In the end the Indianapolis Colts, holding the number one pick, took Manning. The San Diego Chargers took Leaf with the second pick and signed him to a five-year contract worth $31.25 million, including a guaranteed $11.25 million signing bonus, at the time the largest ever paid to a rookie—that is, until the Colts paid Manning even more: $48 million over six years, including an $11.6 million signing bonus.
You probably know how the story unfolded. Peyton Manning is the Zeus of NFL quarterbacks, a four-time MVP winner (the most of any player in history), a Super Bowl champion, riding shotgun on the express bus to the Hall of Fame. And Leaf? As we write this, he is currently out on bond as he defends himself against burglary and drug charges in Texas. (He was sentenced to probation after pleading guilty to illegally obtaining prescription drugs.) In the summer of 2009, he was arrested by customs agents as he returned from Canada, where he had been in drug rehab. He played his last NFL game in 2001, his career marked by ineffective play; injuries; toxic relations with teammates, coaches, and the media; and a general lack of professionalism. Leaf once complained of wrist pain to avoid practice but reportedly had played golf earlier in the day. Another time, while serving a four-game suspension for insubordination and ordered to rehabilitate a shoulder and wrist, he was videotaped playing flag football with friends.
After three disastrous seasons, the Chargers released Leaf, which actually broke with convention. Time and again, ignoring what economists call sunk costs, when a high draft pick underperforms, teams tend to keep investing in him for years in the hope that he’ll turn it around. The team rationalizes: We paid a ton to get him; we have to keep trying to make it pay off. But the money’s gone, and continuing to invest in the underperforming player is simply making a bad decision worse. Ever ordered a food item, bitten into it, and found it tasted awful? How many of us eat it anyway because, well, we paid for it? Or how often do we insist on holding on to a stock we bought for $50 that is worth $40 today and $30 tomorrow? Is the off-tasting sushi really going to taste better if we keep eating it? Is the company that is tanking really going to see its stock price return to $50? The money is gone; we may as well cut our losses now. But we hardly ever do. Everyone hates admitting a loss, football executives included.
To the credit of the Chargers organization, they learned from their costly mistake. In 2000, still recovering from the Leaf fiasco, the Chargers finished 1–15 and were “rewarded” with the top pick in the 2001 NFL draft. They traded the pick to Atlanta for the Falcons’ number five as well as a third-round pick, a second-round pick in 2002, and Tim Dwight, a wide receiver and kick return specialist. The Falcons used the top pick to select Michael Vick; the Chargers used that fifth selection on LaDainian Tomlinson, who would become the most decorated running back of his generation. To satisfy their quarterback needs, they waited until the first pick of the second round and tapped Drew Brees, who would go on to become the 2010 Super Bowl MVP, albeit for a different team, the New Orleans Saints. Remember, too, the Chargers were willing to give up Eli Manning in 2004 for what was effectively Shawne Merriman and Philip Rivers. Add up the values from both trades and they follow McCoy’s chart almost exactly. Imagine how much better the Chargers could have done if they’d known the chart was flawed.
Beyond the money, overinvestment in high draft picks can have other real costs. Pampering the first-round pick—treating him differently from the sixth-rounders who’d be put on waivers for a comparably dismal performance—exacts a price on team performance and morale. It also forestalls taking a chance on another athlete. The more chances given Ryan Leaf, the fewer chances afforded his backup. And it’s not just the team that has drafted the player that’s prone to this fallacy. Even after Leaf’s miserable performance and behavior in San Diego, three other teams gave him another shot. They recalled the player he was in college. They still coveted his size, strength, and athleticism and believed the hype. Never mind the clear evidence that he was a bust. “It’ll be different here,” they told themselves. Only, of course, it wasn’t.
How do teams know when they’re getting the next Peyton Manning and when they’re getting the next Ryan Leaf? They don’t. There are simply too many unknowns and too much uncertainty to know whether you’ve drafted a great player or a bust. The only certainty is that you will pay dearly for both. Manning and Leaf were both very expensive, but only one of them was able to perform. Also, the uncertainty regarding whether Manning was better than Leaf is not uncommon. Take a look at the following table and compare the number one picks over the last decade with the players voted offensive and defensive rookies of the year in the subsequent year as well as other players at the same position available in that draft who made the Pro Bowl.
