Invisible Influence

by Jonah Berger

  Take teams losing by a point. Everything else would suggest these teams should be about 7 percent less likely to win than teams ahead by a point. Controlling for how good each team was, whether they were playing at home or not, and all the other factors, out of one hundred games, teams losing by one point at halftime should have won seven fewer games than teams winning by a point.

  But they didn’t.

  In fact, teams that were losing by a point were actually more likely to win. Not only did being behind increase a team’s chance of winning (by around 8 percent), but, compared to their opponents, teams that were behind by one actually won more games. Even though they tended to be worse teams and had to score more points than their opponents to win.17

  If you had to bet money, betting on the team down by one at halftime would be a safer bet.II

  * * *

  Why does losing lead to winning? To find out, we had people play a simple game.

  Imagine sitting in front of a computer keyboard. On the left side of the keyboard, right below the letter Q, is the A key. Toward the bottom of the keyboard, right between the letters V and N is the B key. Place one finger on the A key, and one finger on the B key, and imagine pressing them, in short succession, as quickly as possible. A, B, A, B, A, B, as fast as you can.

  Every time you press those two keys in order, you get a point. The faster you mash on those two keys, the more points you get. Not the most fun game in the world, but pretty easy to play.

  Now imagine that you are competing against someone else who is playing the same game. There are two thirty-second halves (or periods of play) divided by a short break (or halftime). Whichever player has the most points at the end of the game wins a small sum of money.

  We told different groups of players different things during halftime. While some players were told nothing, other players were given competitive feedback. Similar to Opower’s energy reports, they were given information about how well they were doing relative to others.

  To examine the effect of being behind, we rigged the competitive feedback. We told players that their opponent had scored one point more than they had so far, and thus they were one point behind. Then we measured how hard people worked in the second half of the game. Whether they increased or decreased the number of keys pressed.

  Thinking they were behind increased motivation. It made people work harder. Compared to participants that received no feedback at all, those who thought they were behind increased their effort more than threefold.III

  * * *

  Competition influences motivation by shaping people’s reference points, or the yardstick they use to measure how well they are doing. When running a 5K race, taking a test, or making sales calls at the office, we often set goals for ourselves. We want to run the race in under twenty minutes, get an A, or bring in ten new clients this month.

  Our performance relative to those goals, in turn, affects how hard we continue to work. Consider the following:

  Chip and George both love to work out and each usually follows a workout plan that involves twenty-five sit-ups a day. One day, Chip sets a goal of performing thirty-seven sit-ups and George sets a goal of performing thirty-three sit-ups. Both Chip and George are tired after performing thirty-five sit-ups and, at most, have the energy to perform one, maybe two more.

  Who do you think will work harder to perform those final couple sit-ups? Chip or George?

  People tend to think that Chip will work harder than George to do the last couple sit-ups because he’s yet to reach his goal.18 He has only done thirty-five and his goal was thirty-seven. Chip is almost there and with just a little more effort he can achieve what he set out to do. While George probably feels good because he’s achieved his goal, Chip may feel unsatisfied because he hasn’t gotten there yet. And that dissatisfaction will motivate him to work harder. Compared to being ahead, being behind is more motivating.19

  The motivating effect of being behind happens not only for the overall goal, it also happens for progress along the way. If our goal is to bring in ten new clients this month, and halfway through we’ve only brought in four, we’ll feel less satisfied than if we’ve already brought in eight. Being behind our ideal trajectory can motivate us to work harder.

  Competition affects motivation for similar reasons. Just as we use certain predetermined goals (thirty-three sit-ups or ten new clients) to determine whether we are succeeding, we often use others as a standard of comparison. Winning a basketball game doesn’t depend just on how many points your team scores; it depends on scoring more than the other team. Is 1,074 kilowatt-hours a lot of energy to use in a month? Hard to say, but if someone’s neighbors are using less than that, people may be motivated to close the gap.

  Sometimes there is a clear and compelling reward for doing better than others. Whoever makes the most sales calls gets a bonus. Whoever shoots the lowest wins the golf tournament.

  Other times, the reward is just the feeling of achievement. Winning is more satisfying than losing. Using less energy than your neighbor feels better than using more.

  Consequently, being behind others can motivate us to perform better. Teams losing by one at halftime came out of the locker room fired up. They played hard and erased most of the deficit in the first few minutes of the second half. Just like the people pressing the A and B keys in our experiment, being behind motivated the players to work harder. And as a result, their teams were more likely to win.

  But is being behind always more motivating?

  WHEN LOSING LEADS TO . . . LOSING WORSE

  Richard “Pancho” Gonzales was one of the best tennis players of all time. Born in Los Angeles, California, in 1928, Gonzales is one of the game’s few Mexican-American stars. His mother gave him a fifty-one-cent racquet when he was twelve years old and he never looked back. Largely self-taught, he learned to play by watching players at the nearby public courts. He was 6’3” by the time he reached nineteen years old, and his height helped him develop a dominating serve that overpowered opponents.

