Don't be despondent! You still have time to get speed and accuracy up. Also, having done the mock test, in theory your DD should feel less nervous for the real thing.
I can't help you with the content of the tests your DD will be taking (different area) but I would suggest you go through a couple of papers she's already completed, or do one together so you can identify the gaps in her knowledge, if there are any. Is she spending too much time on any particular type of question? She should mark these and come back to them later, not get held up there. It could be that she's like my own DD and just makes silly mistakes because she's rushing through it and not reading the questions properly. This thread has some tips on silly mistakes:viewtopic.php?f=19&t=15128&hilit=silly+mistakes
If you search through the sections there are lots of educational sites posted for fun games which help build up mental calculations. DD loves: http://www.arcademicskillbuilders.com/
Remember these scores are relative to that test. These test results must be considered in context of the real tests to be taken. The Sutton mock NVR is 30 questions. For example a child getting 10 questions wrong scores 67%. If the real test is 60 questions then 10 wrong would be 83%. It doesn't follow that they would only get 10 wrong, but equally it doesn't follow that they will get one third wrong either.
Is the 326 cut off standardised scores for all subjects? The standardisation used by Sutton mocks looks a bit odd to me as well, as if the DCs ages are not taken into account. If you look at all the 100% maths scores they all have a standardised score of 131 which is the higher end of the scale (ie younger children). Equally, all VR scores of 51% have a standardised score of 81. Not sure how this works, but it can be misleading if this is what you are looking at.
The benefit of these mocks imo is that they give excellent practice of a test environment and also can help you identify areas to work on or improve.
I hope this helps! Good luck.