What about ethics? Could it be that U.S. players are more upstanding? If that were the case, we might expect American players to obey drug rules across the board. Yet they don’t. Looking at suspensions of players for recreational drugs—drugs of abuse or recreation that have no performance-enhancement benefit, such as marijuana†—we find that almost all recreational drug use occurred among U.S.-born players. To date, no Dominican or Venezuelan baseball player has been suspended for recreational drug abuse.‡ Among the drug suspensions handed down to players from these countries, all were for performance enhancement. Among U.S. players, we found no relation between recreational drug use and economic background. Players from poorer neighborhoods were not more likely to use recreational drugs—just PEDs.
The stigma of using drugs may indeed differ across countries, and there may be a self-perpetuating effect. The greater proportion of Latin players who use such drugs, the less stigma it carries. The potential shame from such use, in theory, is lessened when “everyone else is doing it.” And the more players who use banned substances, the more pressure it puts on others to cheat. If your class is being graded on a curve and you know many peers plan to cheat on the upcoming test, you’re probably more tempted to cheat, too, just to keep up.
What begins as a result of economic incentives can grow into a self-reinforcing culture in which drug use is not only tolerated but expected. Soon athletes aren’t cheating to get an edge; they’re cheating simply to avoid falling behind (see professional cycling).
How big an effect do performance-enhancing drugs have? Answering this is difficult because it’s hard to know the counterfactual: What would a player’s performance have been had he not taken drugs? Although we don’t claim to have a definitive answer, here’s some empirical evidence. We looked at all major and minor league players and asked a simple question. Controlling for as many variables as possible—age, experience, height, weight, position, country of origin—how do the players who have been suspended for drug violations stack up against other players? What level within baseball’s league system (from the Rookie league to Single-A, Double-A, Triple-A, and the Majors) did they achieve relative to other players with the same characteristics who didn’t test positive?
As we discovered, suspended players were much more likely to achieve a higher level in the baseball league system. Specifically, an athlete suspended for using performance-enhancing drugs was 60 percent more likely to achieve the next level than an athlete who wasn’t suspended. To gain some perspective on this, an additional year in a league increases the chances of making it to the next level by only 20 percent. So performance-enhancing drugs have three times the effect of an extra year of experience. For players born outside the United States, PEDs have an even greater effect, increasing the chance of a player moving up to the next level by almost 70 percent. In addition to the science, the data support the claim that steroids work.
To make this point another way, imagine a set of high school seniors of the same age, quality of education, and grade-point-average. Group A cheated on the SAT, and Group B didn’t. If on average the students in Group A were admitted to higher-ranked colleges, you would reach the inexorable conclusion that cheating on the SAT had a benefit.
Could the reason that players caught using PEDs have achieved a higher level be simply that enforcement and drug testing are toughest at the highest levels in baseball? In other words, are baseball players cheating equally at all levels, but there’s simply more scrutiny at the top?
In fact, just the opposite is true. There seems to be more vigilance at lower levels of baseball. The Major League Baseball Players Association, which has fought rigorous testing, represents only players in the Major Leagues. Without the full protection and advocacy of the union, minor leaguers have been subject to considerably more stringent testing—from a longer list of banned substances to more frequent testing. Not surprisingly, minor leaguers fail drug tests twice as frequently as do their Major League counterparts.
In light of what we’ve learned, it’s also not surprising to find that, controlling for country of birth, age, and position, the height and weight of players caught using PEDs are much lower than those of other players. Shorter players and players who weigh less are far more likely to use a banned substance. Intuitively, this makes sense, and it also makes sense from an economic perspective. If performance-enhancing drugs improve power, for instance, it is the stature-challenged who would seem to benefit the most from such a boost. This evidence also seems to fly in the face of the “recovery from injury” claim. Unless recovery time is correlated to physical size—which no respected medical professional seems to claim—the explanation for the relationship between PED use and an athlete’s frame is that players are taking it for power and mass, not to shorten recovery time from injuries.
In a lot of ways, athletes who use performance-enhancing drugs behave like borrowers in a financial market, leveraging an asset with risky debt. Plenty of market speculators invest more than 100 percent of their money by using leverage. If they have $10 and borrow $20 from the bank, they can invest $30, three times the original investment. The downside is that they have to pay the borrowed money back later (with interest), which exposes them to more risk: They could lose more than their original $10. When an athlete leverages his effort with PEDs, he is in effect taking his initial genetic inheritance—his body, his talents, his training, and his work ethic—and buying on credit. If his investment pays off, he’ll exceed what he started with. If his marker is called and he can’t cover his debts—that is, if he fails a drug test—he stands to lose not only his initial investment (his natural health) but his reputation and future as well. For some athletes, the risks and costs are too severe and outweigh the potential benefits. For others, including older athletes and players from poorer backgrounds, the potential benefits outweigh the costs.
