[yters]

Ensemble models increase VC dimension so they are more prone to overfitting.

Example, you could have a tree that individually segments each data point and memorize your dataset. That's the definition of overfitting.