As I mentioned in the blog, EPA number is meaningless. It's like I see a green apple, others (ie, Tony Williams) see a green apple, but I tell myself and everyone else that the apple is red because EPA says so. That's nonsense.
Biggest problem with EPA rating is that it's a hodge podge of some arbitrary "stuff" and not enough granularity. The test I and Tony Williams perform is least variable possible given the circumstance. What EPA should do is produce graph like I made: mi/kWh over speed and added power use (also over temperature) which shows what the car is capable of with and without hodge podge. In effect, looking at EPA is like looking at box of chocolates: you never know what you get.
For example, I got 5.3 mi/kWh average after 16K miles. Assuming 80% efficient charging (L2 is actually over 85%, DCFC over 90%), that works out to 4.24 mi/kWh, or 142.9 MPGe with 33.7 kWh/gal. That's over 30% better than EPA's 109 MPGe. For some who got over 6 mi/kWh (4.8 mi/kWh after 80%, 162 MPGe) after 10K miles, that's almost 50% better than EPA rating. Box of chocolates and red apples, indeed.
To be fair, EPA rating works "ok" with gas cars. Every gas car I've driven was close to EPA rating, though rarely any ever got above it. EPA was bit optimistic when it came to gas cars, but they're way pessimistic random number when it comes to EV, at least with regard to SparkEV, probably others as well; you can't capture efficiency that could vary by over 90% with just two numbers.
However, forum post you link is interesting. It does seem to show Ioniq beating SparkEV in efficiency at 70 MPH. But if you read further down, another posted "In 5 degree temperature at up to 60 mph motorway I got 147 miles. Shows 4.9 miles/kWa on the trip".
https://speakev.com/threads/hyundai-ioniq-electric-range.21219/page-5#post-396857
The battery is 147/4.9 = 30 kWh, not 28 kWh. Then doing the math with % used at 70 MPH gives 5.06 mi/kWh and 4.79 mi/kWh. That's still better than SparkEV at 70 MPH. But Tony Williams got 5 mi/kWh with SparkEV at 62 MPH while Ioniq "up to 60 MPH" got 4.9 mi/kWh, which is better than Ioniq. Tony was running more favorable close to 20C, so the jury is still out.
Even if Ioniq or another car turns out to be more efficient than SparkEV, that doesn't change the fact that it was the most efficient car in history up to that point. Records are meant to be broken.