In this study, we present a framework to investigate the impact of dehazing algorithms on the performance of visual object tracking task using a synthetically generated hazy video dataset. The dataset is generated using a 3D development engine where photo-scanned environments are used for creating photo-realistic scenes. The dataset contains clear and hazy pairs of image sequences with moving objects. The positions of the moving objects for each frame are automatically exported as metadata, simplifying the typically laborious process of manual annotation. We applied various dehazing algorithms on this dataset and quantitatively evaluated their performance based on multiple metrics using clear reference scenes. Then, we tested several state-of-the-art visual object tracking algorithms on both hazy and dehazed images to determine the effect of dehazing on tracking performance. Our findings reveal that while dehazing algorithms improve visual clarity, they do not always enhance the performance of visual object tracking methods. This suggests that dehazing performance does not necessarily correlate with the functional gains in object tracking task.