Construction industry has found that cost overruns and schedule delays are recurrent problems within the sector. Considerable cost and schedule deviations are issues at the project level that permeate organizations and seriously affect their financial performance. These adverse deviations are evidences that the traditional project control systems fail to predict, promptly and effectively, cost and schedule deviations at completion of capital projects. Even more, it seems that the current assessment methodologies do not inform how well the control system has ensured the expected cost and schedule at completion throughout the life cycle of the project. The present study inquired into current practices of forecasting and project management to see if the predictability of cost and schedule deviations of construction projects is timely and accurate. Additionally, this research investigated and assed the most relevant change reasons associated to the forecast deviations, as well as, the most critical underlying factors that jeopardize the accuracy and timelines of these predictions. To attain these proposed objectives, a correlation research was mainly designed and complemented with a cross-sectional design. An electronic questionnaire was the primary data instrument adopted to collect the data, which finally, were analyzed by bivariate and ordinal correlations statistical techniques. The results demonstrated that cost timeliness and predictability -understood as the addition of cost predictability index plus schedule predictability index, were significant predictors of the likelihood of achieving small cost and schedule deviations at project completion, less than 5%. Although there was no evidence enough to demonstrate the statistical significance of schedule timeliness, the study considered and showed its benefit as component of the predictability index. The study identified the most significant change reasons related to forecast increments (cost and schedule) and classified their relevance by owners and contractors. The results demonstrated that change reasons with low frequencies implied high impacts and those with high frequencies implied medium and low impacts. The study showed there are underlying factors that noticeably influence the predictability performance, which must be tracked and controlled to avoid adverse deviations.