Smoltz for Alexanderby Dave Studeman
February 21, 2005
It was 1987, and things were happening in the American League East. The Brewers tied the MLB record by starting the year with 13 consecutive victories, eerily foreshadowing the Orioles' 21-game losing streak at the beginning of 1988. The Mattingly/Henderson/Winfield/Randolph Yankees surged into first after the Brewers found their intrinsic net worth, but injuries took their toll on the Yanks, and they eventually finished nine games off the pace.
By early August, the Blue Jays, featuring that awesome outfield of Bell (the AL MVP), Moseby and Barfield, had taken over the lead, followed closely by the Detroit Tigers. The Tigers had started the year with an 11-19 record but, led by the phenomenal Alan Trammell (who probably should have won the AL MVP), they were right behind the Jays in early August. The rest of the year turned into an awesome, dramatic pennant race between the two teams.
The Tigers made a couple of trades in early August to bolster the team for the pennant race:
- Darnell Coles (who went on to compile 40 Win Shares in his career, but none above a bench-player level) and minor leaguer Morris Madden to the Pirates for Jim Morrison (a grand total of two WS remaining in his career).
- Minor Leaguer John Smoltz to the Braves for Doyle Alexander (Win Shares to come).
On September 23, the Tigers were just a half game behind the Jays for first place. The two teams faced each other in a four-game series in Toronto, and the Blue Jays won the first three games, all one-run nail biters. The second game, in which the Jays scored three runs in the bottom of the ninth to win 3-2, was particularly dramatic. The Tigers won the final game, 3-2 in extra innings, as Alexander went 10.1 innings, allowing just two runs, but the Jays seemed safe with a 2.5-game lead and six games to play (seven for the Tigers).
However, the Jays were swept by the Brewers in the next three games, and the Tigers split a four-game series with Baltimore, setting up a final three-game showdown between the two teams in Detroit. Has there ever been a better season-ending series?
Alexander opened it with 4-3 win for Detroit and the two teams were tied for the Eastern lead. The Tigers won the next game 3-2 in 12 innings thanks to Trammell's game-winning single, and the final game was a fantastic pitcher's duel between Frank Tanana and the great Jimmy Key. The Tigers only mustered three hits against Key, but won the game 1-0 thanks to a home run by Larry Herndon and a dominant complete game by Tanana.
Two fantastic head-on series by the pennant contenders at the end of the year, all games determined by one run. It's no wonder that the team page for the Tigers over at Baseball Reference is sponsored by someone who says, simply:
Favorite. Team. Ever. What a fantastic end to a fantastic season.
And it's no wonder that Toronto fans have probably stopped reading by now.
Remarkably however, when you add up all the Win Shares involved in the Smoltz/Alexander trade, it ranks as one of the worst in baseball history from the Tigers' point of view. And I'm not the only one who has said so; the Baseball Page lists the Smoltz/Alexander trade as one of the worst in the history of the Detroit Tigers. Another article rates it the fourth-worst deadline deal of all time, further citing Alexander's poor record in the 1987 postseason (they didn't make it to the World Series).
On the other hand, I've received e-mails from Tiger fans saying that they would make the same trade again. And this thread highlights the mixed feelings many Tiger fans have about the trade, calling it the best short-term and worst long-term trade in Tiger history.
Which makes me wonder: can we reconcile these two points of view? Can we take the crude Win Shares measure I introduced last week and refine it through the lens of this trade?
Of course we can! Let's try these steps on for size; each step is added to the previous one, and the noted impacts are cumulative:
1. Let's first total up the Win Shares of each player after being traded. That's easy: Alexander created 29 Win Shares from the trade to the end of his career two years later. Smoltz has compiled 234 since the trade, and he's still going. That's a 205 Win Share lead for the Braves.
2. Let's compute how much those Win Shares contributed beyond a bench player (WSAB). This is a very crude approach, but I calculate 148 for Smoltz and 14 for Alexander, based on a "bench" starting pitcher contributing six Win Shares and a reliever contributing five Win Shares a year. So that's a 132 Win Share lead for the Braves.
3. Let's look at the "present value" of those Win Shares. Today's Win Shares are worth more than tomorrow's Win Shares, for a number of reasons. I'm going to take a simple approach here, and simply say that today's Win Shares are more valuable than tomorrow's because today's bottom line is worth more than tomorrow's bottom line. Some people call that the "time value of money" but you probably call it interest. Using a 5% interest rate for Win Shares, I get a 1987 value of 96 Win Shares for Smoltz, and 14 for Alexander. That's an 82 Win Share lead for the Braves.
