Those figures seem off to me when applied to the smart metric time periods which mark night at 23:00 - 08:00, or 37% of the day. Short of having an EV I can't see any pattern that would put so much energy use at that time, while asleep, with no lights or applications in use.
Even less so for anyone who works from home.
Really though if smart plans were better for consumers, it would be easy for providers to do a you could have saved X over the last billing period by being on a smart plan. Instead they can only show these you could save Y IF you moved to the plan AND moved some of your load from day to night time.