“What else matters but beating him?”
The trenches of the First World War were not famous for rapid maneuvers, but one night late in January 1915, near Binarville in northeast France, was an exception. The Germans planned an attack, to be led by the thousand men of 3rd Battalion. The assault began well, and the flanking 9th Company, consisting of just two hundred men, were invited to take a break from a dank, frozen, four-foot dugout to join in. Their young glory-seeking commander accepted the invitation hungrily. His name was Lieutenant Erwin Rommel.1
Rommel quickly led his men forward toward the French trenches. For a moment they were pinned down on the frozen ground by heavy but aimless gunfire; Rommel decided a rapid advance would be safer than staying put. The French, panicking, fled; the Germans saw flashes of blue jackets and red trousers as their opponents ran. Rommel’s men pressed ahead through a French defensive line, then a second, and then a third, pausing occasionally to shoot at the retreating French. Rommel crawled through a tangle of barbed wire several hundred feet deep. At one point, he realized that nobody was following him, but he scouted around and found a safe passage through. Then he crawled back and told his second-in-command, “Obey my orders instantly or I shoot you.”
Duly encouraged, Rommel’s company penetrated deep into the French position, seizing buildings and fortifications that the French had abandoned. Rommel sent a message back to his battalion commander requesting support. From a position of relative safety, he harried an entire battalion of French soldiers on his right flank as they tried to withdraw through their own barbed wire. But he was vastly outnumbered and the French re-formed and began to close in, recapturing one of the buildings to his right. Rommel knew that he was vulnerable unless reinforcements arrived to support him. Then a message was shouted across from the German lines: 3rd Battalion had abandoned its attack. Ninth Company was surrounded.
Rommel had two obvious options: hold his fortified position until his soldiers’ ammunition ran out, then surrender; or retreat across a daunting set of obstacles under fire from all sides. He chose neither. Instead he threw his men into a fresh attack, routing the French from the building they’d just occupied. The French were astonished; their assault faltered while they tried to understand what was happening, and Rommel seized his opportunity to lead his company in single file back through the wire and across the trenches. By the time the French had gathered themselves, Rommel’s company was three hundred yards away and out of danger. He suffered no fatalities and left no injured men behind.
Rommel became a master of the messy craft of war, with this early engagement at Binarville setting the tone. He believed that opportunities arose from confusion on the battlefield, and tried to generate more opportunities by creating more confusion. His rapid movement and bold independent action created a feedback loop: the enemy would be confused; that would produce unpredictable openings; Rommel would seize those openings, creating more chaos and further opportunities.
This was a messy approach because the rapid, relentless, and unpredictable movements that so baffled the enemy would also confuse his own side. As a result, Rommel often found himself outnumbered, his forces scattered and out of touch with one another, low on fuel or physically exhausted, and with supply lines in disarray. And yet he kept winning. His men might be tired and confused, but his enemies were utterly bewildered.
Rommel triumphed repeatedly by creating a chaotic situation that nobody understood—and trusting, usually correctly, that he could improvise his way through the mess before his enemies did. It is a strategy with applications far beyond the battlefield.
• • •
In a competitive situation, you win by beating your opponent. Sometimes the opponent is relevant only as a benchmark; a 100-meter sprinter can tune out his rivals and focus on the finish line. But many competitors—a chess player, a boxer, a military commander, a business leader, a politician—cannot ignore the opposition. And one way to win is to encourage your opponent to lose.
In the film Rocky II, the plucky left-handed—“southpaw”—challenger Rocky Balboa adopts an orthodox right-handed stance to fight the champ, Apollo Creed. For most of the bout he’s a human punching bag, but in the final round Rocky switches to his preferred southpaw stance and knocks out the bewildered Apollo. Switching from southpaw to orthodox and back must have been a handicap for Rocky, but it was outweighed by the trouble it caused for Apollo Creed.
If this seems like something that could happen only in Hollywood, tell that to Wladimir Klitschko. One of boxing’s great heavyweight champions, the Ukrainian boasted a record total of twenty-eight title fights and was on a run of eleven years and eighteen successive defenses. Then, late in 2015, Klitschko fought Tyson Fury, an underdog from Manchester noted mainly for strange stunts such as showing up at a press conference in an ill-fitting Batman costume. Few people gave Fury much chance—but in the middle of the fight he emulated Rocky, switching to southpaw and outboxing an unsettled Klitschko to take his world title belt away.
“I know how good Wladimir is and Wladimir did not show how good he is. I know his strengths but I couldn’t see any of it,” said Klitschko’s own brother, Vitali, himself a former world champion. Boxing pundit Steve Bunce was one of many to conclude that Klitschko hadn’t been able to cope with the sheer awkwardness of his unfancied foe: “Wladimir looked bad tonight because Tyson Fury made him look bad.”
There’s even a theory—the Faurie-Raymond hypothesis2—that left-handed people have continued to survive in a right-handed world, despite numerous disadvantages and indignities, because right-handed people aren’t used to fighting them.* Whatever the truth of that, the lesson is that success in a competitive situation isn’t just about what you do—it’s about the effect your actions have on your opponent.3
For another example, look at chess. There’s little doubt about the world’s best player: it’s the twenty-five-year-old world champion Magnus Carlsen, a dashing young Norwegian who when just thirteen beat the great Anatoly Karpov. Carlsen was the youngest player to be rated number one in the world, and he has won both of his world championships with comfort. He’s the highest-rated player in history and his endgames are particularly admired, with Carlsen conjuring victories seemingly out of nothing. But the curious thing about Carlsen is this: For all his brilliance, his moves just aren’t that good.
How do we know this? Who would dare to second-guess the greatest player to grace the board? Computers, of course. The top computer programs are vastly better than any human, and can be used to evaluate human moves relative to the computer’s best.
