Why the difference in readings?......

It's certainly not stupid and you will get a variance all the time. The monitors are approx 15% percent out either way. What you take is the lower the reading the closer it is. So for example if you have a reading around 7.2 after 2 hours after food and it's within your range then it doesn't have to be so accurate.
You also get rogue readings occasionally. Just retest and if your are within 2.0 of your pre meal reading that's good control.

Let's say the 'real' number was an average of the two: 7.7

+15% of 7.7 = 7.7 + 0.15 x 7.7 = 8.9
-15% of 7.7 = 7.7 - 0.15 x 7.7 = 6.5

So, your readings of 7.2 and 8.2 are both 'correct' as they are within the accuracy of the meter.

When I do two readings I either throw out the one that seems to be obviously wrong, or take an average.

Hi Izzy,

It's a pain isn't it? The reality is that our little home meters aren't that accurate, and we just have to accept it - even though it is very frustrating.

So the best way to use them is by spotting trends, while treating individual readings with a pinch of salt.
For instance:
I eat the same breakfast 4 days in a row. Each day the reading goes up by between 2.8 and 3.5 mmol/thingy.
I decide that is too much of a rise (aiming for no more than 2mmol After a meal)
So I cut the carbs in my breakfast, and test for another few days.
This time, the readings go up by 1.3-2.2 two hours after eating.
Result! I'm on target.

So it isn't the individual readings that are important. It's the trend.

That's why we all plot the readings on apps and spreadsheets (well, maybe not ALL ), so that we can see the trends as they happen.

Hope that helps.

The blood can indeed be different as well.

Capillary blood can have a higher reading than venous blood, the fasting is supposed to be most accurate.
But exercise can change the BG levels, temperature can, so it's possible to use the muscles on one side, (cleaning teeth, showering, running your hand under the top etc) and get a change. That coupled with variation from strip to strip can amplify that change.