Taking a "present value" approach to a trade is a very important step for trades of this sort. Smoltz didn't begin to contribute on a major league level until 1989, two years after the deal was consummated.
4. Let's include only those years in which a team could have been certain to have both players on the roster. One of the complaints I received about the Major League Trading Balance Sheet was that we included all of a player's future performance in the calculation, when everyone knows that a player is a free agent after his first six years and can then move onto any team he wants. Although Smoltz was from Michigan, he might have left the Tigers once he reached free agent status. Why include those later Win Shares in the calculation?
Alexander had signed a two-year deal with the Braves before the 1987 season, so let's include only those two years in our calculation, as well as only the first six full years of Smoltz's career. Now, Smoltz has a 38-12 Win Share lead over Alexander, and the difference between the two has declined to 26 Win Shares.
5. Finally, let's consider each team's competitive position at the time of the trade Alexander was 36 years old in 1987, and the Braves were 69-92, fifth in the NL West. Why not play the youngsters and save some money by trading the Alexander contract? Conversely, the Tigers were at their peak -- Trammell was 29 and Whitaker was 30. It was time to reach for the brass ring. Any complete assessment of a trade like this has to take these relative positions into account.
To quantify this impact, I'm again going to revert to my financial roots. See, a team that is in contention (over 82 wins) makes about six times as much money for each additional win as one out of contention. I've referenced this article before, and it's a critical idea for linking performance with actual worth to a team. It allows you to say that Alexander's nine WSAB in 1987 were worth SIX times as much to the Tigers as to the Braves.
Now, that sort of seems like a lot to me, so I'm going to use the following table to give a weight to each player's yearly Win Share totals, based on his team's winning percentage that year.
W% Min W% Max Weight .000 .400 1 .400 .500 2 .500 .525 4 .525 .550 5 .550 .575 6 .575 .600 7For example, a team with a winning percentage of .510 will attribute a weight of four points to a player's Win Shares. One with a .450 record will attribute two. The overall difference in this table between over/under .500 teams is less than six -- it's probably more like three. But it's seven for a great team vs. a bad team. My hope is that you can accept this more conservative approach.
Alexander not only made a fantastic contribution to the Tigers in 1987, he was also a key starting pitcher for them in 1988 when they went 88-74 and finished second to the Red Sox in the East. They then fell to a .360 record in 1989, and flirted with contention a couple of times between then and 1993 (the end of Smoltz's six years). So they have the highest weight given to years in which Alexander was great and Smoltz wasn't yet contributing in the majors.
On the other hand, during Smoltz's first two full years in the majors, his teams had a winning percentage around .400, but they then exploded to the top in 1991, finishing first in the NL West, and remained in contention for the next four years. So Smoltz contributed heavily when the Braves needed him, and when Alexander was no longer pitching.
The net effect of this is that dealing Smoltz was a great deal for the Braves, as he accumulated 168 weighted Win Shares for the Braves, while Alexander would have accumulated 12. That's a difference of 156 weighted Win Shares in favor of the Braves.
And for the Tigers, it was also a good deal. Alexander accumulated 74 weighted Win Shares for them, while Smoltz would have accumulated 66. That's a difference of eight weighted Win Shares in the Tigers' favor.
That's right. If you are willing to jump through all those hoops with me, this deal was a Win/Win outcome! Despite the lopsided nature of the initial Win Share totals, the Tigers actually made themselves a good deal. The Braves' side of the deal was much better, but that doesn't detract from the Tigers' side of the math. If a deal has a positive variance in your favor, you made a good deal for yourself even if the other team has a bigger variance.
In the end, this mathematical exercise is just that: exercise. But I wanted to see if you could make a reasoned, quantified argument that the Tigers actually came out ahead of where they would have been by trading Smoltz for Alexander. And it turns out you can.
References and Resources
All of this trade analysis stuff is a joint venture between myself and Mike Carminati, who is also posting a series of articles on the same subject. I can't thank Mike enough for his hard work and guidance. Also, thanks to the many people who have sent or posted suggestions about trade evaluations.
All the information about the end of the 1987 season is courtesy of Retrosheet.
Here's a list of the AL Win Shares Leaders in 1987:
Player WS ------- --- Trammell 35 Boggs 32 McGwire 30 Puckett 29 Molitor 29 Clemens 28 Mattingly 27 Yount 26 Bell 26 Evans 25
Dave was called a "national treasure" by Rob Neyer. Seriously. Comments about this article can be sent to him through the miracle of e-mail.