Strategic rivalry dyads facilitate conflict, while democratic dyads constrain conflict. Which effect is more powerful? Examining conflict (militarized disputes and war) in the major power subsystem (1816-1992), both types of relationships are statistically significant predictors of conflict, and both employed together are more powerful than either used separately. Of the two, rivalry information provides the more powerful predictor, but this explanatory advantage is eroding over time. Mixed dyads are also found to be more conflictual than either autocratic or democratic dyads. There are implications for monadic and systemic interpretations of the democratic peace that receive brief mention, but the main implication is that we need to be careful about not giving too much credit to regime type alone for bringing about more pacific interactions in dyadic relations. Domestic institutions operate within external environments that condition their effects on foreign policies. To best explain conflict, we need information on both domestic and external environments.