The objective of this study was to develop equations for estimating farm-specific cost of days open. The equations were constructed using a whole farm stochastic simulation model previously described by Bewley et al. (2010) and Liang (2013). Ten thousand iterations were run for lactations 1 to 5 with the mean cost per day open as an output. Stochastic variables expected to have potential effects on cost per day open were collected from each iteration. Those variables included: rolling herd average milk production, age at first calving, mature cow live weight, heifer calf value, bull calf value, semen cost, days in milk dictating an open cow as a reproductive cull, milk production level dictating an open cow as a production cull, veterinarian costs, discount rate, milk price, feed price, replacement price, cull cow price, voluntary waiting period, estrus detection rate, and conception rate. The GLMSELECT procedure of SAS 9.3 (SAS Institute, Inc., Cary, NC) was used to analyze the effect of stochastic variables and 2-way interactions on the mean cost per day open for each lactation. Variables remained in the model when significant at P < 0.05. The R² of the resulting models were 0.57, 0.54, 0.64, 0.85, and 0.63 for lactations 1 to 5, respectively. The models were used to develop deterministic, lactation-specific equations for cost of days open. These equations are available in an online spreadsheet at: http://afsdairy.ca.uky.edu/CostOfDaysOpen. To demonstrate use, mean US Holstein herd data from 2015 DairyMetrics (Dairy Records Management Systems, Raleigh, NC), 2014 Food and Agricultural Research Policy Institute (Columbia, MO), and published literature were entered into each equation. Mean cost per day open for lactations 1 to 5 was $2.44, $2.82, $4.42, $4.54, and $3.32, respectively. These new, robust regression equations for cost of days open account for the complexities of varying market and herd conditions. The equations can estimate cost of days open in partial budgets without the costs or computing time required for stochastic simulations.