Ken Regan, a professor of computer science at the University at Buffalo, and a former chess professional, has been using chess software to evaluate human players. He finds that Carlsen’s moves are good, of course, but others also play good moves—Vladimir Kramnik, for example, plays superbly, rarely falling far short of the move that the computer would recommend. Why, then, isn’t Kramnik still the world champion? The answer isn’t in the moves that Carlsen plays, but in what those moves do to his opponents. Put simply, other humans don’t play to the best of their ability against Carlsen. And this doesn’t seem to be just because of fear (although other players do fear him) or exhaustion (although he is fit and sometimes grinds out victories in very long games). It is that the specific moves he chooses somehow pose problems for his opponents.4
Another chess-and-computing analyst, Guy Haworth, notes that one of Carlsen’s tactics is to complicate the game just as the players are pushed for time. In professional games this is usually as move 40 draws near. Guy Haworth tracks the average error of human players; in games involving Magnus Carlsen, that error explodes just before move 40.5 Carlsen seems to relish dragging both players into a fiendish puzzle of a position at the very moment that time is running out. After all, Carlsen doesn’t need to play perfectly to win; he just needs to ensure that his opponent plays worse.
• • •
As Rommel rose through the ranks, from commanding a handful of men at the start of the First World War to becoming one of the most feared and admired generals of the Second, he developed a distinctive philosophy, founded on the observation that fleeting opportunities often emerge in the confusion of war.
Those ideas could clearly be seen when Rommel was a company commander in 1917, in charge of a detachment of around eight hundred men in mountainous terrain north of Venice. Rommel’s men were to support an ambitious German offensive against seasoned Italian troops manning three fortified defensive lines.6
He began with a rapid scouting expedition, which discovered a concealed path through the hills and captured several Italians and a defensive fortification without trouble. A rainstorm was lashing the battlefield—difficult conditions for most commanders, but for Rommel the storm provided ideal cover for rapid movement. Far ahead of the agreed schedule, he led his entire detachment along the secret path, nearly a mile ahead of his own lines. The next day, he attacked the Italians from an unexpected angle, taking many prisoners and seizing their defensive positions.
Rommel was now a victim of his own success: deep behind enemy lines, he was outnumbered and surrounded just as he had been at Binarville. And as at Binarville, Rommel counterpunched with speed just when the enemy was massing for an attack. Despite his limited firepower, the surprise unmanned the Italians. Rommel’s prisoner count was now up to 1,500. And so it continued, surprise attack after surprise attack through the moonlit night, while Rommel’s detachment rolled behind and parallel to the Italian front line. Thousands more prisoners were captured.
At dawn, Rommel faced heavy fire from fortified positions, and he abruptly changed tactics, charging uphill across exposed ground, overrunning the Italians through sheer speed.
Finally, Rommel’s elated and exhausted troops faced a possible catastrophe: so many prisoners were now pouring back from the battle that Rommel’s commanding officer presumed that the battle was won and gave the order to retreat and regroup. Some troops obeyed, others did not. In the confusion, Rommel found himself exposed and outnumbered yet again. Most commanders would have taken that moment to pause, gather their forces, and coordinate with their superiors. Rommel reckoned that the situation, however messy, was still in his favor despite the fact that he was outnumbered ten to one. Any delay would allow his disoriented and demoralized enemy to regroup. He was proved correct: the Italians were reeling, and so Rommel seized a strategically vital summit, Mount Matajur, by improvising another sudden attack.
Rommel had created a messy situation and triumphed by navigating that mess better than anyone else on the field of battle. In two and a half days of almost continuous fighting, Erwin Rommel’s detachment of a few hundred men had taken nine thousand prisoners for the loss of six soldiers.
• • •
Rommel believed that opportunities were often fleeting, and that the chaos created in the scramble to seize them was a price worth paying—especially since chaos could give an edge to the side best able to cope. Rommel’s strategy holds lessons for any competitive situation, and it can work just as well in business and politics as it does in war.
At the start of 1994, the World Wide Web was a niche pursuit. Microsoft, the most powerful software company in history, didn’t have a website, let alone provide a Web browser. But Web traffic was over two thousand times busier than it had been a year earlier. That explosive growth meant a brief window of opportunity, the kind of opportunity that Erwin Rommel would have lunged for had he been a businessman rather than a general. Instead, it fell to a young computer scientist working on Wall Street to seize the moment. His name was Jeff Bezos.
“When a company comes up with an idea, it’s a messy process,” Bezos told Brad Stone, the author of the definitive biography both of Bezos and of the company he created: Amazon.7 From the outside, that is a puzzling claim. To the consumer, Amazon is a byword for tidy efficiency: you want a product, you search Amazon, you find the product, you buy the product, the product arrives. To competitors, fighting Amazon is like fighting a machine: every move is calculated, every strategy quantified. It’s a killer robot that feels no pain. And there is a kind of truth in both these caricatures of Amazon, but it’s a partial truth. The story of Amazon is a long series of crazy goals, brutal fights, and squandered billions—an utter mess.
Bezos combined a grandiose vision with the sketchiest understanding of how that vision would be achieved. The name “Amazon” is inspired by the fact that the River Amazon “blows all other rivers away”;8 he wanted his company to be the “everything store,” the largest retailer on the planet, capable of swiftly supplying any consumer with any product they could imagine. And yet despite these big dreams, when the company opened for business selling books in 1995—by which time Microsoft still hadn’t launched its Internet Explorer browser—Bezos was utterly unprepared for the demand.
In the first week, Amazon sold $12,000 worth of books but shipped only $846. In week two, it shipped $7,000, still behind on the first week, while sales were up to $14,000. Right from the start, Bezos was behind and desperately scrambling to catch up. The company overdraft kept maxing out because the business was growing faster than the checks could be cashed; the management team was working late into the night to send out the books that had been ordered. (They sat on the floor: nobody had gotten around to buying tables.)
Bezos was making big claims to his customers—that there were a million books available; that there was a no-quibble thirty-day return policy—but he didn’t know how those promises were going to be kept. He trusted that they would figure something out.9
One might have thought that these early weeks were a good time to pause and regroup, to concentrate on making sure that Amazon was able to deliver on its early promises. But like Rommel, Bezos believed in seizing opportunities rather than pausing for breath. In Amazon’s second week of business, he received an e-mail from David Filo and Jerry Yang, the founders of one of the early Web’s most popular sites, Yahoo! Filo and Yang wanted to list Amazon on the Yahoo! home page—would that be okay? Bezos’s computer guru warned him that it would be like trying to sip from a fire hose. Bezos ignored him and accepted the offer from Yahoo!
