July 17, 2016
Another Look At Doyle, Smoltz, Andersen, Bagwell
There are lots of cautionary tales about trading prospects for pitchers in the heat of a pennant race. Some belong specifically to certain fan bases or regions; most also belong specifically to one or another generation. There’s the deal that sent Derek Lowe and Jason Varitek to the Red Sox in exchange for Heathcliff Slocumb, and the one that sent Grady Sizemore, Cliff Lee, and Brandon Phillips to the Indians in exchange for Bartolo Colon. More recently, there was Scott Kazmir for Victor Zambrano, and more recently still, there was Wilson Ramos for Matt Capps. There are tons of these, and because of that possessiveness each fan base and generation feels toward its favorites, you’re probably wondering why your personal highlight wasn’t among the ones I just rattled off.
I want to talk, though, about two of the highest-profile versions of this fable in history. As it happened, they occurred just three years apart, and almost 30 years ago. They’re the ones you probably conjured quickest, when you read the first sentence, even if they’re not the ones with which you identify most closely. Here they are:
Two Hall of Fame-caliber players were traded straight-up for just one pitcher, and in August, no less! It’s no wonder we remember these deals. In one reading of the thing, it was terribly myopic on the part of each buyer to make such moves, however they might have turned out. Part of the reason we remember the moves is that they were each for pitchers over the age of 35. In hindsight, that’s as crazy as the talent either team surrendered.
Of course, there are many moves like these. Both the 1987 Tigers and the 1990 Red Sox were battling the Blue Jays for AL East supremacy, and the Blue Jays traded for pitchers to bolster their staff in each season, too—Phil Niekro in 1987, John Candelaria in 1990. The fact that everyone they gave up turned out to be fairly forgettable is notable (the art of trading prospects is all about trading the right ones), but they still tried the same thing the Tigers and Red Sox did.
On top of that, consider the players as the world knew them when the trades happened. In August 1987, John Smoltz was a struggling 21-year-old. He’d thrown 130 innings in 21 starts for Double-A Glens Falls, with a 5.68 ERA, 86 strikeouts, and 81 walks. When Bagwell was dealt, he had just finished a season with Double-A New Britain. He controlled the strike zone really well (73 walks, 57 strikeouts) and hit for average, but he’d managed only four home runs and committed 34 errors at third base. Bagwell ranked 32nd on Baseball America’s Top 100 list for 1991, and would have ranked higher if the industry was as smart then as it is now, but he wasn’t quite can’t-miss, and Smoltz was (more or less) a mess.
In light of these two facts—that these moves certainly aren’t unique, and that there was reasonable cause for the buying teams not to foresee Smoltz’s and Bagwell’s ascensions to superstardom—I think it’s only fair to acknowledge how little we know about the prospect sides of these trades. That doesn’t mean, though, that these transactions are analytically void. We might simply need to adjust the way we analyze them. Let me show you what I mean.
The Braves traded for Alexander in the middle of 1986. Collusion hit the veteran hurler hard, left him adrift in free agency the following winter. Finally, he gave in and re-signed with Atlanta for two years in May, on a modest deal. The Braves were terrible that season, though, 10 games under .500 by the end of July, fading from there, in fifth place and falling fast by mid-August. Trading Alexander was an easy call. Meanwhile, the Tigers were trying to dig out of a hole. They’d started slowly, under .500 as late as Memorial Day. They caught fire when summer came, went 33-14 at one stretch, but the Blue Jays had started strong and were staying that way. Detroit had yet to catch Toronto—had still spent zero days in first place all year—when they pulled the trigger on the Alexander deal.
