Look, I'm sorry but a lot of your info is just wrong.
As this is a guide for coloring & texturing it would be nice if the information was more accurate for players, and I'm just trying to help you out here by providing the correct numbers for levels, etc. as soon as I have confirmed them.
The values of success are approximate values since every click produces different success (except the 323 cans/textures per click - that is always 100%).
While 323 will usually come out as 100%, there is a small risk of failure with that amount. You have even confirmed that youself before:
....Since I was using 323 textures per field and that is supposed to guarantee 100% I was quite surprised to get a result of 100% and 99.8%.
My numbers of average saturation % come directly from MA, you can simply check these in game yourself. The percentages are shown if you hover just below the "succes rate" bar:
As you'll notice when you check it in game, with 323 cans it will show 99%, this is because the game rounds down. It's actually 99.xxx%. That is why on occasion it can still fail eventhough it usually gives 100% results.
For example: With 100 cans/textures per click you can end up with 10%, 20%. 35.9%,90%.... its all over the map. It will never be near your average of 76%.
I've had clicks with 300 cans that failed to 30% saturation and clicks with 150 cans that were 100%. It's random.
How many clicks have you done for every amount of cans/textures to be so certain of your numbers?
I'm not the first person to test the amounts, Sarah did research on this before too:
https://www.planetcalypsoforum.com/forums/showthread.php?96981-Actual-CoS-in-Coloring-amp-Texturizing
Sarah's findings are similar to mine and the succes rate shown in game.
I've started keeping track a few weeks ago, to be able to provide players with even more insight in how the amount of paint cans / textures used will affect the outcome. (Because I always leave the final decision of what amount to use in the hands of my costumers). My experiment will continue for several more months to get even more accurate results.
My data so far is based on 200.000+ textures used, with 1300+ clicks:
As you can see, there are clear patterns emerging with different amounts. it's not all over the map:
(ignore the 150 and 250 data lines as I haven't finished them yet.. smaller amount tested)
I'll take 200 cans as an example here:
As you can see, most of the results are well over 90%, with only a few percent of the results ending up below 75%.
To show it in a nice graph to make it more clear:
There is just no way you could have tested this and come out with an average that low........
As for:
For example: With 100 cans/textures per click you can end up with 10%, 20%. 35.9%,90%.... its all over the map. It will never be near your average of 76%.
I really have no clue where you are getting your numbers from.. 200+ attempts so far and my lowest result was 39.3%.
That is just an extreme outlier, which hardly has any effect on the average.
To compare the average results found :
As you can see, the averages from my data are within 1% of MA's given numbers. Sarah's are all within 4% so I'm guessing her sample was a bit smaller then mine.
Yours however are up to 21% lower, that's statistically impossible.
My own amounts experiment is still ongoing. It will probably take me a few more months to finish it, with even larger samples of different amounts. All data gathered so far can be found on my website:
http://valnor.webs.com/amountssuccesrateinfo.htm
Will update once we have all the exact numbers 
Can you post a screen shot?