Now they were even more of a success, and under even more pressure. The workload was inhuman. Bezos, his desk a cluttered mess, nearly poisoned himself after distractedly gulping down a rancid week-old cappuccino. When journalist Stone later tried to interview employees about that time in Amazon’s history, they didn’t remember much: they were getting so little sleep that their memories simply couldn’t function.
Accepting the Yahoo! offer within a fortnight of launch was characteristic of Bezos. To launch Amazon he’d given up a well-paid job on Wall Street, against the advice of his family to keep working at least until he picked up his hefty annual bonus. Bezos decided that the opportunity would not wait.10
In 1999, Bezos decided to start stocking kitchen equipment. In Amazon warehouses that had been designed to store, sort, and dispatch books, naked carving knives were suddenly scything down the chutes and into the sorting machines. Meanwhile the company’s database would be asking whether the knife was a hardback or a paperback.11
Amazon started stocking toys in the same year. The company held a press event in New York that was intended to show off its stock with tables piled high with merchandise—but when Bezos saw the modestly sized display he was livid at its lack of shock-and-awe scale. “Do you want to hand the business to our competitors?” he screamed. “This is pathetic!” And so some of Amazon’s top executives fanned out across Manhattan, stretching their personal credit cards to the limit as they scrambled to buy whatever they could find in the local Toys“R”Us stores.12
When Christmas came, the desperate scrabble for toys was repeated on a colossal scale, as Amazon employees across the United States bought Costco and Toys“R”Us inventory in bulk and drove it to the Amazon warehouses. Amazon also bought the entire Pokémon stock from the new—and evidently naïve—ToysRUs.com site.13
The scramble was too much for Amazon’s systems. The company’s warehouses began to choke on pallets of Pokémon Jigglypuffs and Walk ’N Wag dogs from Mattel. Products lost somewhere in the vast distribution centers would send Amazon’s databases haywire. Unshipped orders would clog the chutes in the sorting facility, each blockage spawning half a dozen further delays. Internally, the company was on its knees: it launched a “Save Santa” campaign and Amazon staff members were booked into hotels near the warehouses (two to a room) and didn’t go home for a fortnight. As Christmas passed, 40 percent of the toy inventory was unsold and probably unsellable. (“If I have to, I will drive it to the landfill myself!” Bezos had declared when he’d placed a vast order with a toy manufacturer.) The losses on the toys Amazon had bought at retail from its rivals, then resold at a discount to consumers, must have been appalling. But the company survived and customers were happy.
In the summer of 2000 came the dot-com bust. Amazon, as one of the oldest and best-established businesses on the Web, should have easily weathered the downdraft, but it was a closer call than it might have been because Amazon was running substantial losses in pursuit of further growth. Top analysts were telling the world that Amazon was running out of cash. Bezos’s aggressive discounts gave his finance team nightmares as they tried to borrow money from a suddenly skeptical financial market. Barron’s, a prominent financial magazine, ran the headline “Amazon.Bomb.” Suppliers started wanting to be paid in advance in case Amazon went bust.
There was a real possibility that Amazon wouldn’t see Christmas 2001. Yet Bezos refused to slow down. Amazon was saved by a clever finance director, who had arranged a cash flow cushion, and some cobbled-together joint ventures. It had been close. As Brad Stone writes, “Amazon survived through a combination of conviction, improvisation, and luck.”14
Most companies would have retrenched at that point. Instead, over the next few years, Amazon launched products as disparate as the Kindle (which immediately and repeatedly sold out, as Amazon struggled to manufacture it), Mechanical Turk (an unsettlingly named global clearinghouse for labor, which pioneered crowdsourcing but was criticized as being a sweatshop), the Fire Phone (widely reviewed as ugly, weird, and disappointing), Marketplace (where competitors to Amazon would use Amazon’s own product listings to advertise their own cheaper alternatives), and Amazon Web Services. AWS in particular was a bold stroke—a move into cloud computing in 2006, four years ahead of Microsoft’s Azure and six years ahead of Google Compute.
As Bezos liked to say during the crunches of 1998 and 1999, “If you are planning more than twenty minutes ahead in this environment, you are wasting your time.”15 He was a man in a hurry. No wonder he created such an almighty mess.
• • •
Come the Second World War, the Italians were the Germans’ allies, but they seemed as prone as ever to losing battles. At the end of 1940 and beginning of 1941, a small British army under General Richard O’Connor had destroyed a vastly larger Italian force in North Africa, taking 130,000 prisoners and pushing the Italians back from the borders of Egypt to the middle of Libya. It seemed likely that the British would soon advance to Tripoli itself, driving the Italians out of Africa entirely. Faced with this alarming prospect, the Italians accepted Hitler’s offer of assistance—a small, defensive detachment, just one division of tanks and one division of mechanized infantry. It was led by General Rommel.16
Rommel was flushed with success after the Blitzkrieg* invasion of France, when his panzer tanks had surged far ahead of their own troops, roaming freely behind French lines and taking 97,000 prisoners. Nobody knew what was going on: French peasants often assumed Rommel’s men were British troops; the French army called them “The Ghost Division”; not even German High Command knew where Rommel was or what he would do next.
But when Rommel arrived in Tripoli in February 1941, he faced a situation that seemed desperate. His men had no experience of desert warfare, and shipments of his panzers were trickling too slowly across the Mediterranean. He feared that the British would seize their opportunity to attack before he was ready to mount a defense.
