seen that West Brom are near the bottom in shots on target, but top half in actual goals. Right now 42.1% of their shots on target are going in, which if it holds would be the highest recorded in the eight years for which we have stats.
This is not in any way to lessen West Brom’s achievement. Tony Pulis has instituted a clear plan and made it work, and how many managers can say that?
Better still, the side has become much more enjoyable to watch. And their defence hasn’t been hurt one bit: they’re allowing roughly the same number of goals as last year, and doing a bit better than that at home. Any successful season needs a dose of luck, and if right now everything is breaking their way, fantastic.
Next season might be a disappointment, though; that’s the way these things usually work. One feels duty bound to say this year looks like an anomaly. But over the years the one manager who has consistently beaten the numbers is Tony Pulis. As a stat lover, I can’t wait to watch him try it again.