<
>

Premier League's unluckiest: Liverpool, Martinez's Wigan, Benteke stand out over past decade

Luck matters more than you think, both in life and in soccer. There have been countless investigations into the role of fortune in football. German researchers found that luck was the most important factor in deciding which team won a given match, while a bunch of other findings, summarized in the book "The Numbers Game," suggest that the underdog is likely to win way more often in soccer -- just south of 50% of the time -- than in any of the other major sports.

In a 2018 study, another German researcher, Martin Lames, discovered that 47% of the goals scored across a pair of Bundesliga and Premier League seasons contained some element of luck. Maybe more staggering, the first goal in the match -- arguably the most important goal -- was most likely to be affected by chance.

- Coronavirus cancellations and reactions across sport
- Hunter: Coronavirus might change football forever
- Karlsen: How clubs are staying sharp during shutdown
-
Stream ESPN FC TV on ESPN+ Monday-Friday for the latest news

This isn't necessarily a modern discovery. Back in 1974, I.D. Hill wrote the following in the Journal of the Royal Statistical Society: "I find it difficult to imagine that anyone, who had ever watched a football match, could reach the conclusion that the game was either all skill or all chance. That both skill and chance are involved seems too obvious."

While Hill was right, we do at least now have some modern tools to help better determine both how lucky a team or player was and how repeatable their performances have been. Mainly, we have expected goals, which determines the likelihood that a given shot will end up in the net based on a variety of factors (distance, angle, nature of pass, etc.) and a historical record of how often similar shots have been converted.

So, as we're all staying at home and social distancing while most football is suspended, it felt like the right time to look at the players and clubs who have been worst hit by the whims of randomness, and award them a retroactive, imaginary honorific. Consider it a celebration of the guys who should have done better than they did and who deserved more than they got.

*The data we'll use from TruMedia extends back to the start of the 2008-09 Premier League season.

Unluckiest Result: Crystal Palace 0-4 Sunderland, 2017

The biggest gap in goals to expected goals came in a 9-1 victory for Tottenham over Wigan in November 2009. Despite the presence of Luka Modric and Gareth Bale on the roster, Spurs sported a starting midfield of Tom Huddlestone, Wilson Palacios, Aaron Lennon and Niko Kranjcar. The star, though, was Jermain Defoe, who scored five goals. Somehow, the score was only 1-0 at half-time.

Despite the nine goals, Spurs created 1.97 expected goals. They weren't lucky to win -- Wigan created only 0.51 of their own -- but the extremely lopsided scoreline wasn't a true reflection of the balance of play. However, since a win was never really in the cards, Wigan don't really deserve to earn this distinction.

Instead, the unluckiest result in the modern era of the Premier League goes to Crystal Palace, who in February 2017 lost 4-0 to Sunderland despite creating 2.76 expected goals and conceding just 0.88. The total "luck gap" as we'll call it -- the combined amount they underperformed at each end of the field -- was 5.88 goals. Only Wigan (6.54) were farther away from their much-more-likely parallel reality.

The cast of characters involved befits the bizarre outcome, too. Sam Allardyce was managing Palace, while David Moyes sat on the Sunderland bench. Moyes' starting lineup included John O'Shea, Bryan Oviedo, Jack Rodwell, Sebastian Larsson, Adnan Januzaj and Defoe (him again!). The bench featured Darron Gibson, Fabio Borini, Joleon Lescott and Steven Pienaar. Palace's lineup wasn't so heavy on "oh-I-remember-that-guy" figures but the bench had three former French internationals: Loic Remy, Mamadou Sakho and Mathieu Flamini.

Christian Benteke took eight shots for Palace all by himself; Sunderland took 10 shots ... total. The Belgian generated 1.33 xG, nearly half a goal more than the entire Sunderland team. Defoe, meanwhile, scored twice, because of course he did. Despite this gift from the gods, Sunderland still finished in last place and haven't been back to the Premier League since, while Palace finished 14th. Benteke ended the season with 15 goals, as did Defoe.

Unluckiest Shooting Season, Individual: Christian Benteke, 2017-18

Perhaps Benteke's performance against Palace was a precursor of what was to come. His 17-18 Premier League season was the worst finishing season in what we'll call the expected goals era. He got on the end of chances worth 11.8 expected goals, but only three of his 60 shots ended up in the net. That's a 5% conversion percentage; league average is more than double that.

- Stream ESPN FC, 30 for 30 Soccer Stories and much more on ESPN+
- A letter from the great quarantine: Missing football, pasta and friendship

The year before, Benteke also trailed his expected goals (15 goals to 18.06 xG) but in his five combined Premier League seasons before the infamous 2017-18 campaign, he was actually slightly ahead: 66 goals on 64.35 xG. After 17-18, though, he lost his place in the Palace starting XI and his shooting boots still haven't come back: he's scored two goals on 7.33 xG over the past two years. Now, his finishing hasn't been good, but it hasn't been this bad, either. These are all the attempts he's taken since the start of 2017, scaled by the quality of chance. You can figure out what color the goals are:

Despite all that, Benteke doesn't hold the record for most total goals missed over a Premier League career. He's only third, at 12.49. Second is Manchester City's David Silva, one of the league's best-ever players, who's scored 57 goals on 70.9 xG. But sitting in first is Hugo Rodallega, who found the back of the net only 29 times despite amassing 44.93 xG, for a whopping discrepancy of 15.93.