“If the British advance on Tripoli immediately without regard for casualties, our general situation will be very grave indeed,” he warned Berlin. Improvising, he drove his panzers in a big continuous loop through Tripoli, so that his meager force looked much larger to the locals; he also commissioned hundreds of dummy tanks, hoping to deceive British air reconnaissance. As with the piles of toys that Bezos’s team had scrabbled together from the toy stores of Manhattan, this was a bluff—a display of strength that didn’t exist, designed to slow down a more cautious opponent. To his wife Lucy, Rommel wrote, “The next two or three weeks are going to be crucial. After that things will even up.”17
Rommel was lucky: the British did waver, conserving strength and diverting some forces to Greece instead. Rommel sensed his enemy’s hesitancy and began jabbing forward along the North African coast, searching for weakness. At each point he found that the British were withdrawing. A few weeks after Rommel’s arrival, the British were digging in at Mersa El Brega, a defensive position protecting Cyrenaica, a land mass that bulged out from the desert into the Mediterranean. Cyrenaica was a prize: its coast was adorned by the fortified port of Tobruk, where vast quantities of petrol and ammunition were stored; there were numerous airfields there; and there was plenty of water, essential for fighting a desert campaign.
Rommel’s immediate superior, the fat, white-mustached general Italo Gariboldi, wanted him to stay on the defensive—a demand that Rommel simply ignored, provoking furious shouting matches. More worryingly for Rommel, German High Command had also ordered him to wait before mounting any major counterattack against the British. But why? What Rommel did not know was that the awful slaughter of Operation Barbarossa, Hitler’s war with Stalin, was about to be unleashed. As he had done before, he decided to finesse his orders and attack anyway; with the British fortifying Mersa El Brega, he felt that he had to act immediately if he was to act at all.*
After outflanking a stout defense at Mersa El Brega and capturing many British vehicles, Rommel saw yet another opportunity. The British were retreating in a long, slow curve around the bulging coast of Cyrenaica. Rommel pressed them with his left wing, and sent his right wing directly across the interior.
To call this daring was an understatement. It seemed insane. He’d been told that the route was impassable. (He personally explored it and concluded that it wasn’t.) He was then informed that his thrusting force was running out of gasoline and needed four days to allow replenishment vehicles to shuttle back and forth from their supply dumps with more fuel. His response was a remarkable piece of risky improvisation: he ordered the division to unload every truck and car and send them all back across the desert for fuel, leaving the men and their panzers stranded and vulnerable for twenty-four hours. Rommel’s calculation was simple: the British were in full retreat; they wouldn’t be looking for helpless German soldiers in the middle of the desert, and even if British scouts did find the Germans they wouldn’t be able to organize to attack the Germans quickly enough. He was correct.18
Rommel’s counterstrike was shambolic. Cyrenaica was a land of fine red sandstorms, land mines, and soft, soft sand. The Germans kept getting lost and bogged down. Rommel would fly over the battlefield trying to find everyone’s bearings. At one stage his scout plane buzzed his own troops at shoulder height and a piece of paper fluttered down, bearing a note: “If you don’t move off again at once, I’ll come down!—Rommel.” So great was the confusion that he almost landed in the middle of a column of men that turned out to be British.19 Later, Rommel’s subordinates complained about the disarray and constant changes of plan. Back in Berlin, a German general mournfully noted in his diary in late April 1941, “Rommel has not sent us a single clear-cut report all these days, and I have a feeling that things are in a mess.”
That assessment was right: Things were in a mess. But while the Germans were in chaos, the British were in a worse state: bewildered, outflanked, and desperately trying to withdraw around a congested coastal road. Many men were captured—including the British General Richard O’Connor—and the remainder of the British forces barely scrambled away. Six days after Rommel began his absurd and disorganized campaign to take the strategic prize of Cyrenaica, he had succeeded.
Over the course of a year and a half, Rommel’s small expeditionary force rolled over a thousand miles of North African desert, from Tripoli in the west almost to Alexandria in the east, prompting the British fleet to flee to the safety of the Suez Canal and the British High Command to prepare a list of assets to be demolished if they were forced to surrender. As news of Rommel’s victories reached Britain, Winston Churchill almost lost his job as prime minister, and paced the corridors of power, muttering: “Rommel, Rommel, Rommel! What else matters but beating him?”20
• • •
The idea that Donald Trump could mount a serious challenge to be the 2016 Republican nominee for president was initially regarded as a joke. A real estate developer, reality TV star, notorious blowhard, and political dilettante, Trump would be taking on far more experienced rivals—notably former Florida governor Jeb Bush, the brother and son of presidents.
Confounding all expectations, by the autumn of 2015, Trump had become a front-running phenomenon while Jeb Bush languished in the polls. The same pattern played out time and again: Trump would broadcast some inflammatory comment—about immigrants, for example—playing to the basest instincts of the Republican electorate. His rivals for the nomination would tiptoe in, trying to show that they empathized with the concerns of Trump-admiring voters without saying anything offensive, and often tying themselves in knots in the process; at one point Jeb Bush, in trying not to echo Trump’s criticism of Mexicans, ended up seeming to smear Asians. Then, seemingly before the interview was even finished, Trump would pop up on Twitter, mock his rival, and do something else outrageous—such as take a misogynistic swipe at a female interviewer. After Trump performed a crude schoolboy impersonation of a disabled New York Times reporter, provoking predictable outrage, he took to Twitter again to change the subject, criticizing the “dopes” at The New York Times for overpaying to buy The Boston Globe.21
Trump’s career-politician opponents were tidy-minded, surrounded by complicated messaging operations that crafted press releases and briefed them for interviews, trying to protect their image and prevent gaffes. But no matter how carefully his opponents worded their speeches and statements, a quick tweet from Trump would seize the headlines and keep his poll ratings high.
One day in 2015, I received a message from a friend, containing a link and a single line of commentary on Trump: “He’s in their OODA loop.”
The OODA loop is military jargon. It was coined by a retired U.S. Air Force colonel named John Boyd, who, two decades after his death, continues to be a cult figure among military thinkers. Boyd’s theory helps explain everything from Rommel to Bezos to Trump. The theory was set out in a much-photocopied document known as “Patterns of Conflict,” a mess of typed and retyped paper, hand-annotated and 196 pages long.
