Unbeaten,
Unlucky,
Underestimated

Let us begin at the beginning, which is to say let us begin with an act of administrative incompetence so thorough, so complete in its indifference to the actual purpose of attending a football tournament, that it remains, seventy years on, a kind of dark comedy that resists improvement in the retelling.

Scotland qualified for their first World Cup in Switzerland in 1954. FIFA permitted each squad to consist of twenty-two players. The Scottish Football Association, in the manner of a committee that has been handed the instructions for a self-assembly wardrobe and decided they look complicated, sent thirteen. Thirteen men to play two group games, in the summer heat of Switzerland, against Austria and Uruguay, the latter being the reigning world champions. The dignitaries of the SFA, meanwhile, had brought their wives, because a World Cup is, after all, a wonderful opportunity to see Switzerland in June, and the football is really just a backdrop to the occasion.

Andy Beattie was the manager. He resigned before the second game, citing his inability to actually manage the squad in any meaningful sense, which was fair enough in the circumstances, if somewhat belated. Tommy Docherty, who played in both matches and later became the most quotable man in Scottish football, recalled that the squad were "knackered by the end of the anthems" in their heavy wool shirts and over-the-knee shorts in the Basle sunshine, while Uruguay appeared in lightweight kit designed for South American humidity, as though they had thought ahead.

The data, to its credit, is not brutal here. Scotland's Elo rating of 1600 made them the weakest team in a group that included Uruguay at 1980 and Austria at 1870. The expected points return was 0.77. They got zero. The delta is minus 0.77, which is a marginal underperformance in a group where they had no business expecting anything at all.

Uruguay beat Scotland 7-0. Austria beat Scotland 1-0. The squad went home having scored nothing and conceded eight. The SFA held a debrief. There is no record of the wives being consulted, though their experience of Switzerland was presumably more enjoyable.

There is no great injustice to locate in 1954. Scotland were not equipped, literally or figuratively, for the occasion. The group was brutal, the preparation catastrophic, the ambition of those who organised the trip directed elsewhere. The data exonerates nobody. But it does confirm that what happened in Switzerland was not a sign of a football nation failing. It was a sign of a football administration that had not yet understood what it meant to participate.

In February 1958, four months before the World Cup in Sweden, the plane carrying Manchester United home from a European Cup tie in Belgrade crashed on the runway at Munich airport. Twenty-three people died. Matt Busby, who had been appointed to manage Scotland at the tournament, survived, but only barely, and only after weeks in which survival itself was in question. He would be in hospital until the spring.

Scotland went to Sweden without their manager. Dawson Walker, a selector, stood in. The word "stood" is doing heavy lifting in that sentence; Walker was not a tactician, not a motivator, not a man who had signed on to lead a national team through a World Cup. He was a committee man asked to do something committees are not designed for. The results reflected this.

And yet this is where the data becomes uncomfortable, because 1958 is the one campaign where the numbers offer Scotland no shelter at all. The group contained France and Yugoslavia, both ranked in the 1820s, and Paraguay at 1640. Scotland came in at 1710, which made them the third-strongest team in a group that was genuinely competitive but not impossible. The expected points return was 2.65. Scotland got one point, drawing Yugoslavia before losing to Paraguay and France. The delta is minus 1.65, which is a strong underperformance, and the data is not inclined to construct excuses for it.

Scotland finished bottom of the group, behind Paraguay. They scored three goals and conceded six. The tragedy of Munich was real and its shadow over the campaign was genuine, but it is a measure of intellectual honesty to note that the underlying quality of the squad was sufficient to have done considerably better, and did not. Some losses are systemic. Some losses are circumstantial. Some losses are just bad football. Scotland in Sweden, with all the compassion one owes to genuine tragedy, managed to produce the latter.

1958 is the aberration in the data. It is the campaign where Scotland were expected to compete and did not, where the margin between potential and performance was largest, where the numbers decline to offer the kind of contextual comfort they offer elsewhere. The data is many things, but it is not dishonest. Scotland underperformed at the 1958 World Cup. That is the honest account, and honest accounts are the whole point.

Jimmy Johnstone had, some weeks before the squad departed for West Germany, been found drifting out to sea in a rowing boat at Largs having mislaid the oars. This is one of those facts that appears, at first glance, to be too perfectly Scottish to be true, and yet it is true, and it is worth dwelling on, because it tells you something about the gap between what Scotland could have been in the summer of 1974 and what the world allowed them to be.