Seen through an economic lens, many examples of drug cheating and deception—however deplorable and amoral we might find them—start to make sense. How many millions of Americans massage their income tax returns, calculating that the gain outweighs the potential penalties and the risk of detection? How many people better their prospects for a choice job by doctoring their credentials, reasoning that the worst thing that could happen would be to lose a job that wasn’t theirs to begin with?
We spoke with Welington Dotel in the spring of 2010 as he was awaiting assignment with the Mariners. He expected to play for either the Clinton (Iowa) LumberKings of the Midwest League or the Everett (Washington) AquaSox, where he had spent the majority of the previous season. Sometimes he felt tantalizingly close to his dream of making the Majors. Other times his goal felt unfathomably distant. Money issues continued to loom large in his life. Having spent most of his signing bonus, he was now subsisting on a minor league salary of a few hundred bucks a week and living with a host family at every minor league outpost. He often phoned his family in the Dominican Republic with news of his baseball career, but they’d yet to see him play a game in the United States. “Too expensive,” he explained. “Maybe they come when I make the Majors!” It is not so hard to see why he might have felt compelled to bend the rules.
As we see it, fans are well within their rights to condemn baseball players for the steroids era. From Mark McGwire on down, the cheaters ultimately made the decision to cheat. But it bears pointing out that Major League Baseball and the players’ association created a system that gave many players the choice between acting immorally and, at least from an economic perspective, acting irrationally.
* The “Report to the Commissioner of Baseball of an Independent Investigation into the Illegal Use of Steroids and Other Performance Enhancing Substances by Players in Major League Baseball” was the culmination of a 21-month investigation that produced the 409-page report released on December 13, 2007. The report named 89 players alleged to have used PEDs.
* We look only at players in the U.S. minor leagues (Rookie, A, Double-A, and Triple-A leagues) and exclude
any suspensions from amateur leagues in other countries, as those leagues may have different drug and enforcement policies.
* Specifically, we used census data on education, employment, and income for the city or MSA (metropolitan statistical area) in which the player was born in the decade in which he was born.
† One could claim that a few drugs of abuse might also enhance athletic performance. But most of the positive tests were for marijuana, which would not be performance-enhancing in almost any competition we could think of, the Nathan’s Famous Fourth of July International Hot Dog Eating Contest notwithstanding.
‡ Kengshill Pujols, however, came close. A Dominican-born pitcher in the Dodgers organization, Pujols was already serving a 50-game suspension for PEDs in 2006 when he was arrested and found to be carrying 118 baggies of crack cocaine in his underwear. Charged with drug possession, he was released by the Dodgers and eventually convicted.
DO ATHLETES REALLY MELT WHEN ICED?
Does calling a time-out before a play actually work?
Plenty of field goal kickers have had rough games, but few have been so spectacularly bad that they’ve inspired an entire Saturday Night Live skit. It was during the 2005 NFL season that New York Giants kicker Jay Feely missed three potential game-winning field goals against the Seattle Seahawks. The Giants ended up losing 24–21. A few weeks later, SNL served up “The Long Ride Home: The Jay Feely Story.” Feely, played by Dane Cook, is traveling aboard the team plane when the flight hits turbulence. He is asked to land the aircraft between two skyscrapers. Naturally, he drives it wide right.
Midway through the 2008 NFL season, it appeared as though Feely was prime for another round of mocking. By then, he was kicking for the other New York team, the Jets. With three seconds to go in an October road game against the Oakland Raiders, Feely trotted onto the field for a 52-yard attempt to send the contest into overtime. Feely went through his routine, struck the ball fairly cleanly, but doinked the kick off the goalpost. The Raiders crowd celebrated. Jets fans groaned.
But wait. Oakland’s coach, Tom Cable, had called time-out before the kick, a spasm of psychological warfare. Feely’s attempt didn’t count. After a brief pause, Feely tried again. This time, he drilled the ball through the uprights, sending the game into overtime. Afterward he explained that he happily welcomed the Raiders’ time-out call. “I heard the whistle before I started, which is an advantage to the kicker. If you’re going to do that, do that before he kicks,” he said. “I can kick it down the middle, see what the wind does, and adjust. It helps the kicker tremendously.”
The Raiders ended up winning the game in overtime—ironically, on a field goal of 57 yards, a heroic distance. But Feely was redeemed. And it was another bit of evidence that questioned the wisdom of “icing the kicker.”
For decades, it’s been an article of faith in the NFL that when Team A faces a pressure-infused field goal to tie or win a game, Team B calls a time-out to “make him think about it” or “plant seeds of doubt.”
But does it work? Does icing the kicker increase the likelihood of a miss? Several years ago, two statisticians, Scott Berry and Craig Wood, considered every field goal attempt in the 2002 and 2003 NFL seasons, playoffs included. They paid special attention to “pressure kicks,” which they defined as field goal attempts in the final three minutes of regulation or at any point in overtime that would have tied the game or given the kicking team the lead. Publishing their results in the journal Chance, Berry and Wood asserted that on pressure kicks between 40 and 55 yards, iced kickers were 10 percent less successful on average. (On shorter kicks, the effect was found to be negligible.) However, the statistical significance of the difference found—amounting to 4 kicks out of 39 attempts—has been questioned.