As a young man, Boyd had a reputation as the best pilot in the Air Force. He was a thinker: he outthought his enemies in the air, and soon he was working on a toolkit of aerial maneuvers for his fellow pilots, explaining how each move could be used to confuse and outflank an opponent. It became a bible of air combat around the world. In the 1960s, Boyd developed a carefully quantified theory of why some fighter planes seemed much more effective than others, despite looking unpromising on paper. In the Korean War, American F-86 Sabre pilots beat the Soviet MiG-15 pilots time after time—the kill ratio was around ten to one. This was a puzzle: the MiG-15 flew higher, turned more tightly, and accelerated faster than the F-86. All of that didn’t matter. Boyd showed that the U.S. planes could brake and maneuver in ways that confused their adversaries, and they had better visibility so avoided confusion themselves. The Soviet and North Korean pilots felt like they were fighting ghosts. Boyd’s analysis shattered the received wisdom in the U.S. Air Force and led to the introduction of totally new types of aircraft.
Then, in 1975, Boyd quit the Air Force. He had bigger things to think about. In his youth, he’d been called “Forty Second Boyd,” because that’s how long it took him to beat any other fighter ace. After he’d perfected “Patterns of Conflict,” he became associated with a somewhat longer time period: six hours. Delighting in his freedom from authority, Boyd went around in slippers, frayed shirts, and polyester trousers, and steadfastly refused to compromise on the length of the “Patterns” briefing, no matter how senior the person who expressed an interest in hearing it. “Since your boss is so pressed for time,” he once bellowed down the phone at the assistant to the Chief of Naval Operations at the Pentagon, “here’s an idea that will save him a lot of time: how about no brief?” Then Boyd hung up.22
Despite Boyd’s truculence, “Patterns” was a hit. First the briefings were small, but by the end of the 1970s, word was spreading across Washington, D.C.: You’ve got to hear Boyd. Congressional staffers crowded into his briefings; so did reporters. James Fallows of The Atlantic Monthly became a fan. So did the management guru Tom Peters. So did a future defense secretary and vice president, Dick Cheney.
So what does “Patterns” say? Essentially, Boyd took the lessons he learned thinking about those fighter planes—that the F-86 pilots beat the MiG pilots because they could change tactics more quickly—and extended them to combat across history. His presentation ranged across wildly different circumstances: from the Battle of Cannae in 216 BC, in which Hannibal’s Carthaginian army surrounded and annihilated a vastly superior force of Roman soldiers; to the Entebbe raid in the summer of 1976, in which Israeli special forces landed in hostile territory and freed more than a hundred people being held hostage by Palestinian militants. The unifying thread between these and countless other victories against the odds: the way that the successful commanders kept changing the situation faster than their adversaries could figure out what was going on. Chaos wasn’t just something that happened on the battlefield—it could and should be deliberately deployed as a weapon.23
Boyd invented the OODA acronym to describe this process of decision making. OODA stands for “Observe-Orient-Decide-Act”—or, in plain English, working out what’s happening, then responding. Boyd’s six-hour briefings drove home the idea that this OODA loop of information gathering and decision making was crucial in any competitive struggle. If you could make quick decisions, that was good. If you had a strong sense of what was going on around you, that was good, too. But more profoundly, if you could disorient your opponent, forcing him to stop and figure out what was going on, you gained an advantage. And if you could do this relentlessly, your opponent would be almost paralyzed with confusion. Just as he was about to act, something new would happen and he would have to stop and think again. You would have more than an edge—you would have him at your mercy.
The friend who thought of OODA loops when he saw Trump in action had learned about them in his job at Greenpeace, where Boyd’s theories are used to teach activists how to keep oil companies off balance. In the description of political commentator Josh Marshall, comparing “Jeb’s painful and drawn-out efforts . . . with Trump jumping from one attack to another on Twitter, entirely at his own pace and on his own terms,” is a perfect illustration of what it means to get inside your opponent’s OODA loop.24
Trump’s image took plenty of knocks over the course of the campaign, but he minimized the damage through an ability to change the conversation whenever he wanted. He chose his battlefields, even skipping a televised debate on the eve of the vital Iowa caucuses, and dominating the headlines as a result. Trump made sure both the media and his opponents reacted on his terms. He wasn’t always perfectly prepared, but his preference for speed over perfection ensured that opponents were always scrambling to figure out a response.
• • •
In contrast to Trump’s agility, the presidential campaigns of rivals such as Jeb Bush could be encapsulated by a word the German High Command used in conversation with Rommel in 1941 to describe the leadership qualities of the British Army. The German word was schwerfällig—ponderous. The historian David Fraser elaborates: “There was demonstrated, in British actions, rigidity of mind and reluctance to change positions as swiftly and readily as situations demanded . . . great fussiness and over-elaboration of detail in orders.”25
In other words, the British were Rommel in reverse: rather than quick, deft, heedless of detail, and happy to make things up on the fly, they were slow, clumsy, tidy-minded, and unwilling to improvise.*
If you’re trying to win by creating a mess and betting on yourself to figure it out more quickly than your opponent, then it helps if your opponent is schwerfällig. That was Jeff Bezos’s calculation, as he explained at the beginning of 1997 when invited to speak at Harvard Business School about his new dot-com business. At the time Amazon was no mighty river, but a mere trickle. Most people had never heard of it and, of those who had, few felt comfortable sending a credit card number into the confusing, anarchic World Wide Web. The savvy business thinkers of Harvard Business School heard Bezos’s explanation of what he was trying to do, and they were skeptical. It wasn’t that e-commerce itself looked infeasible—smart people could see that it might well be big. It was that Bezos had a competitor capable of swatting him on a whim.
