I was bored at work one day and I wrote a little script to calculate these probability charts. The X-axis is your dice pool and the Y-axis is the probability of succeeding at the test with the given threshold number. So for example, to find the probability of succeeding in a test with a threshold of 4 and a dice-pool of 9 with no edge, you would look at the graph with the title "4 harder" and then see how high the black curve is on the Y-axis when you look above the X-axis value of 9.....and we see it's about 0.35 or 35%.
This might be a good guide to understanding thresholds...just translate it to probability of success, that's what I do.
(Ignore the small-scale wiggles in the red curve, I used Monte Carlo method to calculate rule of six cuz it's faster..... just smooth it out mentally. Also, the 2 dice you add to your dice-pool is already included in the red curve....i.e. If my original dice-pool is 8, I'd look at the value on the red curve for 8, not for 10...that would be the value with using 4 edge.)