Simple portfolios provide useful benchmarks to gauge the added value of complex models that generate the risk of overfitting. The simplest strategy—an equally-weighted portfolio—has been shown to be one of the best performers.
Famously, de Miguel, Garlappi, and Uppal (2009) compared the out-of-sample performance of portfolios produced by various mean-variance optimizers, including robust Bayesian estimators, portfolio constraints, and optimal combinations of portfolios, to the simple 1/N rule. They found that the 1/N portfolio produced a higher Sharpe ratio than each asset class position, explained by the high cost of estimation errors that often outweighs the benefits of sophisticated optimization out-of-sample.
The 1/n portfolio is also included in the efficient frontier figure above.