Still think it’s easy to pick the best players? If you look at the top picks in the NFL draft from 1999 to 2009, not a single one was named rookie of the year on either side of the ball. More damning, many of the top picks have turned out to be busts. Of the last 11 number-one-picked players, eight have been quarterbacks. Four of them—Tim Couch, David Carr, Alex Smith, and the beleaguered JaMarcus Russell—came nowhere close to justifying the selection. Of the four remaining quarterbacks, it’s too early to tell what will become of Matthew Stafford in Detroit, and though Carson Palmer and Michael Vick have each been to the Pro Bowl, both have also spent considerable time on the sidelines, Palmer because of a gruesome knee injury and Vick because he was incarcerated for nearly two years and suspended from the league for his involvement in an illegal dog-fighting scheme. That leaves only one number one quarterback pick, Eli Manning, who has started the majority of games for his team since his debut.
But again, divergent as their careers have been, all the number one picks were paid handsomely. So for the teams selecting at number one, the best-case scenario is that you get a good player for an expensive price. You buy a Camry at Porsche prices. Worst-case scenario, you pay a lot of money and get nothing in return. You pay the price of a Porsche for a clunker. What you will never get is a great player at a cheap price. You never get the Porsche at the clunker price in the early rounds.
In the 2010 draft, the trend continued as the St. Louis Rams selected quarterback Sam Bradford with the number one overall pick (making 9 of the last 12 first picks QBs) and promptly signed him to the richest contract in history—five years at $86 million with $50 million guaranteed.
Even with successful high picks on the order of Manning and Palmer, the question isn’t how much they cost in terms of salary but also how much they cost in terms of the draft picks you could have taken instead. In 2005, the San Francisco 49ers drafted quarterback Alex Smith with the first pick; the only quarterbacks from the 2005 draft to have made a Pro Bowl are Aaron Rodgers (number 24 pick), who made it for the first time in 2009, and Derek Anderson, the eleventh quarterback taken that year. In 2000, defensive end Courtney Brown was chosen as the number one pick. He never made a Pro Bowl. But Shaun Ellis and John Abraham, the second and third defensive ends taken in that draft, did make numerous Pro Bowls. As did Kabeer Gbaja-Biamila, the twelfth defensive end taken that year with the 149th pick, and Adewale Ogunleye, who wasn’t even drafted that year, meaning at least 24 defensive ends were chosen before him. Bottom line: In football, it’s very hard to tell who is going to be great, mediocre, or awful.
So what should a team do if it’s blessed (which is to say, cursed) with a top pick? Trade it, as the San Diego Chargers learned to do. Drafting number 10 and number 11 instead of number 1 is a much better proposition. The Dallas chart shows that the value of these players should be the same, but the reality is quite different. Teams get far more value from having picks at number 10 and number 11 than they do by taking a chance on one pick at number one. With two “draws,” the chance of having at least one of the two picks succeed is much higher, and the cost is the same or less. Factor in the potential for injuries and off-the-field trouble and it becomes even more apparent that having two chances to find a future starter is a much better proposition than havi
ng only one. Also, the team avoids the potential of a colossal and public bust like Ryan Leaf. Even if the later picks flop, fans won’t care nearly so much as they will when the top pick is a bust.
Perhaps NFL owners have different objectives, but it’s safe to assume they want to win or make money, and probably both. Following the chart meant that they lost on both counts. They overvalued top talent and, even when a pick happened to pan out, paid dearly.
For a franchise willing to ignore convention and depart from the chart (or improve it), the payoff can be huge. A team that discovered the chart was flawed—that it overvalued top draft picks—could trade its high picks for many more lower picks. It wouldn’t be taking the sexy picks and exciting fans by drafting Heisman Trophy winners and standouts at the NFL combine, but as Massey and Thaler’s research shows, it would field better teams and win more games. If you look at the teams that did trade down in the draft or traded current picks for a greater number of future picks, the researchers showed that those teams improved their winning percentages significantly over the four years after each trade.
