We present a machine learning framework aimed at increasing performance seizure detection systems. Despite the diversity of ML-based available, the methodology to select an optimal classifier or ensemble model for seizure research is not commonly known. Our study is aimed at bridging this gap by showing use of a statistically guided machine learning framework that has delivered reliable on-field performance. We empirically examined performances of 11 different classification algorithms, with additional 13 variants, executing 300 machine learning models against 6 fixed preictal windows (ranging from 5 minutes to 60 minutes) and 3 different sets of features as part of our study. Using a base of 55 subjects, of which we used 43 subjects to train and validate multiple ensemble models and classifiers, we achieved upto 98.3% F1 score, with 94.8% chance level performance on our data, and a False Positive Rate (FPR) of 0.0156/hr. We applied the top ensembles and classifier variants on 12 completely new subjects, unseen by the models to test their on-field efficacy. One of our models detected all 12 seizures on the unseen data and predicted 11 of them before electrographic onset. Average prediction time of models across all subjects was 31.56 minutes. At an individual level the earliest prediction was recorded at 176 minutes, which is the highest known seizure prediction time recorded in literature. Our analysis showed that a 30 minute window could be the most ideal preictal window for training seizure models. The significance of our work is that we are reporting the highest F1 score of 98.3% on seizure data so far recorded with sample size of higher than 10. We are able to show high performance of seizure models using wearables data, detecting seizures at par with neurologist verified EEG onset time, and predicting seizures up to 176 minutes before onset, which is significantly higher than previously reported scores. Our work lays important foundations for using wearables for actual monitoring of PWEs in out-of-hospital settings, which might be an important step in ensuring their safety and quality of life.