Johnstone made the squad. He was, in his prime, perhaps the most gifted footballer Scotland had produced since the invention of the position. He dribbled as though the ball were attached to his boots by a length of string he alone could see. And he went to a World Cup as part of a Scotland squad that, on paper, was genuinely capable of causing damage. Bremner, the captain, willed things into happening through sheer force of personality. Dalglish was twenty-three and already understood the game at a frequency most players never reached. They left Hampden in front of a crowd of thousands who believed.

Scotland were drawn in a group with Brazil, Yugoslavia, and Zaire. Their Elo rating of 1810 made them third in the group behind Brazil at 1960 and Yugoslavia at 1870, with Zaire, newly arrived on the world stage, at 1430. The expected points for Scotland were 3.24. They accumulated four. By any measure, Scotland overperformed.

They beat Zaire 2-0, which was the required result and nothing more. They drew Brazil 0-0 in a match that Billy Bremner described as the best Scotland had played in living memory, which tells you something about the quality of that Brazilian side and something else about how well Scotland contained them. They drew Yugoslavia 1-1. They were unbeaten.

The mechanism of their elimination was goal difference. Scotland needed to beat Yugoslavia by three goals in the final game. They drew 1-1. The margin they needed had been borrowed by the match against Zaire, where Scotland's 2-0 victory was not enough, because Brazil had beaten Zaire 3-0 and Yugoslavia had beaten Zaire 9-0 and the arithmetic simply did not add up in Scotland's favour regardless of how the football had actually gone.

More than ten thousand people met them at the airport on their return. Billy Bremner was photographed clutching his head in his hands after the Yugoslavia game, the reality of what the goal difference meant having just arrived. The photograph is one of the most Scottish images ever taken, and it depicts not a failure but a man processing an injustice that had no adequate response.

The data is unambiguous. Scotland overperformed their expected points by 0.76. They outplayed their Elo rating. They were eliminated by a rule, not by a result. Every other team in that tournament had been beaten at least once. Scotland had not. They nonetheless caught the plane home first.

Ally MacLeod was a man who believed. He believed in Scotland. He believed in the squad he had assembled. He believed, publicly and with the enthusiasm of a man who has not yet encountered the specific obstacle the universe has arranged for him, that Scotland could win the World Cup. The actual quote, often inflated in the retelling, was: "I honestly think that if Scotland play reasonable form at all, we will qualify and I think a medal of some sort will come." The myth became more than the man, as myths do, and it attached itself to an entire nation's sense of hope and tipped it, briefly, into something approaching delirium.

Thirty thousand people came to Hampden for the send-off. There was a novelty song in the charts. There was a bus around the stadium. There was, somewhere in the middle of it all, a football team who were about to encounter Teofilo Cubillas at the height of his powers, which is to say they were about to encounter one of the finest technical footballers South America had produced in a decade, on a day when he had decided to demonstrate exactly why that was true.

Cubillas scored twice against Scotland. The second was a free-kick struck with the outside of his right boot from twenty-five yards that flew into the top corner, changing direction twice in flight, a goal now regularly cited among the finest ever scored at a World Cup. Willie Johnston then failed a drugs test for a cold remedy and was sent home in disgrace. Scotland had lost 3-1 to Peru. The narrative was set.

But here is the number that dismantles the narrative. Peru, in that group, across their three games, performed to a delta of plus 1.79. They were expected, on the basis of their Elo rating going into the tournament, to accumulate 3.21 points. They accumulated five. Peru beating Scotland 3-1 was not Scotland having a bad day. It was Peru having the tournament of their lives, against everyone they played. Cubillas scored five goals in the group stage, including two against the Netherlands, who reached the final. Peru were an anomaly. Not Scotland.

And then came Gemmill. Against the Netherlands, with Scotland needing three goals to qualify and the odds stacked to the horizon, Archie Gemmill took the ball in the 68th minute and did something that will be replayed as long as football is played and footage exists to show it. He went through three Dutch defenders, dummied the fourth, and chipped the goalkeeper. 3-1. Scotland needed one more. They didn't get it. Rob Rensenbrink scored at the other end. Scotland went out. They drew their group, Scotland did, with the Netherlands, on points. They went out on goal difference. Again.