Over the last decade, two teams in particular went “off the chart,” as it were, and created a new model, placing less value on the top picks: the New England Patriots and the Philadelphia Eagles. Not surprisingly, those two teams have two of the top winning percentages and five Super Bowl appearances between them since 2000. Tom Brady, one of the few quarterbacks hailed as Peyton Manning’s equal, a former MVP and three-time Super Bowl winner? He was drafted in the sixth round of the 2000 draft with the 199th pick and thus was obtained cheaply, providing the Patriots extra cash to collect and keep other talent to surround him. Teams that traded current draft picks for future ones benefited in subsequent years, too, and again, the Patriots and Eagles were at the forefront. (Not coincidentally, their coaches, Bill Belichick and Andy Reid, have enough job security to afford the luxury of a long-term focus. Reid even has the additional title of executive vice president of football operations.)*
Which teams are on the other end of the spectrum, routinely trading up in the draft to get higher picks and overpaying for them? The answer is unlikely to surprise you: the Oakland Raiders and the Washington Redskins, who collectively have the fewest number of wins per dollar spent.†
As for the Cowboys, as we write this, their quarterback is All-Pro Tony Romo, who currently owns one of the highest passing ratings of all time. Never mind the Cowboys fleecing other teams with a pricing system for draft picks. After starring at tiny Eastern Illinois University, Romo wasn’t selected at all in the draft, so Dallas simply acquired him as a rookie free agent in 2003 and nourished him. One of Romo’s favorite targets, wide receiver Miles Austin, was also undrafted when he left tiny Monmouth University. He, too, was spotted by the Cowboys and signed as a rookie free agent. It seems the Cowboys may have found other ways to find value among new players outside the draft, deviating from the system they created.
* Ironically, Washington used those eight picks to remarkably bad effect, selecting bust after bust. Williams, meanwhile, led by the rapper Master P, negotiated perhaps the most lampooned contract in sports history, a deal laden with performance-based incentives, few of which Williams managed to meet, thereby sparing the Saints millions. So what should have been a disastrous trade for New Orleans was more or less a wash, equally bad for both parties.
* Bill Belichick and Andy Reid are also two of the least conventional coaches in terms of their play-calling (both go for it on fourth down more often than average) and are, of course, routinely criticized for it, especially when it fails.
† Ironically, the Redskins had extensive discussions with Massey and Thaler early on in their research, and the two professors met with team owner Dan Snyder and his football staff. After receiving the advice to go off the chart and trade down in the draft and give up current picks for future ones, the Redskins did exactly the opposite.
HOW A COIN TOSS TRUMPS ALL
Why American Idol is a fairer contest than an NFL overtime
It’s one of the great ironies in sports. For 60 minutes, the gladiators in the NFL risk life and head trauma, bouncing off one another, driving opponents into the ground, and generally purveying violence and mayhem. They’re caked in blood and sweat and dirt and grass. It’s all part of the spectacle that makes professional football so indefensible to some and so compulsively watchable to the rest of us.
And then, if the two teams are tied after regulation, these fierce and brutal struggles are decided largely by … the flip of a coin. After that gamelong physical exertion, the outcome ultimately comes down to dumb luck. Sure, the winning team has to kick the ball through the uprights or, in rare cases, march into the end zone, but that’s mostly a formality. Win that arbitrary coin toss at midfield and earn first possession of the ball in overtime, and victory is usually yours. As the broadcaster Joe Buck once quipped before the Oakland Raiders and San Francisco 49ers were about to begin overtime: “[Here comes] one of the biggest plays of the day, the coin flip!”