To bring it full circle: Rodallega was in the starting lineup for Wigan when they lost to Tottenham 9-1. And to bring it back to the present: At the time of publishing, Sheffield United's David McGoldrick is setting the pace for most xG (7.07) without a goal. The record in a 38-game season is Stewart Downing's 2011-12 campaign (4.47 xG) with Liverpool, and behind that it's... Stewart Downing's 2008-09 season with Middlesbrough (4.26). The current Premier League season must be finished for no other reason than to give McGoldrick a fair shot at erasing his name from this list.

Unluckiest Shooting Season, Team: Liverpool in 2011-12

Ah yes. The 2011-12 season was supposed to be the implementation of a grand vision. Liverpool's new director of football, Damien Comolli, was building a team based on "Moneyball"-esque principles. He wanted players who recovered possession in the final third, whipped the ball into the penalty area and won header after header. He used data to identify transfer targets with these characteristics -- Charlie Adam, Andy Carroll, Luis Suarez, Stewart Downing and Jordan Henderson -- and then it ... didn't work.

It turns out that a heavy reliance on crossing -- especially crosses that leads to headed shots -- is an incredibly inefficient approach. In other words, the data was right, but the conclusions were wrong. The team finished eighth, Comolli was out of a job by April 2012 and the manager, club legend Kenny Dalglish, left a month later. Adam, Carroll and Downing would all be gone soon, too, but who knows how things would have turned out with perhaps a little more luck.

Liverpool created 72.40 xG but scored only 47 times. That's a gap of 25.4 -- no other team in the dataset has experienced anything above 20. The only players to score more goals than their xG were Steven Gerrard, 32-year-old Craig Bellamy and Sebastian Coates, who scored his only goal for the club on that thunderous scissor kick against QPR. In addition to his finishing woes, Downing somehow created 56 chances for teammates, and not one was turned into a goal.

Beyond Downing, both Carroll (4 G to 11.62 xG) and Suarez (11 to 18.29) scored more than seven goals less than their expected number, while Dirk Kuyt wasn't much better (2 goals compared to 8.29 xG). At the time, these were three international-calibre players, falling on their faces, time after time after time. It seems unlikely that it was totally random, and the tactics themselves would feel like something from the Stone Age today, but this extreme degree of underperformance is unheard of and unsustainable. Over the next two seasons, Suarez would go on to score 54 goals on just 42.45 xG.

Had Liverpool just converted their chances at an average rate in 2011-12, they likely would've finished in the top four, and both Comolli and Dalglish would've stayed another season. Hell, maybe they win the league the next season or the season after. Perhaps Carroll and Suarez become a legendary, beloved striker pairing that revolutionizes global tactics. Or maybe it still all falls apart, and the club's timeline gets delayed by a year, they never challenge for the 13-14 title, and their managerial search just doesn't happen to sync up with the availability of Jurgen Klopp. The 11-12 season was painful viewing for Liverpool fans, but without it, they might not be where they are today.

Unluckiest Defensive Season, Unluckiest Overall Season and Unluckiest Overall Club: Wigan from 2010-13

Listen, I try to be objective. I try to provide historical context. I try to use data. I try to think of things holistically. But man, if there's one team that I'm convinced made some kind short-term-benefit-but-long-term-ruin, sell-your-soul kind of deal with a guy in a cloak or some devious, horned half-goblin-half-human who runs the underworld, it's Wigan.

They won the 2013 FA Cup Final against a Manchester City team that had Yaya Toure, David Silva, Sergio Aguero and Carlos Tevez in its starting lineup. And they did it the same year they were relegated from the Premier League after an eight-year stint in the top flight. They haven't been back since.

What's more amazing, though, is that they weren't relegated sooner.

As I wrote recently, they're not only one of the few teams who escaped the relegation zone over the final 10 weeks of the season; they did it two years in a row. The first escape was in 2009-10, when they allowed 79 goals despite conceding just 43.38 expected. That gap of 35.62 is the biggest on record. Same goes for their actual goal differential compared to their xG differential. Their actual goal differential was a league-worst minus-42, compared to an expected differential of minus-9.76. Overall, that's a gap of 32.24; no other team has ever been above 26.

When teams are that unlucky, they almost always get relegated. Somehow, Wigan didn't, and it likely speaks to the unique, attacking-oriented, bottom-of-table tactics Roberto Martinez was employing at the time. And of course, we're describing a very specific, kind of inspiring version of luck here: The unluckiest teams are the ones who had one unfortunate season on either the offensive or defensive end, dropped out of the league and never came back.

The same goes for individual players; Benteke is able to miss so many good chances only because he kept getting on the end of those chances in the first place and because his managers saw fit to continue giving him playing time despite the fact that he kept missing chances. Meanwhile, the only reason Wigan were able to rack up a Premier League-worst xG-to-G deficit of 69.29 over their final five seasons in the top flight is that despite all the bad luck, they were still able to do just enough to get by.