Come draft day, however, Savage didn't heed his inner caddie. Instead, he grabbed his driver and took the third-largest swing of the weekend: Michigan receiver Braylon Edwards, who scouts say drops too many passes, but should fetch $20 million in guarantees. Last year's third overall pick, Larry Fitzgerald, banked as much from the Arizona Cardinals.
The stroke may also have cost him more than his shirt: The Washington Redskins reportedly offered multiple picks for the one Cleveland used to select Edwards.
Rejecting a trade was precisely the wrong move, say Cade Massey and Richard Thaler, professors of economics at Duke and the University of Chicago, respectively. Their research suggests that it won't matter if Braylon Edwards becomes the next Paul Warfield: The Browns, needing both, will have overpaid.
"You can't judge [Savage] too harshly," says Massey. "He's just doing what everyone else does in his situation."
Massey and Thaler co-authored The Loser's Curse: Overconfidence vs. Market Efficiency in the National Football League Draft, using years of data on draft trades, player performance, and compensation to tackle the question: Are the top picks really worth it?
They found -- get this -- that the last pick in the first round is worth more to his team than the first. So highly valued and overpaid are top picks that no matter how well they perform, they're worth less than any second-round choice too.
The premise is simple: Because the NFL imposes a salary cap, giving too much to the unproven few means the rest of the roster must be built on the cheap.
In 2003, for example, the last year that Tim Couch, Courtney Brown, and Gerard "Big Money" Warren were Browns, their combined cap figure -- $16 million -- swallowed more than 20 percent of the team's $75 million limit. The team simply couldn't afford to pursue more proven free agents like tackle Wayne Gandy, linebacker Shawn Barber, or cornerback Dre Bly, any of whom could have been significant upgrades.
The result? Five wins.
Though the Browns recently have shown no qualms about tipping for gristle, wise teams, say Massey and Thaler, should not use their high first-round picks. Rather, they should exploit those willing to pay, trading down the first round for more picks in the second and third, where a player's "surplus value" -- his bang per buck -- is greater.
"Both the 9th pick and the 25th pick are individually worth more than the third pick," says Thaler. "Not trading the third pick was the biggest mistake they could make." (The Browns declined repeated interview requests.)
Does this suggest that the Browns, who passed on Donovan McNabb, Corey Simon, and Richard Seymour for Couch, Brown, and Warren, now merit a case study? Should the professors' book title be revised to The Lerners' Curse?
"Actually," says Massey, "my first reaction was that we'd solved the mystery of the Bengals."
Nine times since their last winning season -- 1990 -- Cincinnati has exercised a top-10 pick. The Bengals earned them, losing exactly twice as many games as they've won (80-160).
What's ironic is that since the Browns' return in 1999, each team has made four top-10 picks. The Bengals have two more wins.
It gets hairier. Dwight Clark and Carmen Policy, hired in 1998 by Al Lerner to build the new Browns, brought with them a combined 34 years of service with the San Francisco 49ers -- too many, apparently, to remember Bill Walsh's legendary 1986 draft.
That year Walsh made six trades, including his first-rounder, to stockpile picks in the third and fourth rounds. Eight of the first nine players he selected were key to capturing that decade's last two Super Bowls. Walsh himself was struggling to keep pace with the Washington Redskins, Super Bowl contestants four times between 1982 and 1991. Their general manager, Bobby Beathard, traded away first-round picks seven times in that span.
Even Bill Rees, hired by Savage to serve as the team's director of player personnel, witnessed Walsh turn the third pick in 2000 into six picks over the first three rounds, four of which became defensive starters.
"It's not just the Browns," says Massey. "Those lessons are available for everybody, but seem to have been learned by very few."
Choosing the best players, he says, means predicting the future and bidding competitively. They're not simple tasks, but, as the data shows, they're just performed by simpletons.
There have been 70 drafts now, but NFL teams have been slow to learn. They're overconfident in their own intuitive assessments and overestimate how strongly other teams share them. All are behaviors consistent with psychological research on irrational decision-making.
Of course, you don't need a Ph.D. to dissect the Browns' cognitive limitations. The team's first six drafts yielded 53 players -- a full regular-season roster -- without any being either productive or popular enough to reach the Pro Bowl. Only three -- Daylon McCutcheon, Dennis Northcutt, and Aaron Shea -- remain from the team's 1999, 2000, and 2001 drafts combined.
"I do think it's safe to say executive talent in this league varies dramatically," says Massey.
The fool's-gold standard, however, is "the chart," invented by Butch Davis's old boss, Jimmy Johnson. Early in his tenure with the Dallas Cowboys, Johnson assigned each pick a relative value by analyzing past draft trades.
After sending his best player, Herschel Walker, to the Minnesota Vikings in 1989 for a package that included six high draft choices, Johnson used the picks to lever his way up and down the 1990, 1991, and 1992 drafts, building a core that won three Super Bowls.
NFL teams now use versions of his chart to evaluate deals. A chart posted recently on nfl.com, for example, values the first pick overall at 3,000 points, the tenth at 1,300 and the 32nd at 590.
The one Savage used probably said that Washington didn't offer enough.
"The chart's not wrong by a little," says Massey. "It's wrong by a lot."
Wrong, yet employed. He and Thaler analyzed 238 trades involving draft picks and found the first pick valued at as much as the 10th and 11th combined. Those values were consistent with the value assigned to picks on the chart.
Going off the chart can also be costly when you're jonesing for a particular player. Remember Kellen Winslow?
Last year, because Davis feared that the Detroit Lions, picking one spot ahead, would steal Winslow, he gave them a high second-round pick to move down and select the player the Lions wanted all along -- Texas receiver Roy Williams -- and pay less.
Based on their research, Massey and Thaler calculated the odds that one player (Winslow, for instance) starts more games or reaches more Pro Bowls than the next player chosen at his position (New England's Ben Watson, chosen 32nd). It was barely better than a coin flip.
For six months of the year, Pro Football Weekly's Nolan Nawrocki compiles two draft books. He powwows regularly with scouts and has a pair of Tivo machines oozing game footage.
It's a complex game, he says. Production can't be ignored. Mental toughness is underrated. Still, a greater determinant of where players land on NFL wish lists is the scouting reports.
Bubble butt. Light in the pegs. Small-area player. Swingman. Waist-bender. Passes the eyeball test. Plays the piano. Has some magic you can't teach. Takes a $25 cab ride to get around him. Quick, shifty waterbug.
"I don't scout hockey," says Nawrocki. "I don't scout tennis. The problem with the NFL is that it's a league made up of overachievers. When you're giving that kind of money to players, are they going to change? Are they going to keep working? You can't measure a guy's heart."
Thaler isn't buying. "Almost any quantitative measures would be better than these," he says.
"Nobody's a genius," says Nawrocki. "I don't care who you are or what your track record is -- there isn't anybody who hasn't missed on some picks."
Savage said that picking Edwards was a "no-brainer." One way or the other, he'll be right.
Subscribe now to get the latest news delivered right to your inbox.