  Gonzales was ranked the best player in the world for a record eight years in a row. He won seventeen major singles titles over the course of his career, including two Grand Slams. When the editors of Sports Illustrated picked their favorite athletes of the twentieth century, they ranked Gonzales fifteenth, saying that if the fate of the earth was on the line in a tennis match, Gonzales would be the man humankind would want serving.

  One of Gonzales’s most unusual matches, though, was in 1969 against Charlie Pasarell at Wimbledon. Gonzales was a forty-one-year old at the time and a grandfather. Pasarell was not only much younger (twenty-five) but had trained under Gonzales, learning his technique by copying the older man’s strokes.

  The match started with each player holding his serve. When Gonzales served, Gonzales won the game. When Pasarell served, Pasarell won the game. This went back and forth. First for five games, then ten, then fifteen. Numerous times Gonzales saved set points to avoid defeat. Twenty games, then thirty, then forty. Finally, with a lob to the back edge of the baseline in the forty-sixth game, Pasarell broke Gonzales’s serve. The first set was his, twenty-four games to twenty-two.

  The second set began a little after seven p.m. It was a gloomy day in London and the light was fading. Gonzales complained about the deteriorating visibility, but the tournament referee ignored him. Whether because he was angry, or couldn’t see, Gonzales lost again, but much faster this time, 1–6. Play was called after the end of the second set.

  The next morning proved better weather, and the players returned to the closely fought contest. Gonzales bent repeatedly but never broke, and the sets crept upward, 6–6, 8–8, and 10–10. Pasarell soon began to feel the pressure of trying to finish off his former mentor. After twenty-nine sets, Pasarell double faulted twice and lost the third set 14–16.

  At this point the tide had begun to turn. Pasarell double-faulted again and lost the fourth set 3–6. Now the match was tied, two sets to two. Gonzales looked tired, leani
ng on his racquet between points and stalling for time. But he would not give up. Pasarell had him on the ropes time and time again, but couldn’t push it through. Gonzales was serving at 4–5 down 0–40 but the lob shots that had worked so well for Pasarell earlier in the match began to falter. Gonzales fought back, and seven deuces later, he won to tie the score at 5–5.

  Pasarell won the next game, but Gonzales again came from 0–40 to tie things at 6–6. Again the momentum shifted back and forth as the game tally went higher and higher. Eventually Gonzales won the final 11 points to win the set 11–9, and the match.

  The contest had lasted more than 5 hours and spanned more than 110 games. It’s one of the longest singles matches in the history of Wimbledon.

  * * *

  Based in part on this epic contest, Wimbledon introduced the tiebreak in 1971. Rather than playing game after game until someone gets ahead by two, for sets tied at six games each, a tiebreak game determines the winner.IV Players alternate serving and whoever scores 7 points first wins (as long as they have 2 points more than their opponent). Tiebreak games can still go on for a while, but they decrease the chance matches go on for as long as Gonzales and Pasarell did.

  Similar to our work on basketball, an economist wondered how losing in tennis would affect performance.20 Take a player who loses a tiebreak: Does that loss influence how well he plays the rest of the match?

  After analyzing thousands of matches, he found that the answer was a resounding yes. But the impact is actually the opposite of what we found in basketball. Rather than leading players to do better, losing led tennis players to do worse. Players that lose a first-set tiebreak lose an extra game, on average, in the second set.

  Why would that be?

  It’s tempting to attribute the differing impact of losing to distinctions between the two sports. Basketball is a team sport, while tennis is an individual one. Basketball games last less than an hour, while tennis matches often go on for two or three times as long. There are a number of other differences.

  But it turns out that the disparity has less to do with distinctions between basketball and tennis and more to do with the size of the discrepancy, or how bad the losers were losing.

  People get more motivated as they get closer to their goal. Take the cards you get at coffee shops, bagel stores, or as part of other loyalty programs. These cards reward frequent patronage with free stuff. Buy nine coffees, get the tenth free. Every sixth bagel is complimentary. Rewards like these encourage people to return to the store, but how motivating they are depends on how close people are to achieving the reward. Compared to people who have just started the card, people who have almost completed it buy much faster.21 Feeling like we’re almost there makes us more motivated, so we come back to the store sooner.

  Animals show the same pattern of behavior.22 Compared to rats that just started running a maze, those close to reaching a reward (cheese, for example) run faster. The closer they are, the more motivated they become.

  In competition, then, it’s not just about being behind. It’s about how far behind someone is. Being down by a little is often more motivating than being down by a lot because people are closer to achieving their goal of winning.

  Take a team losing by one point at halftime. They’re almost there. They’re just like the rat that runs around the corner and sees the cheese. If they play good defense, and hit an extra shot, they can close the gap. If they give just a little bit more effort, they can go from losing to winning. As a basketball announcer might say, they’re so close they can almost taste it.

