Friedman was a Chicago monetarist, not just an Austrian hack. Pure monetarism doesn't always work, as it ignores irrational behavior, but it's not _useless_. Monetary models work great in steady-state conditions for fine-tuning and in cases where the monetary supply results in high inflation.
The 2008 economic crisis demonstrated that brilliantly. The IS-LM model correctly predicted the outcome of fiscal expansion (lack of inflation, despite trillions in stimulus) and the futility of monetary measures (negative rates in Europe did not result in economic growth).