Just my opinion, - but the key to success with engine-off landings is preventing the engine from going off.
Let's talk about Robinsons, as they do more than half of all single-engine helicopter hours and more than 90 percent of private training.
I'll say it again - as far as I can ascertain, no R22 pilot who got into autorotation has ever been killed on landing, either in countries where EOLs are mandatory or elsewhere. They die because engines fail, and they don't react quickly enough.
Is it possible to genuinely teach students to react in time? I don't think so. Bearing in mind the strictures against surprise throttle chops, it's difficult even to demonstrate engine failure. Requiring a pilot to be sufficiently keyed up, every second of every minute of every flight, to react in time is also asking the impossible.
But it is also true to say that most engine failures in Robinsons are caused by carb icing.
Therefore, it is important that the student know that if they forget everything else they were told and shown, they remember carb heat. Stamp it on their foreheads, write it on the bubble in chinagraph, go bananas when they forget.
Being able to land blue-side-up engine-off is a virtual irrelevance. Shame to smash up so many good machines practising shutting the stable door, again and again and again.