Simulating debate: Bonds and Pittsburgh

Had Bonds played his entire career with the Pirates, he would be a bit closer. Bonds, in fact, would have entered the 2007 season with 745 career homers, just 10 shy of Aaron's benchmark.

Confused? Let us explain.

The Pittsburgh Tribune-Review asked Diamond Mind Baseball, a computer software division of Simnasium, Inc., to project Bonds' career accomplishments had he never left the Pirates in 1993. Via computer simulation, stats guru Luke Kraemer of Diamond Mind calculated that Bonds would have smacked 11 more home runs from 1993-2006 had he never signed with the Giants. However, Bonds would not have broken Mark McGwire's single-season record in 2001, hitting 68 homers with the Pirates, as opposed to 73 with the Giants.

The projection did not include the early portion of this season, during which the real Bonds -- as opposed to the computerized Barry -- added one more homer to his glowing resume.

View this PDF for a look at the statistical breakdown.

Had Bonds made Pittsburgh his permanent baseball home, not only would he have hit more home runs, he'd have helped the Pirates win 86 more games over those 14 seasons. And forget about that string of 14 consecutive losing seasons. Bonds would have helped the Pirates to four winning seasons and a pair of .500 finishes.

When Bonds left, he had hit 176 home runs and won two NL MVP awards in seven seasons. Over the next 14 seasons in San Francisco, he hit 558 more homers and collected four more MVP's. How different would his numbers look had Bonds never left for greener pastures? He began 2007 with 734 home runs. A difference of just two home runs more per season in Pittsburgh over the 14 seasons he spent in San Francisco (disregarding his abortive 2005 season) would have meant breaking the record in 2006. A difference of two less per season may have put the record out of his reach entering 2007.

Why might Bonds have done better or worse playing for the Pirates than the Giants? When imagining how players would have performed in different times and places, context is everything. Even the casual fan may know, for example, that you have to take into account the boost that Colorado Rockies players get from playing half their games in Coors Field when you compare their hitting stats to players on other teams.

Baseball statisticians have measurements for how the characteristics of ballparks impact hitters called "park factors." They can measure the overall impact of a park on offense compared to the league average, the specific impact on the relative frequency of singles, doubles, triples and home runs and any difference in impact for each of these measures between left-handed and right-handed batters.

From 1993-2006, the average park factor for Three Rivers Stadium and PNC Park for home runs by left-handed batters was 104 (100 represents the league average). Over the same period, for Candlestick/3Com Park and PacBell/SBC/AT&T Park, it was just 83. Based on those figures alone, all other things being equal, Bonds would have been better off in Pittsburgh. So much so that you might assume that Aaron's all-time record would already be his.

View this PDF for a look at the statistical breakdown.

Of course, all other things rarely are equal. Using the Diamond Mind Baseball simulation software, we replaced the Pirates' regular left fielder with a Barry Bonds "clone" (we kept Bonds on the Giants as well) and replayed the 1993-2006 seasons 20 times each. The Computer Manager for the Pirates was instructed to play Bonds with the same frequency each season that he actually played in those years for the Giants. Then, we averaged the results for Bonds and the Pirates for the 20 seasons we simulated for each year.

Bonds did average more home runs for the Pirates in our simulated seasons than he actually hit for the Giants, but only 11 more overall (putting him closer to, but not beyond, Aaron's record heading into 2007). The Pirates also averaged nearly seven wins better per season with Bonds than they did without him, not including the 2005 season.

View this PDF for a look at the statistical breakdown.

Why might Bonds not have done even better had he played all of those seasons for the Pirates rather than the Giants? Here are a few possibilities:

• When playing for the Pirates in our simulated seasons, he was facing the Giants, instead of the other way around.

• From 2001, when the NL changed to an imbalanced schedule, the simulated Bonds, playing for the Pirates in the NL Central Division, lost over a dozen games total in Coors Field and Bank One Ballpark compared to the real Bonds playing in the West Division. (While SBC Park in San Francisco is perhaps the toughest in the National League for a left-handed home run hitter, and PNC Park is about average, Coors and the BOB are two of the friendliest.)

• The Giants and Pirates played different schedules each season A detailed game-by-game analysis might well find that the simulated Bonds, playing for the Pirates, faced tougher pitching for a left-handed home run hitter overall than the real Bonds faced for the Giants.

The year the imbalanced schedule was introduced also was the year Bonds hit a mind-boggling 73 home runs. In our simulated seasons, however, he averaged just 68 for the Pirates. One other explanation why Bonds didn't do relatively better for the Pirates in our simulated seasons overall is suggested by this extraordinary performance in 2001.

Baseball analysts have developed increasingly sophisticated methods for projecting the future performance of players. This development has been driven by the needs of baseball front offices, faced with increasingly expensive multiyear personnel decisions, and the voracious appetite of fantasy baseball addicts for an information edge. However, in any given season, real or simulated, an extraordinary athlete may surpass even the most optimistic expectations and smash all previous records. Our method of replaying and averaging the results of many seasons smoothes out such once-in-a-lifetime performances.

It may simply be that Bonds has been such an extraordinary hitter that, despite the relatively unfavorable conditions he faced in San Francisco, he already was performing near the limits of what was possible. He simply may not have gotten the same advantage hitting in Pittsburgh that other merely mortal left-handed power hitters would have.

• Most Read
• Most E-mailed
• Latest News