Tiger Woods and other golf superstars who stand to win millions on inch-long putts apparently are subject to the same fear and aversion to risk that can afflict investors and managers. Taking the safe route, however, has its own costs, according to new Wharton research.

In a working paper titled, Is Tiger Woods Loss Averse? Persistent Bias in the Face of Experience, Competition, and High Stakes,”  (PDF) Wharton operations and information management professors Devin Pope and Maurice Schweitzer examine putts during pro golf tournaments and determine that even the best golfers systematically miss the opportunity to score a “birdie” — when a player sinks a ball in one stroke less than the number of expected strokes for a given hole — out of fear of having a “bogey” — or taking one stroke more than what is expected. According to the researchers, for many, the agony of a bogey seems to outweigh the thrill of a birdie.

The researchers calculate that this type of decision-making bias costs the average professional golfer about one stroke during a 72-hole tournament. For the top 20 golfers, that translates to a combined loss of about $1.2 million in prize money a year. According to the paper, golfers frame their approach to putting based on the risk of coming in worse than “par” — or the number of strokes a professional golfer would be expected to take to complete a given hole. The researchers’ analysis shows that golfers avoid the possibility of loss by playing conservatively when they have the opportunity to do better than par, but will try harder if they are at risk of coming in worse than par.

Pope says that the study explores “loss aversion” — a bias in decision-making that is an important element in the growing field of behavioral economics, which explores how human psychology impacts markets and business. “This research provides evidence that people work especially hard in order to avoid losses,” says Pope. In the current economy, he adds, “a lot of people … are very determined to get out of the loss domain and return to where they were a couple of years ago.”

The study’s findings challenge theories that suggest bias in decision-making does not persist in markets and that when biases exist, they tend to erode with competition, large stakes and experience. Where better to examine the theory than the competitive, high-payout world of the PGA (Professional Golfers’ Association) Tour?

“The [study] demonstrates judgment biases in a high-stakes setting with experts,” says Schweitzer. “The bottom line is this: If Tiger Woods is biased when he plays golf, what hope do the rest of us have?”

The Dreaded Square

Many people assume that experts and professionals don’t show biases in a specialized area to the degree that a person without experience in that subject would exhibit, Pope says. “People assume CEOs have less biases than their employees and yes, that’s probably true, but it doesn’t mean CEOs don’t have biases as well.”

The researchers based their analysis on a rich dataset compiled by the PGA. Each year, the association’s tour is made up of 40 to 50 individual tournaments in which about 150 golfers compete. The four-day tournaments paid out, on average, a total of approximately $5 million in 2008. The top two-thirds of tournament players share in the purse, with the winner typically receiving 18%.

The data was compiled by approximately 250 workers hired by the PGA to gather detailed information about tournament play. To get the data, the tour mounts lasers around each hole of a course to measure and record within an inch the coordinates of each ball after every shot. Pope and Schweitzer used data from 239 tournaments between 2004 and 2009, concentrating on 2.5 million putts attempted by 421 professional golfers who each made at least 1,000 putts.

To focus on the issue of loss aversion, the researchers examined the putts in the context of the par set for each hole. According to the researchers, the par number creates a reference point that clearly distinguishes a loss from a gain. On their scorecards, golfers circle holes they score under par. If they shoot over par, the score is framed in a dreaded square. “Though golfers should only care about their overall tournament score, golfers may be influenced by the salient, but normatively irrelevant, reference point of par when they attempt putts,” the authors write.

Most of the putts in the data were for par (47%) or birdie (39.8%). The approach to each hole taken by golfers, relative to par, provides a way to measure loss aversion. Pope and Schweitzer used the PGA data to determine whether a golfer was playing it safe by making a putt that would end up just in front of the hole, in order to set up a sure next shot. Using data measuring the force of a stroke and position of a ball before the putt, the researchers determined that, on average, golfers make their birdie putts approximately two percentage points less often than they make comparable par putts. “This finding is consistent with loss aversion; players invest more focus when putting for par to avoid encoding a loss,” the researchers write.