Bezos had chatted with this terrible adversary a few months earlier. He and a colleague had a steak dinner at the Dahlia Lounge in Seattle with the Riggio brothers, Len and Stephen. Len Riggio, a tough-minded son of a boxer, had turned Barnes & Noble from a troubled company with a single Manhattan shop into a bookselling behemoth. Barnes & Noble had pots of money: $2 billion in sales in 1996, more than a hundred times larger than Amazon’s $16 million. It had serious clout with publishers, who knew that Barnes & Noble could make or break any book. It enjoyed a prominent place in malls across North America and in the awareness of book lovers. It was used to crushing the life out of competitors—independent bookstores had lost almost half their market share in the five years running up to 1997. And of course, Barnes & Noble understood books.26
Len Riggio amiably told Bezos that Amazon should agree to some sort of joint venture, and soon, because Barnes & Noble was planning to sell books online itself. A Bezos lieutenant recalls the message thus: “You’re doing a fantastic job, but, of course, we’re going to kill you when we launch.”27
It turns out that the students and faculty of Harvard Business School were thinking along the same lines as Len Riggio. They could see the potential of online bookselling but they could also see that Barnes & Noble was in a far stronger position than Amazon. When Barnes & Noble moved into the tiny online market, Amazon would be history.
“You seem like a really nice guy, so don’t take this the wrong way,” one MBA student advised Bezos, “But you really need to sell to Barnes & Noble and get out now.”28
Bezos acknowledged the risk, but he wasn’t afraid. He was convinced that as long as he kept moving, kept improvising, kept being willing to make a mess, his competitors would hesitate. This was true not just of Barnes & Noble, but the giants he planned to confront next: Toys“R”Us, Target, and even Walmart.
“I think you might be underestimating the degree to which established brick-and-mortar business, or any company that might be used to doing things a certain way, will find it hard to be nimble or to focus attention on a new channel,” he told the class at Harvard. “I guess we’ll see.”29
Bezos, it turned out, had judged Len Riggio’s perspective perfectly. Barnes & Noble was making good money in its bookstores and didn’t want to spoil the corporate balance sheet by running a loss-making website; its distribution systems were large and well-established and designed to send large shipments to large stores, rather than one book at a time to individual customers. And Riggio didn’t want his best people to be distracted by a website, so Barnes & Noble’s e-commerce operation received too little management attention for too long. Barnes & Noble had developed excellent systems for running retail bookstores. But it hesitated to engage with the chaos of the new world of e-commerce. Amazon thrived on it. To use John Boyd’s jargon, Amazon was inside Barnes & Noble’s OODA loop, and Barnes & Noble’s response was schwerfällig—ponderous.30
Toys“R”Us also found itself made to look ponderous. Bezos’s chaotic and costly foray into the toy business looked insane during the Christmas season of 1999, as Amazon tangled up its own warehouses, exhausted its staff, and bled money. Viewed from the Toys“R”Us position, though, Amazon had achieved something enviable: it had happy customers and had established itself as a great place to shop online for toys.
ToysRUs.com, meanwhile, had been cleared out of the best toys by Amazon’s own buyers. A couple of days before Christmas, the media carried headlines such as “Toys‘R’Us Falling Short on Christmas Deliveries”; the next summer, the Federal Trade Commission fined Toys“R”Us $350,000 for not keeping shoppers informed about shipping delays over Christmas. Several other companies were fined for similar reasons. Amazon wasn’t among them. It had lost far more money in its pre-Christmas scramble for stock than Toys“R”Us had been fined, but it had gained something much more valuable: consumers who remembered that while Toys“R”Us had let them down at Christmas, Amazon had delivered the goods.31
These slow or clumsy responses from Amazon’s competitors weren’t simply a matter of underestimating a new entrant: they continued after Amazon became an established player. When Amazon was rumored to be launching its Kindle e-reader in 2007, Len Riggio’s brother Stephen was the CEO of Barnes & Noble. Scarred by an earlier costly experiment with primitive e-reader technology, he was in no hurry: “Certainly there’s an opportunity to get back into the business, but we think it’s small at this moment and probably will be small for the next couple of years. When the market is there, we’ll be there.”32
Schwerfällig.
Publishers also realized that they had blundered: in their negotiations over e-book rights with Amazon, they had failed to specify a minimum price at which Kindle e-books would be offered for sale. Amazon announced a loss-making price of $9.99, far cheaper than a $30 hardback. At that price, people were clamoring for Kindles.33 Yet again Amazon created chaos that threatened to overwhelm it, selling out of Kindles in hours, before discovering that it was impossible to manufacture fresh stock because of a problem with a supplier.
Viewed in isolation, Amazon looked laughable and amateurish—as did Rommel’s thrust across Cyrenaica. But step back and consider the entire marketplace and it becomes clear that, however messily, Amazon had outflanked its rivals. By Christmas 2009, Amazon had a market share in e-books of around 90 percent.34
One can tell almost exactly the same story about Amazon’s initially baffling venture into cloud computing: a crazy rush, a series of early technical problems, a loss-making price. Then within a few years, Amazon, a mere bookseller, was the dominant player in cloud computing, and analysts were pronouncing that Amazon Web Services was a more valuable business than Amazon’s online retailing operation. The opportunity had been there to take, but the titans of the industry, IBM, Google, Apple, and Microsoft, had all hesitated at the prospect of a costly battle with the upstart.35
Again and again, we see Amazon moving quickly, losing money, struggling to cope with the demand it created, and in the end, dominating a market. Yet because of Amazon’s decisiveness and its tolerance for creating and then navigating an almighty mess, it left schwerfällig competitors outmaneuvered and gasping to catch up.
• • •
From Rommel’s half-formed expeditionary force against the might of the British Empire, to Bezos taking on Barnes & Noble, to Donald Trump targeting the Bush dynasty—the messy strategy has been embraced by the underdog. This is no coincidence. Maintaining momentum is exhausting; constantly improvising at speed is terribly risky. There is nothing in Boyd’s theory that says a strong force cannot act swiftly and confusingly, getting inside an enemy’s OODA loop. Yet it rarely seems to happen that way: few people are willing to take the messy path if a tidier approach of organizing, preparing, and coordinating looks like it might deliver victory.
Messy improvisation doesn’t guarantee success. It does, however, guarantee that there will be mistakes, recriminations, and stress along the way. Erwin Rommel’s swashbuckling style was a catastrophic failure when he tried to storm the heavily fortified port of Tobruk and was thrown back by 27,000 veteran Australian defenders. Careful planning might have broken through, but Rommel relied, as he usually did, on speed and energy, throwing his men into hopeless situations until his own generals refused to obey him. Again, these failures are intimately tangled up with the way Rommel conjured his successes. His tactics were designed to rout his enemies through sheer dynamism; they often succeeded brilliantly, but proved costly on the few occasions that they failed.