Nick Stamm of STATS, Inc., found that pressure kicks—defined as above except within the last two minutes of the game rather than the last three—in the NFL regular season from 1991 to 2004 showed an insignificant difference between iced and non-iced kicks. The conversion rate on iced kicks was 72 percent; for non-iced kicks, the rate was 71.7 percent. Stamm’s work suggests that at best, icing the kicker does not diminish his chances of success.
We undertook our own study, using NFL data from 2001 through 2009 and using Stamm’s standards for pressure kicks as well as some of our own—looking at kicks in the last two minutes, one minute, 30 seconds, and 15 seconds of the game. First, we looked at instances when the team on defense called a time-out right before the kick (icing the kicker) and compared that with instances when they didn’t. We then controlled for the distance of the field goal attempt so that we could compare the same field goal from the same distance when one kicker has been iced and the other hasn’t. Simply put, we found that icing made no difference whatsoever to the success of those kicks. NFL kickers being iced are successful from the same distance at exactly the same rate as kickers who are not iced. The following table shows the numbers.
NFL FIELD GOAL SUCCESS WHETHER OPPONENT CALLS A TIME-OUT OR NOT
In some instances, icing the kicker may exact a psychological price. In other instances, it may backfire, as it did with Jay Feely, giving the kicker the equivalent of a free dress rehearsal. In the vast majority of cases, the kick will be successful based simply on mechanics and how cleanly the ball is struck (and how well the ball is snapped and placed by the holder), not on whether the kicker had an extra 90 or so seconds to consider the weight of the occasion.
Former Tampa Bay Buccaneer kicker Matt Bryant in an interview with the Tampa Tribune summed it up this way: “I think when you’re at this level, nothing like that should matter. If it does, you probably don’t belong here.”
Kickers aren’t the only athletes opposing coaches try to ice. Take the waning seconds of an NBA game. A team is whistled for a foul. Just as the free throw shooter steps to the line, the opposing team reflexively calls a time-out to ice the shooter.
Just like icing the kicker in the NFL, it’s a dubious strategy. We examined all free throws attempted in the last two minutes of the fourth quarter or overtime of all NBA games from 2006 to 2010 when the teams were within five points of each other—in other words, “pressure” free throws. When a time-out was called just before the free throws (icing), the shooter was successful an average of 76 percent of the time. When there was no time-out called, the free throw percentage was … 76 percent.
Next, we looked at only the first free throw attempt—the one directly after the time-out—and ignored the second since a player might have adjusted his shot after the first attempt or his nerves might have settled. There was, again, no difference. We even looked exclusively at situations in which the score was tied and thus a made free throw would have put a team in the lead. Again, there was no difference between shooters who were iced and those who weren’t.
There might be valid reasons for a team to call a time-out before a high-pressure kick or free throw. The coach might want to devise a strategy to block the kick or set up a play in the event of a miss. The defensive team might want to create the appearance that it’s doing something rather than standing by idly. They might want to ensure that the rights-paying television network has the opportunity to air an additional series of commercials—annoying the fans in the process. But they shouldn’t expect the time-out to have any bearing on the subsequent play.
Icing doesn’t freeze a player or heat him up. You might call it a lukewarm strategy.
THE MYTH OF THE HOT HAND
Do players and teams ride the wave of momentum? Or are we (and they) fooled into thinking they do?
In the sprawling clubhouse of the New York Mets, David Wright’s locker is featured prominently, square in the middle of the room, near the front door. The symbolism is unmistakable. Wright isn’t just a spectacular young third baseman and a handsome, genial guy born without the jerk chromosome. He is the face of the franchise. So it is that his locker is positioned in such a way that the media can always find him for a quote, teammates can observe how profession
ally he comports himself, and Mets employees can locate him when they need to corral him to meet with corporate sponsors or tape a promotional video or sign the cast of a season ticket holder’s kid.
But early in the 2010 season there were few good vibes emanating from Wright’s double-wide clubhouse stall. He was struggling at the plate: smacking nothing but air with his swings, grounding feebly into double plays, and taking pitches for called third strikes. After going hitless in four at-bats the previous night, Wright arrived for a late April game against the Chicago Cubs hitting .229, a full 78 points below his career average of .307. Only 11 of his 48 at-bats had yielded a hit. As with most baseball players, his mood moved in step with his batting average. He still answered questions from the media but did so in a clipped manner, staring ahead vacantly and shifting uneasily in his chair. He cut back on the sponsor grip-and-grins and the video promos and the other extra-baseball obligations that fall to a franchise’s star player. He’d broken with routine and arrived at Citi Field early that day to take extra batting practice at the team’s indoor cage, hoping “to get my swing back to where I want it to be.”
Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won Page 22
