It is not good trying to engineer your way into a one shot success by creatively applying OTS equipment to new expanded missions or trying to do too much on a dime without allowing budgeting for failures. The OTS equipment is limited in supply, so failure testing is too costly. The 737max was engineered to save money as was the Titanic explorer, and probably the latest example, the Peregrine lunar mission. My guess is the Peregrine mission was funded by all the add-on missions and they probably also had to work on a dime. Musk has it right. Keep testing and failing cheaply, till you end up with a robust resilient system. Start simple and cheaply and fail, fail, fail until you fully understand the engineering and work out all the kinks. Sending Astronauts around the Moon on a second shot, with such a complicated system, is like throwing a hail mary in a football game hoping nothing goes awry. I have never been comfortable with that approach, and I agree with delaying for more robust testing, if that is their approach. Better yet, would be to have SpaceX send a few more dummy capsules around the Moon to test the heatshields. His launches have to be a lot cheaper.