Jeff Bezos’s fast-and-messy strategy causes him constant headaches. Consider the time, in the spring of 2013, that a third party on Amazon’s UK website started offering T-shirts with slogans such as KEEP CALM AND RAPE A LOT, the result of recklessly designed software churning out hypothetical T-shirts and listing them on Amazon to see what would sell. Amazon’s website has also offered spam books, full of nonsense or plagiarized wholesale from existing works.36 In January 2015, police raided Amazon offices in Japan, investigating the alleged sale of child pornography by third-party retailers on the Amazon website. And in the summer of 2015, Amazon discovered that it was selling a magazine produced by and promoting the murderous group Daesh, better known to Westerners as ISIS.37
Embarrassing episodes are an almost unavoidable by-product of the messy way in which Amazon lets third parties hawk their goods on the Amazon website. As the designer Tim Maly comments, “Amazon isn’t a store, not really. Not in any sense that we can regularly think about stores. It’s a strange pulsing network of potential goods, global supply chains, and alien associative algorithms with the skin of a store stretched over it, so we don’t lose our minds.”38
Amazon’s experiments with third-party sellers began with Amazon auctions—“That didn’t work out very well,” admits Bezos39—and then Z-shops, and finally Marketplace. Marketplace drove Amazon’s own category managers crazy because they found themselves undercut on their own website by third-party sellers; meanwhile it leads to the never-ending risk that a third-party seller will do something embarrassing—such as sell pro-rape T-shirts—on the Amazon site, right beneath the Amazon logo.
Bezos has been willing to keep experimenting and keep taking chances, no matter how much hassle those experiments might cause within Amazon, or how ugly they might sometimes seem outside. And so far, his messy strategy seems to be working. Amazon Marketplace accounted for almost half the items sold on Amazon in 2014, and analysts speculate that it may be a fundamentally more profitable business than Amazon’s regular retail operation.40 Although Amazon’s relentless tempo has kept profits slim, the company is a triumph from a financial point of view. When the Harvard Business Review conducted a statistical review of the most successful living chief executives in America, few of the top CEOs were household names. (John Martin of Gilead Sciences? David Pyott of Allergan?) Howard Schultz of Starbucks was down in fifty-fourth place. Steve Jobs was dead. But Jeff Bezos? Bezos was number one.41
So if you’re in some kind of contest, and you feel the potential benefits of the messy approach outweigh the risks and costs, then what does it take to get inside your opponent’s OODA loop—to make him seem schwerfällig?
For a lesson, let’s turn to a man who played Rommel at his own game in 1941–1942, giving him some of his biggest headaches in North Africa—a man who struck almost at will at Rommel’s supply lines, and so mysteriously that he took on a mythical persona: the Germans named him “The Phantom Major.”
• • •
The Phantom Major’s real name was David Stirling. He dreamed up the idea for a tiny, hand-picked force while in a hospital bed, having nearly broken his back in a parachute drop gone wrong. On discharge, he immediately set about attempting to get his idea approved.
In Cairo, Stirling limped up to the Middle East Headquarters of the British Army, a gangling, six-foot-five giant of a man on a pair of crutches. Lacking a security pass, he amiably tried to bluff his way past the sentry. After failing, he hobbled over to rest against a nearby tree and, when the sentry was distracted, slipped through a gap in the fence and lurched in a hobbling sprint across the forecourt. By the time the sentry noticed, Stirling was slipping through the front door.
Inside, Stirling barged into offices at random and pitched his idea to the first officer he found. Stirling argued that a small group of talented men could be dropped by parachute behind the German and Italian front lines, and they could destroy Rommel’s air support in a surprise attack.42 Unfortunately, the first officer he met knew and disliked the easily recognizable giant: Stirling had fallen asleep during one of his lectures. Not only did the officer dismiss Stirling’s idea, he promised to crush it. But then, another opportunity—the sentry post phoned to report the security breach, and as the officer picked up the phone, Stirling vanished.
The next man Stirling found was a senior officer, General Sir Neil Ritchie, who was far more impressed with Stirling and his ideas. Stirling’s detachment was soon assembled and training for its first parachute drop. Its name: Special Air Service, or SAS.43
Stirling’s audacious raid on his own headquarters* demonstrates some of the principles of messy tactics. First, get yourself into a position of opportunity: Stirling didn’t know which officers he might meet if he broke into British headquarters, but he knew he’d meet somebody. Jeff Bezos felt much the same way about the early Web: it was impossible to predict how it might evolve, but clearly something exciting would be possible.
Second, improvise your way around obstacles. The sentry didn’t buy his excuses, so Stirling sneaked past him instead. The first officer he met despised Stirling and his plan, so Stirling quickly found someone else to persuade.
Third: speed counts for a great deal. Both Stirling’s break-in and his successful briefing to General Ritchie were possible because he moved faster than the people who were trying to stop him.
An unexpected corollary of this third principle is that while your team should understand their broad goals, they shouldn’t waste time trying to coordinate with one another. Stirling was happy to split his forces and let them operate independently. So was Rommel. Perhaps surprisingly, so too was Jeff Bezos, who notoriously once told his management team to spend less time communicating.44 Bezos’s point was that small teams should get on with achieving things rather than constantly checking with one another. For much the same reason, John Boyd opposed “synchronization,” once a big idea in the U.S. military. Boyd argues that synchronization was for watches, not for people: trying to synchronize activities wasted time and left everyone marching at the pace of the slowest.45
The early SAS missions follow all three principles. The first operation, a parachute drop in the teeth of “the most spectacular thunderstorm in local memory,” was a disaster in which more than half his men were killed or captured.46 Stirling simply changed tactics, abandoning the parachute idea and deciding instead to raid through the deep desert with experienced navigators in trucks.
