First, think about current political conditions. How likely is nuclear war now? Your answer for this will go on the left-hand side of the graph, near 2010.

Then, think about what conditions may be like in 2070. Maybe nuclear war will become more likely as weapons technologies proliferate. Maybe nuclear war will become less likely as global political institutions improve. Your answer for 2070 will go on the right-hand side of the graph, near 2070.

• Claim: Stanford Professor Martin Hellman, who has researched the topic for decades, estimates a 1% annual probability of nuclear war between major nuclear powers. Reviewing the literature, he found no other references on the annual probability of nuclear war.
  Implication: Although 1% is probably more likely than 0.001% or 50%, little serious research has been done and probability distributions should be wide.
  Source: Hellman, Martin E. "Risk Analysis of Nuclear Deterrence." The Bent Spring 2008: 14-22. Defusing the Nuclear Threat. 9 Aug. 2008 <http://www.nuclearrisk.org/paper.pdf>.

• Claim: Many experts estimate the annual risk of a nuclear terrorist attack at around 1%-5%.
  Implication: Rogue nuclear bombings, which may lead to nuclear retaliation and subsequent escalation into full-scale nuclear war, are not impossible and may be quite likely in the next 10-20 years.
  Source: Lugar, Richard G., United States Senator for Indiana. The Lugar Survey on Proliferation Threats and Responses. Washington, D.C.: Office of Senator Lugar, 2005.
  9 Aug. 2008 <http://lugar.senate.gov/reports/npsurvey.pdf>. Pg. 22.

• Claim: The US has expressed willingness to use nuclear weapons under very broad circumstances, even when no other WMDs are involved.
  Implication: Generals and other officials are not that reluctant to use nuclear weapons, making a nuclear escalation probable.
  Source: U.S. Department of Defense. Doctrine for Joint Nuclear Operations. Joint Publication 3-12. Final Coordination (2). Mar. 15, 2005. 9 Aug. 2008 <http://www.globalsecurity.org/wmd/library/policy/dod/jp3_12fc2.pdf>.

• Claim: During the Cold War, a US research paper used estimates ranging around 1%-5% annually for nuclear war risk models.
  Implication: High political tensions can lead to significant annual probabilities of nuclear war.
  Source: Avenhaus, Rudolf, Steven J. Brams, John Fichtner, and Marc D. Kilgour. "The Probability of Nuclear War". Rep.No. 86-24. C.V. Starr Center for Applied Economics, New York University. 5 Aug. 2008. IDEAS. RePEc. 9
  Aug. 2008 <http://ideas.repec.org/p/cvs/starer/86-24.html>.

• Claim: Even a relatively small nuclear war could disrupt global climate for a decade or more.
  Implication: Indirect damage from nuclear war could greatly exceed direct destruction.
  Source: "Regional Nuclear War Could Devastate Global Climate." ScienceDaily. 11 Dec. 2006. Retrieved 9 Aug. 2008. <http://www.sciencedaily.com/releases/2006/12/061211090729.htm>.

• Claim: Between 1956 and 1979, there have been at least 20 mishaps that could have led to accidental nuclear war between the US and USSR. Even if the chance of each one leading to a war was small, when so many pile up, the overall chance of disaster may be substantial.
  Implication: Even if there are no outright hostilities, a full-scale nuclear war may occur simply by accident. Once a war gets started, it may be difficult to stop.
  Source: Phillips, Alan F. "20 Mishaps that Might Have Started Accidental Nuclear War." Nuclear Age Peace Foundation. Jan. 1998. Retrieved 9 Aug. 2008. <http://www.wagingpeace.org/articles/1998/01/00_phillips_20-mishaps.htm>.