From what I have been led to believe, the tests were designed to be read with the tube being held up against the chart - and having sunlight - full spectrum light - shining in from the front.
Since the majority of us do not have the luxury of running outside into the sunshine to read each test, over the years people have come to read the tests by moving the tube in and out from the chart, in as bright a light as possible.
The colors are difficult to tell apart - particularly at the higher levels. The way I look at it is that each of those "colors" actually covers a very wide range of ppm of whatever you are looking for. The test is an approximation of what you have - not an exact reading. In the lower concentrations - ie: 5, 10 or 20 of nitrate, the colors are pretty easy to tell apart. When you get up into the high concentrations they are really hard - but you know that once the card reads really red - a red that is difficult to tell apart from the other reds - you are already in the "oops - hafta change the water right away!" range anyway.
With a reading that is really high, you will not be able to lower it much with just one 50% water change. You can count the 50% water changes you need to get the reading below 20ppm and you will have a much more accurate idea of just how high your reading was. For example: 300ppm - one change is 150ppm, 2 changes is 75ppm, 3 changes is about 38ppm, 4 changes is about 19. It took 4 50% changes to get it below 20 ppm. You can go backwards and find out exactly what you had at the beginning.
I guess the main point I am trying to make is that any color red (or whatever your particular test uses) that is so red you cannot tell it apart from the other reds is too high and needs to be addressed. The exact numbers cease to be very important....