Schweitzer notes that the behavior reflects the bias toward avoiding loss — in this case, missing par and scoring a bogey — over the potential to score more in the overall tournament, which is what ultimately matters. “Loss aversion is the systematic mistake of segregating gains and losses — evaluating decisions in isolation rather than in the aggregate — and over-weighting losses relative to gains,” he says.

In a business context, par might be equated to quarterly earnings or investors’ approach to selling or holding on to stocks depending on how much they paid for the shares initially. “In golf, players evaluate their performance on individual holes rather than focusing on their overall performance,” Schweitzer says. “In business, people look at performance in a particular quarter, how a specific stock has performed or how a particular account has performed. People make mistakes when they view related decisions independently.”

Pope and Schweitzer note that their research results are in line with Prospect Theory, an important concept in economics which was developed by psychologists Daniel Kahneman and Amos Tversky in 1979. Prospect Theory predicts that people become more risk averse when they are recording gains than when at risk of suffering a loss. Consistent with Prospect Theory, “[if] professional golfers use par as a reference point, they should be more cautious when putting for birdie (in the gain domain for a specific hole) than when putting for par. Specifically, conditional on missing a putt, we find that golfers hit birdie putts less hard than they hit par putts and are more likely to leave birdie putts short of the hole than par putts…. [We] demonstrate that players sacrifice success when putting for birdie to avoid difficult follow-up putts.” Moreover, this pattern of behavior “decreases expected profits” for the golfers, the paper concludes.

Overconfidence and Nervousness

Pope and Schweitzer attempted to head off challenges to their work by controlling for numerous factors. In their analysis, they control for the ball’s position on the green, whether the player is better with a driver than with a putter, whether experience on a particular green changes behavior, and differences between holes.

They also ruled out psychological explanations, including overconfidence and nervousness. To test for overconfidence, the researchers questioned whether after hitting a well-placed shot, golfers become “overconfident or cocky in a way that harms their performance on their next shot.” However, the data shows that when golfers score one stroke less on a hole relative to average performance, they are likely to score fewer strokes than average performance on the next hole. “Absent a story of loss aversion,” the authors write, “overconfidence or cockiness cannot account for our findings.”

As for nervousness, the paper posits the idea that golfers value a birdie putt more than they value shooting par, and that they might become nervous and “choke” when aiming for a birdie. The research found that although amateur golfers infrequently take birdie putts, professional golfers attempt nearly as many birdie putts as they do for par. Even when a successful birdie or par putt would place golfers in a similar position, they are still more likely to make the otherwise identical putt if it is for par.

The notion that nervousness plays a role is also inconsistent with other results. “First,” the paper states, “players hit their birdie putts shorter than they hit otherwise similar par putts. Second, we demonstrate that even the best golfers, including Tiger Woods, exhibit this bias in early rounds of tournaments. Third, the difference between par and birdie putts diminishes across rounds. Nervousness cannot explain these findings.”

Finally, in addition to the quantitative analysis, the authors cite anecdotal evidence from Woods himself. After playing a round in 2007, the PGA’s most winning active player said: “Any time you make big par putts, I think it’s more important to make those than birdie putts. You don’t ever want to drop a shot. The psychological difference between dropping a shot and making a birdie, I just think it’s bigger to make a par putt.”

The authors note that golf has been the subject of other recent economics research. According to Pope, he and Schweitzer were interested in exploring the subject because of the rich PGA dataset. The practice of recording par for each hole also provides a fixed reference point that does not change given varying circumstances, as might be the case in many other tests of economic theory.

While Pope does not golf at all and Schweitzer is only an occasional player, they say the subject is well-suited to their research goals. “It is a rich domain for studying individual decision making. We have data that measure the outcomes of hundreds of thousands of similar decisions that were made by talented, experienced and highly motivated individuals.”