Scotland's actual delta in 1978 was minus 0.28. They were expected to accumulate 3.28 points. They got three. That is essentially as expected. That is a team performing to their level in a group where one opponent dramatically exceeded theirs. The humiliation of 1978 is largely a fiction constructed around a magnificent Peruvian anomaly and one extraordinary goal that arrived too late.

In the eighteenth minute of Scotland's second group game in Malaga, against Brazil, David Narey received the ball at the edge of the area and struck it with the outside of his right foot in a manner that suggested he knew exactly what he was doing. The ball went into the top corner. Scotland led Brazil 1-0. The BBC studio, watching from London, contained Jimmy Hill.

"A toe-poke," Hill announced, with the confidence of a man who had decided that the most appropriate response to a Scottish player scoring against Brazil at a World Cup was to diminish it before the echo had finished. The phrase passed into Scottish football mythology not because of what it said about Narey's goal, which was magnificent, but because of what happened next.

Brazil replied four times in the next thirty-four minutes. Zico in the 33rd. Oscar in the 47th. Eder in the 63rd. Falcao in the 67th. Each goal more emphatic than the last, each one a kind of aesthetic argument against Jimmy Hill's original thesis. By the end it was 4-1 and the camera had found Hill again, whose expression suggested he had perhaps reconsidered the wisdom of his editorial timing. The goal was thereafter known, in Scotland, as "the goal Jimmy Hill replied to." It is one of the finest pieces of collective retroactive comedy in the nation's sporting history, and it contains a serious truth: the goal was not a toe-poke. It was a 25-yard strike into the top corner from a man who knew exactly where the goalkeeper was and struck the ball over him with the outside of his right foot. Narey was a composed, intelligent defender for Dundee United who played that shot as though he had rehearsed it, because at some level he had. Hill was wrong. Narey knew.

What the goal could not change was the architecture of the group. Scotland entered the tournament at 1820, behind Brazil at 2030 and the Soviet Union at 1870 but ahead of New Zealand at 1400. They beat New Zealand 5-2, lost to Brazil 1-4, and drew the Soviet Union 2-2 in the final game, needing a three-goal win to advance. The expected points were 3.17. Scotland got three. Delta: minus 0.17. The data says nothing happened here except the expected thing.

Graeme Souness, Kenny Dalglish, Joe Jordan all played in that tournament. It was not a weak Scotland squad. It was a Scotland squad in a group where the level above them was simply too high for the margin to close, and the data confirms this. They performed as expected. The level was not enough. Those are different problems with different remedies, and it is worth being clear about which was which.

1982 is, in the long accounting of Scotland's World Cup history, a campaign without injustice or excuse. They played to their rating, the rating was not sufficient, they came home. The toe-poke was a beautiful goal. Jimmy Hill was wrong. Both of these things can be true simultaneously, and are.

Graeme Souness was standing outside the medical room at Ninian Park when Alex Ferguson arrived to find out what was happening. Souness was in tears. That told Ferguson what he needed to know. He walked into a dressing room full of players who had just qualified for a World Cup and he told them that Jock Stein had died. The photographs of that night show men in the particular posture of people receiving news they know they will carry for the rest of their lives.

The wreath from the players at the funeral read: "We will miss you, boss."

Ferguson took Scotland to Mexico. He took them to a group that, by the metrics, was the hardest group draw Scotland received in eight World Cup campaigns. Denmark were at 1880, West Germany at 1970. Scotland came in at 1780, Uruguay at 1740. The expected points were 2.47.

What the expected points could not have anticipated was what Denmark were about to do. Denmark's campaign delta in Mexico was plus 2.68, which is the highest overperformance of any team in any of Scotland's eight World Cup groups. They beat West Germany 2-0 in the group stage, which was not supposed to happen, and the consequence of that result was to make Scotland's already difficult group considerably more so. The group table Scotland played in was harder, in practice, than any reasonable reading of the four teams' ratings suggested it should have been.

Scotland lost to Denmark 0-1, lost to West Germany 1-2, and drew Uruguay 0-0. They accumulated one point. The expected return was 2.47. The delta was minus 1.47, which registers as an underperformance, and it was. But the shape of the group had been bent by Danish overperformance, and the context of what Scotland were carrying into those games was the weight of a dead manager and the grief of men who had not been given adequate time to process it.

