Last week, with the Giants struggling, Brian Sabean was asked about potential trades. The GM acknowledged that he has the payroll room to add players, but cautioned that he wouldn’t expect anything soon, since this time of year isn’t conducive to trading:
The best way to put it is, the price of business right now is prohibitive. Thus, we’re not going to do business, to spend all our money now or clean out our farm system.
You’re not going to do anything without getting your clock cleaned as of this date. If somebody’s going to trade somebody this early, they’re going to have to win the deal and you’re going to have to let them win the deal.
A week later, the Cubs sent Scott Feldman to the Orioles for Jake Arrieta, Pedro Strop, and two international bonus slots. (They also traded Carlos Marmol to the Dodgers.) According to Sabean, the Orioles should have had to let the Cubs clean their clocks to get that deal done, but the exchange didn’t seem obviously lopsided. While the trade was skewed toward Chicago in terms of team control years, the Cubs’ prizes were a pair of flawed pitchers who had allowed a combined 38 runs in 47 innings for the Orioles this season.
So with the help of BP’s Andrew Koo and Retrosheet's transaction logs, I went looking for some sign that buyers have historically paid a much higher price when they’ve added talent well before the deadline. We looked at trades from 1986—the first year that the deadline was July 31—to 2006, which was long enough ago that it’s safe to assess both sides. Sorting the trades into two buckets, "early" (last week of June-first week of July) and "late" (last week of July), we added up the WARP produced for the buyers and sellers by the players they bought and sold.
Sabean’s story is consistent. In nearly two decades as the Giants’ GM, Sabean has made only a couple of minor moves from the last week of June through the first week of July. He acquired Alan Embree in late June of 2001 and traded Bobby Estalella to the Yankees for Brian Boehringer early in July of the same season. He also moved Bengie Molina to Texas for Chris Ray and Michael Main on July 1, 2010.
However, he has a long, successful track record of deadline trades: Darryl Hamilton, Jim Stoops, and a minor leaguer for Ellis Burks in 1998; Nate Bump and Jason Grilli (over a decade before he was good) for Livan Hernandez in 1999; Armando Rios and Ryan Vogelsong (also a decade before he was good) for Jason Schmidt and John Vander Wal in 2001; Jesse Foppert and Yorvit Torrealba for Randy Winn in 2005. (He was also involved in the so-called “White Flag Trade” of 1997, though that one ultimately worked out okay for the White Sox.) And that’s just through 2006; he’s done more good work at the deadline during the last several seasons (which have worked out okay for San Francisco).
Our method was somewhat crude—it’s difficult to tell programmatically whether a traded player became a free agent and re-signed, especially in years for which we don’t have contract info. (And if a traded player did re-sign as a free agent, we don’t know whether the trade made it easier for the acquiring team to bring him back.) It’s also hard to account for the fact that a team might trade one of the players it acquired and extract further value from him that way. But on the whole, there was no obvious trend toward early buyers being fleeced. They may have given up a little more, but they also got a little more back, since their rentals lasted a little longer.
Of course, that’s not inconsistent with what Sabean said. The high asking price early in the trading season—as well as the greater uncertainty about which teams might be buying and selling—might explain why last-minute swaps are so much more common. In the 1986-2006 period, there were over three times more trades made in the last week of July than there were in the last week of June and first week of July combined. The trades that do go down early are likely the ones with more reasonable asking prices, which would explain why the data on trade winners and losers doesn’t show any discernible direction.
To wrap this up, here’s the best “early” (end of June/beginning of July) trade made by a buyer from 1986-2006:
July 5, 1987: Giants trade Chris Brown, Keith Comstock, Mark Davis and Mark Grant to the Padres for Kevin Mitchell, Dave Dravecky, and Craig Lefferts
At the time of the trade, the Giants were 5.5 games back in the West, the Padres were 17.5 games back, and a 25-year-old Mitchell was hitting .245/.313/.398. After joining the Giants, Mitchell hit .306/.376/.530. The Giants won the division, and the Padres finished last. Over his 4 ½ seasons with the team, Mitchell produced 21.5 WARP, making two All-Star teams and winning an MVP award. Dravecky and Lefferts contributed another four wins from 1987-1989, which is roughly what the Padres got out of the four players they received.
And here are a few of the worst:
June 27, 2002: Indians trade Bartolo Colon and Tim Drew to the Expos for Cliff Lee, Brandon Phillips, Grady Sizemore, and Lee Stevens
The Expos were farther from first place than the Indians—and trying not to be contracted—when they made this trade. They finished 19 games back of the 101-win Braves in the East and behind three other teams in the NL Wild Card race. Colon pitched pretty well, but he wasn’t worth one of the best prospect hauls in history.
June 30, 2006: Indians trade Eduardo Perez to the Mariners for Asdrubal Cabrera
Perez hit .195/.304/.241 for Seattle in 102 plate appearances down the stretch, then called it a career. Cabrera debuted for Cleveland the next season and has been worth 14.2 WARP (and counting).
July 6, 1986: Braves trade Duane Ward to the Blue Jays for Doyle Alexander
The Blue Jays, who were 10.5 games back, traded Alexander to the Braves, who were 3.5 games back. Alexander gave Atlanta 1.6 WARP that season, while Ward produced 12.8 for Toronto over the rest of his career, which was all spent with the Blue Jays. Doyle then re-signed with the Braves, who traded him to the Tigers for John Smoltz in August of 1987. The moral of the story: whatever you do, don’t trade for Doyle Alexander at the deadline.
Thanks to Andrew Koo for research assistance.