Microtubules, built from tubulin subunits, are biological polymers intimately involved in numerous key cellular processes. Though microtubules have been studied for decades and implicated in many diseases, there are still many unanswered questions about their function. One key behavior of microtubules that is still poorly understood is dynamic instability: microtubules cycle through periods of length growth and shrinkage seemingly at random. Microtubules require the energy of GTP hydrolysis to function, making them steady state, not equilibrium, enzymatic polymers. Enzymes are typically studied at steady state, but the fact that tubulin serves as both an enzyme and a substrate make using standard enzyme approximations for tubulin difficult. The dynamic instability of microtubules is commonly studied in isolation experimentally or via simulations, but due to practical constraints, most of what we know about their dynamics comes from wild type protein isoforms. These reasons and more have made it difficult to connect the behavior of tubulin subunits to those of microtubule polymers and generate generalizable principles that govern the behavior of dynamically unstable polymers.In this dissertation, I use three types of simulations of dynamic instability that mimic the behavior of microtubules to answer questions about dynamically instable polymers to answer a series of questions about microtubule behavior. Using simulations allows for the exploration of parameter spaces that are difficult to achieve in an experiment. Simulations are effective because the allow the observer to look at multiple perspectives of a system simultaneously. For example, the lack of clarity in the definition of microtubule characteristics like "steady state" leads to contradictory experimental results, where these different definitions can lead to confusion. Experimentalists commonly refer to "the" singular steady state, but can actually be referring to different steady state behaviors. To clarify the definition of microtubule steady state, I have identified and named four different steady states a system of microtubules can achieve and demonstrated both how microtubules behave in these steady states and how to set up an experiment to ensure that it reaches each of these steady state. A second problem with how microtubules are studied is that there are unavoidable thresholds that must be applied to quantify their behavior. Some of the effects of these thresholds have been studied before, but to establish the effects of these thresholds more comprehensively, I have used a model of dynamic instability where all DI events are known before analysis has begun to test the efficacy of different ways of analyzing microtubule behavior. This model has shown that the application of any measurement-based thresholds can have a significant impact on measured catastrophe and rescue frequencies. Next, I have developed a method to correct for the thresholds applied to experimental data and extrapolate a threshold independent catastrophe frequency that is more accurate than what would be measured using standard methods. Lastly, experiments and simulations have shown that dynamic instability transitions are not random as was initially hypothesized, and that the rate of transition to shrinkage increases the longer the microtubule exists (that is, the microtubule "ages"). In this thesis I have reported that the aging behavior seen in microtubule dynamic instability experiments can be explained at least in part by measurement thresholds that can be removed in simulations but cannot be avoided experimentally. As part of this work, I have demonstrated how methods to measure DI are influenced by the thresholds mentioned in our earlier analysis. Taken together, these results provide new methodologies to study microtubule dynamic instability and ways to remove experimental and user constraints on the analysis and reporting of DI results.