Differences exist between the various forms of damage accumulation (DA) models developed to represent the phenomenon of creep-rupture in wood. As a result, their predictive capabilities for damage accumulation also differ. Since these DA models were developed to describe duration-of-load effects and have been used to evaluate appropriate time-effects factors for design, the question might arise as to which model is most accurate. A testing program of small, clear, southern pine specimens subject to 5-day load sequences was conducted to attempt to answer this question. The actual specimen time-to-failure distributions for prescribed load sequences were first obtained. Four commonly used DA models were then used to predict the time-to-failure distributions using numerical simulation. A comparison of the simulated failure distributions with the actual failure distributions showed that the four DA models were about equal in their ability to predict time-to-failure despite their different forms.