  Compare that to a team that’s further behind. Say, losing by 8 points. They’re still in the game, but they’re not almost there. They have to make a number of stops on defense, make a number of extra shots, and maybe even go on a run. There is a lot between them and winning. They may be able to smell winning, but they’re too far away to taste it.

  When we’re further back, it’s harder to muster that extra motivation. The team down by 8 would still like to win, but they are so far behind that winning seems less likely. And it’s harder to encourage that extra effort if we’re not sure it will make much of a difference.

  Along these lines, social comparisons not only increase motivation, they can also decrease it.

  Rather than being down by 8 points, imagine being down by 20 or 25. You’re so far back that the chance of winning seems remote. So many things would have to go right for you to catch up that you doubt it’s even possible. So you begin to give up. In situations where success starts to seem impossible, motivation decreases. Competition becomes demotivating.

  And that’s what happened to tennis players who lost the tiebreak. Winning the match didn’t become impossible, but it became a lot more difficult. For best-of-three set matches, two sets is enough to win. So someone who just lost the first set tiebreak went from being almost halfway to winning the match to being halfway to losing it. They’re not just behind by a little, they’re behind by a lot.V

  This rapid shift in relative performance should be particularly demotivating. Sure, being behind doesn’t feel good, but it feels particularly bad when you were almost ahead by a bunch and then lost it. It’s like thinking you’re the top choice to be promoted and then finding out you’re actually at the bottom of the list. Being at the bottom never feels good, but it’s particularly bad when the top seemed oh, so close.VI

  * * *

  Not surprisingly, being too far behind can also lead people to quit.23 To give up and stop trying altogether.

  But interestingly, being far behind others isn’t the only reason people quit. Quitting also depends on how well people expected to do compared to others in the first place.

  In tennis, one player is often designated the favorite. They are ranked higher based on their recent performance in other matches. Similarly, compared to an upstart no one has ever heard of, most people expect an incumbent politician to win (as long as his or her tenure in office has been good).

  But while favorites should perform better, this expectation often brings along additional baggage. People expect them to do well and this makes the potential of losing (and violating those expectations) even worse. If an underdog loses, it doesn’t reflect that badly on the underdog. They were expected to lose, so losing doesn’t change how people see them. But if the favorite loses, it has a more negative impact on others’ impressions. They were expected to win, and anything less signals that the favorite might not have been so great after all.

  Consequently, competitors may search for a way to self-handicap. An excuse in the event of poor performance.

  Someone worried about blowing a big presentation, for example, might paradoxically stay out late the night before because it creates a handy external attribution for failure. If the presentation goes badly, he has an excuse. Rather than indicating something about their ability, there is now another explanation for any potential failure: If I hadn’t been out late, I would have done just fine.

  Quitting serves a similar function. Rather than sticking it out and losing, quitting allows competitors to preserve the notion that if they had just kept going, they would have won. That they were actually the stronger competitor, even though it didn’t play out that way in the end.

  Researchers found that favorites are more likely to quit for exactly this reason.24 Compared to underdogs, tennis players favored to win were more likely to quit mid-match. Players who were ranked higher going in were more likely to throw in the towel, both literally and figuratively, particularly if they lost the first set.

  For players who were supposed to win, but now looked like they might actually lose, quitting became a way to save face.

  * * *

  People and organizations often drop out of competitions. Basketball players pull up lame after shooting a jumper and sit out the rest of the game. Politicians drop out of the race to spend more time with their families. Companies take their name out of consideration for a contract to focus on other strategic priorities.
>
  In some cases, quitting is warranted. The player hurt their leg, the politician loves their family, and the contract just wasn’t in line with where the company saw its business going.

  But in other cases, quitting provides a clever defense mechanism that enables people to avoid failure. It allows us to preserve the notion that we could have been successful if we had tried. That if we had just kept competing, and pushed through it, we would have emerged victorious.

  PUTTING SOCIAL INFLUENCE TO WORK

  Where do these thoughts lead? Whether trying to inspire a sales team to work harder or encourage students to learn more, social comparisons can be a powerful motivating force. Giving people a sense of how they stack up against their peers can encourage them to work harder and be more likely to achieve their goals. At the same time, though, if not carefully designed, social comparisons can lead people to get disheartened, give up, and quit.

  Unfortunately, many companies and classrooms use a winner-take-all model. The person who makes the most sales this quarter gets promoted. The top student is named valedictorian and speaks at graduation.

  While this strategy motivates people who have a chance at the top slot, it often demotivates those who feel they have no shot at winning. Someone who has only half as many sales as the leader may think they are so far back that they just give up. Students that are getting Cs or Ds may feel similarly. Getting an A seems impossible, so why keep trying?

  One way to encourage perseverance is to shrink the comparison set. Breaking larger groups up into smaller ones based on performance. Golf tournaments organize participants into groups of similar skill. This encourages golfers to compare themselves to others of similar ability, which decreases the chance they feel far behind and helps maintain motivation.