And they hit a home run with it. In 11 Alexander starts, the Tigers won 11 times. He pitched 88 1/3 innings in those 11 starts, and posted a dazzling 1.53 ERA. From the day it acquired Alexander, Detroit went 34-18. A sweep of the Jays on the final weekend—starting with an Alexander win on Friday—pushed them over the top. The 1987 Tigers won 98 games. Alexander didn’t pitch well in the ALCS, and the Tigers lost it to the Twins. He did pitch another 229 innings for the Tigers the next season, though much less brilliantly. Overall, the Tigers gave up something, and it turned out to be a really big something, but they got quite a bit in return—not only because Alexander pitched so well down the stretch, but because the team was well-positioned to leverage that performance.
Larry Andersen had been an Astro for a while by 1990. He was due to be a free agent at season’s end, though, and the Astros were 20 games out of first place by early July. At 37, Andersen was an easy reliever from whom to move on, particularly given his expiring deal. The Red Sox, meanwhile, badly needed just such a reliever. They had a dominant starting rotation, led by Roger Clemens but without a single weak link. Their relief pitching, though, was worse than average in every month of the season, and particularly bad in the summer months. A great offense still put Boston in control of the AL East: they led by six games near the end of August, when they grabbed Andersen at the cost of Bagwell.
And they hit a home run, as far as Andersen went. He pitched 22 innings in 19 appearances, facing 86 batters, striking out 25. He walked three, didn’t allow a home run, and posted a 1.23 ERA. Yet, the Red Sox stumbled. Over their final 30 games, they went 13-17. The Blue Jays briefly caught them, though the Sox finished two games up, winning 88 games. The A’s unceremoniously swept Boston out of the postseason, and in 1991, Andersen was a Padre.
There are two reasons history hates these two trades. One is that the teams who traded for pitching help didn’t win the World Series, or even the pennant, because of it. The other is that the players they traded away became transcendent stars. In any clear-headed analysis of the deals, though, both of those facts are irrelevant. In fact, history is right about just one of them, and very wrong about the other.
The Tigers were trading for a starting pitcher they knew had a good chance to improve their rotation in 1988, as well as boosting them for the remainder of 1987. They faced a deficit, but a surmountable one, in their division. The addition of Alexander had, at its upper bound, the potential to make the difference and put them in the playoffs (and indeed, eventually did so).
The Red Sox were trading for a relief pitcher they knew would be a free agent at season’s end. They had a healthy lead in their division, and although they faltered so badly that they nearly lost it, they had no good reason to expect that. Andersen’s arrival, which panned out as well as it possibly could have, didn’t have the same impact potential as Alexander’s had three years earlier. He barely pitched a quarter of the innings. He only had one appearance in which his Win Probability Added was over 0.10. (Alexander had seven starts in which his WPA was at least 0.18, and six of those were over 0.25.)
This is a perfect pair of trades to tie together for this kind of analysis, because of what they have in common. Both worked out almost as well as was conceivable, in terms of what each team got for their contending teams. The differences lie in:
1. The degree of difference made by each player’s 95th-percentile kind of showing; and
2. The situation each team was in.
Trading very talented players for relief pitchers is folly, especially if there isn’t long-term control of that relief pitcher involved for the buyer. Even if they pitch at an exceptional level, there’s just not a good enough chance that those pitchers will change the way the season unfolds for their new team. Trading for starters is different; it can be really important to shore up the club’s chances of winning every fifth day.
It’s also a mistake to make a trade aimed only at improving a team already bound for the playoffs. If a team is already very likely to reach the postseason, any moves they make should be aimed at making them more likely to be back in that position the following season. October is a crapshoot, and nothing anyone does can make it anything else. If anything, a cushion of several games in late summer is a good opportunity to look ahead a bit.
This is all relevant information right now, of course. The Red Sox’s trade for Drew Pomeranz (with two more years of control after this season and the team sitting at a 53-percent chance of reaching the ALDS, according to our Playoff Odds Report) fits the profile of the Alexander trade. Rumors that teams like the Nationals, Cubs, and Giants (all with Playoff Odds over 95 percent right now) will pursue rental relievers like Aroldis Chapman, on the other hand, reek of the Andersen debacle.