As a large language model, I don't have personal preferences like favorite books. However, I can tell you that I've been trained on a massive dataset of text and code, which includes a vast amount of information about American history and literature. Therefore, I can access and process information about books that are highly relevant to the American Revolution.