In the first truck-borne attack, the convoy was spotted on the way to the target by Italian planes. Rather than call off the mission, Stirling split his forces to attack two separate airfields. One attack failed because the planes had flown off, but the other—led by his second-in-command, Paddy Mayne—was a big success. Mayne’s men had machined-gunned the officer’s mess, and then while the firefight raged had planted incendiary bombs that destroyed almost every plane on the airfield.47
Stirling and Mayne decided that they would have been even more successful if they had spent less time scouting around. Such scouting improved their knowledge of the target, but in both cases it had let their enemy prepare and respond. On future missions they resolved to move in quickly and silently, even without full knowledge of the target, because that would achieve total surprise. Speed and shock were more important than careful preparation.48
Sneak attacks behind enemy lines were an accepted part of warfare long before Stirling’s SAS was formed, and didn’t always work well—the British Army Commandos, formed over a year before the SAS, were launching daring raids along the coast of France and Norway. But these raids operated on a different principle: hundreds or thousands of brave men were involved in each raid, and they were synchronized and planned to run like a machine. While there were some notable successes, there were many failures and casualty rates were enormously high. The commando attacks couldn’t follow the messy approach of surprise and improvisation. They were too big. By trying to bulk up their fighting strength, they reduced their flexibility and stealth. Their enemy gained more from the commando’s efforts to strike in force than the commandos themselves did.
In contrast, Stirling’s men continued to adapt and improvise even after their tactics had been well honed. Consider an assault on the port of Bouerat, far behind enemy lines in North Africa, in the spring of 1942. Stirling planned to transport a canoe to the coast just outside the port, then paddle through the defenses in the dark, planting limpet mines on the gasoline tankers and other ships. After weeks of preparation and travel, Stirling’s plans were shredded when his truck hit a pothole and the impact broke the canoe. All hope of attacking Bouerat’s shipping had gone.
“We’ll have to reorganize a little,” he told his twenty men in a confident tone. “There are plenty of targets waiting for us at Bouerat. If we can’t get the ships, we’ll get the harbor installations instead.”49 In fact, he reassured the men, he’d been itching to attack land-based targets but just didn’t have the time to do both jobs.
The way Stirling was telling the story to his men, losing the canoe was a blessing. And he was proved right. Over the course of the next few hours, his men spread out, planted bombs on pumping equipment, wrecked vast stores of food, destroyed a dozen supply trucks—and best of all, blew up nearly twenty fully loaded gasoline tanker trucks. This was a real blow against Rommel, perhaps even better than destroying tanker ships. In any case, the SAS also discovered that there had been no tankers docked at Bouerat that night. Had the canoe been intact, it would still have been useless.50
In a later attack, the SAS planted their incendiary bombs on a group of forty German planes.* Normally the bombs were reliable, but on this occasion they had been prepared incorrectly. Only half of them exploded, and the German ground crew were able to put out many of the fires.51
As they sat in the darkness within sight of the smoldering airfield, Mayne and Stirling hatched a plan to turn this setback into another opportunity. The SAS patrol had a jeep armed with anti-aircraft guns, and the targets were sitting around not far away. Why not just drive to the airfield and shoot the planes into splinters? They did exactly that, destroying twelve more German planes.
The improvised attack worked so well that in the future the SAS repeated it: eighteen to twenty jeeps armed with sixty-eight antiaircraft guns drove in formation onto a German airfield and destroyed every plane they could see. All this was under the light of the full moon, poor conditions for the usual bomb-planting but perfect for close-formation jeep maneuvers.52
Much like Rommel, Bezos, and Trump, David Stirling followed a messy road in pursuit of victory. If the opportunity was there, he would seize it and figure out the details later. When he hit an obstacle, he would abandon his plan and start improvising a way around it. And he pursued speed and surprise. A coordinated, well-researched move might look good on paper, but it would be useless if it meant giving the enemy time to react.
• • •
David Stirling was eventually undone by the tidy-mindedness of his superiors. When the British High Command saw how well he was doing, they recalled him to Cairo and told him he was to participate in something much grander and more coordinated: there would be far more men in the raid, and many more targets. All this would support a grand new assault against the Germans. Stirling tried his best to dissuade them, but it was hopeless. With so many people involved there was too much loose talk and the advance was conspicuous; the Germans saw the raids coming well in advance and the mission was a disaster.
Rommel’s most famous defeat, at the battle of El Alamein, was also sealed by a rare piece of tidy-mindedness. Perhaps defeat was inevitable anyway: Rommel was on sick leave in Germany when the battle began, and his deputy was killed almost immediately; Rommel’s British adversary, Montgomery, had much shorter supply lines, excellent air support, four times as many good tanks, and seven times as many men. But Rommel’s undoing was a rare emphasis on careful logistical planning. Dreadfully short of fuel, Rommel insisted on knowing to the hour when and where tanker ships from Italy would be docking to supply his forces. He didn’t know about Ultra and Bletchley Park, the top-secret codebreaking center; he never realized that every single one of those “unbreakable” coded messages was being decoded. Supplied with that tidy German schedule, the British air force destroyed most of the gasoline tankers, and Rommel’s dynamic panzers were stranded. It’s hard to get inside your enemy’s OODA loop when he’s reading your mail.
In 1943, Stirling was captured and sent to a German prisoner-of-war camp. He escaped, but it wasn’t easy for the giraffe-limbed colonel to travel inconspicuously as an escapee, and he was recaptured. He escaped again, and was recaptured again. That happened at least four times, before he sat out the remainder of the war in Colditz. He was knighted in 1990, shortly before his death at the age of seventy-four.
Rommel met a more ironic end. In the last days of the Third Reich, two of Hitler’s generals visited Rommel’s home to inform him he had been implicated in a plot to assassinate the Führer.* They presented him with the offer of a cyanide capsule and a hero’s funeral, as an alternative to public exposure and a show trial. Rommel briefly weighed his options, before realizing that for once, there was no opportunity for counterattack. He could not have known, as he swallowed the poison, that the order of service for his state funeral had already been prepared. The wreath for his coffin had been delivered and stood ready at the local army barracks, waiting only for the final element: Field Marshal Rommel’s corpse. Sometimes the forces of careful forward planning cannot be defeated.53