None of that is an excuse. But all of it is true. Ferguson, to his credit, never offered those factors as mitigation. He was not that kind of man. He understood that football is played regardless of what has happened beforehand, and that the measure of a team is whether it can set things aside and compete. Scotland in Mexico could not quite do that. Given what they had set aside, and what they had been asked to set aside, the honest response is not condemnation. It is something closer to understanding.

Andy Roxburgh, in the week before Scotland played Costa Rica in Genoa, told a press conference that kick-off was "at a Costa Rican kind of time." It was an attempt to lower expectations, to suggest that whatever happened in that first game, conditions had been arranged in favour of the opposition. It was not a remark that aged well. It was not a remark that aged at all.

Scotland had nineteen shots to Costa Rica's four. Luis Gabelo Conejo, the Costa Rica goalkeeper, was a man briefly granted the particular superpower of being completely unbeatable. Juan Cayasso scored the only goal. Scotland lost 0-1. The following morning, the debate in Scotland was about whether this was the worst result in the national team's history, and various candidates were advanced and argued over with the kind of energy usually reserved for more important matters.

And then Costa Rica beat Sweden. And then Costa Rica held Brazil. And then the data, when it was eventually assembled with the benefit of sufficient distance, told a story that the morning-after arguments had not considered.

Costa Rica's campaign delta was plus 2.48. They were expected, on the basis of their Elo rating, to accumulate 1.52 points in that group. They accumulated four. That delta of plus 2.48 is the second largest overperformance of any team across all eight of Scotland's World Cup groups, behind only Denmark's +2.68 in Mexico four years earlier. More than Peru in 1978. Costa Rica in 1990 were not merely having a good tournament. They were producing one of the most statistically freakish three-week performances in the competition's history, against all comers, including Brazil, who reached the quarter-finals.

Sweden's delta in the same group was minus 2.83, a strong underachieve, the worst performance by a team in any of Scotland's eight groups. The team that Scotland beat in their second game, 2-1, to give themselves a chance of advancing, were simultaneously having one of the worst tournaments any team in Scotland's World Cup history had managed. The group was chaos. Not Scottish chaos. Universal chaos, arranged with particular cruelty.

Scotland lost to Brazil 0-1 in the final game, needing a result that the group's peculiar mathematics had made almost irrelevant. Costa Rica had already done the damage. Scotland's delta for the campaign was minus 0.98, which registers as underperformance, and it was. But the honest account of Italy 1990 requires acknowledging that Scotland were undone by a team whose overperformance was the second most extreme in the history of their World Cup campaigns, behind only the Denmark side that had dismantled West Germany four years before. That is not consolation. It is context.

The opening game against Brazil, at the Stade de France in Saint-Denis, was the kind of occasion Scottish football had been building towards for four years, and it began, almost against all expectation, with the suggestion that this might be different. John Collins scored from the penalty spot. 1-1. For a few minutes, in front of a full house and the watching world, Scotland were drawing Brazil at a World Cup.

Tom Boyd then turned the ball into his own net. Brazil won 2-1. Scotland's campaign delta for 1998 is minus 2.89, the worst figure in the entire dataset across eight tournaments. There are no contextual asterisks to place alongside it. There is no opposing team whose anomalous overperformance distorts the picture. There is no administrative catastrophe in the background, no resigned manager, no absent genius. There is just a Scotland team that was expected to accumulate 3.89 points and accumulated one, because they were poor, and the data does not offer comfortable alternatives to that word.

The draw against Norway yielded a point that temporarily kept hope alive. Then Morocco. 3-0. Craig Brown's squad, assembled with diligence and care but lacking the concentrated quality of the sides that had preceded it, was simply not equipped for a group that contained Brazil and, as it turned out, Morocco, who were considerably better than Scotland on the day that mattered most.

Scotland's Elo rating going into France 1998 was 1810, which made them the second-strongest team in a group containing Norway at 1780 and Morocco at 1700, with only Brazil at 2050 ranked above them. They should, on paper, have been competitive. They were not. The 28-year wait that followed was not a continuation of a pattern of bad luck or anomalous opponents. It was the consequence of a performance that the data, without pleasure and without ceremony, marks as the worst in the country's World Cup history.

There is a kind of respect in saying that clearly. The analysis that precedes and follows these paragraphs is designed to offer honest recontextualisation where honest recontextualisation is warranted. It is not designed to construct a hall of mirrors in which Scotland never did anything wrong. They did things wrong in France. Badly wrong. That is the honest account, and it ends here, with the bus home, and the long years that stretched out from Saint-Etienne.