Credit Brian Burke, an aerospace engineer, a former F/A-18 carrier pilot in the U.S. Navy, and the current overlord of the website advancednflstats.com, for logging the hard miles here. Burke determined that between 2000 and 2009, 158 NFL games, including the playoffs, went to overtime. Two of those games ended in a tie. In one game, the Detroit Lions won the coin flip and chose neither to kick nor to receive but rather what side of the field they preferred to defend. (Not surprisingly, they lost.) In the other 155 games, the team that won the coin flip won the game 96 times, a 61 percent clip. As Burke correctly pointed out on his website, “Don’t be tricked by people that say ‘only 61 percent.’ If we agree 50 percent would be the fairest rate, you might think 61 isn’t very far from 50. But that’s not the right way to look at it. The appropriate comparison is 61 percent versus 39 percent, the respective winning percentages of the coin flip winners and losers. That’s a big advantage—over 3:2 odds.”
What’s more, in 58 of the 158 games, or 37 percent of the time, the team that won the coin flip won the game in its first possession. Think about this for a second: Teams battled their guts out for 60 minutes over four quarters and were tied with their opponents. Then, 37 percent of the time, one team lost in overtime without even touching the ball. Is it any wonder that of the 460 coin toss winners in NFL history, only 7 of them have elected to kick off and play defense first in overtime play?
David Romer, the Berkeley economist who encouraged more teams to go for it on fourth down, has a way to make NFL overtimes fairer: Change the spot for the initial kickoff. As it stands now, the kicking team boots from the 30-yard line. At this distance, it’s difficult to kick the ball into the end zone for a touchback, so the receiving team often gets the chance for a strong return. Romer claims that moving the kickoff up just five yards to the 35 would trigger a significant increase in touchbacks so that the receiving team would begin at the 20-yard line, the “break-even point” where the team on offense and the team on defense are equally likely to score next.
Chris Quanbeck, an electrical engineer and rabid Packers fan, offers a more radical, and intriguing, suggestion: Auction off the first possession of overtime, using field position as currency. Want the ball first? How far back are you willing to start your first drive? If we accept Romer’s premise that the 20-yard line is the break-even point, if Team A is willing to start on its own 15-yard line, Team B might happily agree to start out on defense. Writing in Slate, Tim Harford, a columnist for the Financial Times, noted this additional benefit: “Imagine the possibilities for stagecraft.… The two head coaches could come to midfield with sealed bids, with the envelopes to be opened by a cheerleader representing each team—a gridiron version of Deal or No Deal.”
Others have suggested eliminating field goals in overtime and mandating that the winning team must score a touchdown. As kickers have improved their accuracy and leg strength, it’s become increasingly easy for the team that wins the coin toss to reach field goal r
ange. What’s more, eliminating the field goal would encourage the kicking team to inaugurate the overtime by attempting an onside kick. (As it stands, it’s a foolish play. If the receiving team recovers, they’re likely to be only 10 or 15 yards away from being in position for a game-winning field goal.)
The other obvious solution would be to adopt some variation of the “Kansas Plan” in college football, whereby each team receives a first-and-ten possession at the opponent’s 25-yard line. Here, at least the team unlucky on the coin toss gets the equivalent of “last licks”—it can’t lose without touching the football on offense.
In response to an Internet discussion, one reader suggested simply letting the fans vote for the winner, in the manner of American Idol. The poster was being facetious, of course. But is it that much more ridiculous than deciding a game—a tightly contested game at that—largely on the basis of heads or tails?
There are other sports that employ the flip of a coin, if not to such dramatic effect. In professional tennis, the winners of the pre-match coin flip have four choices: They can serve, return, choose one side of the court, or forgo the choice entirely. The overwhelming majority of players opt to serve first, and this makes sense. In ATP matches in 2009, servers won 78.4 percent of the time.
At the 2010 NFL owners’ meetings, the league passed a change to the overtime rule in playoff situations. The team losing the coin toss will have a chance to score if the opposing team kicks a field goal. But if the team that wins the coin toss scores a touchdown on its first possession, the game will be over. If both teams exchange field goals, sudden death commences, with the first team to score again (even if it’s just a field goal) winning. Teams voted the modification in by a margin of 28–4, in part because of the data showing how often the initial random overtime coin flip determines the game’s outcome. “Plenty of people on the committee, myself included, are so-called traditionalists,” Bill Polian, the Indianapolis Colts’ president, told reporters. “I am proud to be one. But once you saw the statistics, it became obvious we had to do something.”
Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won Page 20
