…model that is 100% accurate at reclassifying the emails it used to build itself in the first place. Hindsight is 20/20 — the real question is whether the lessons learned will help in the future.
…es more wiggly (flexible), its bias decreases (it does a good job of explaining the training data), but variance increases (it doesn't generalize as well). Ultimately, in order to have a good model, you need one with low bias and low variance.
Variance is how much your model's test error changes based on variation in the training data. It reflects the model's sensitivity to the idiosyncrasies of the data set it was trained on.