I do understand, to a point, statistics. Do you?
It's part of the work I do in a scientific field. So, I do a lot of research looking at various data and have to understand whether something is statistically significant, or down to pure chance, or even somewhere in between. The way this is usually done is by applying probability values to it and looking at things like "Confidence Intervals". A confidence interval is a range of values so defined that there is a specified probability that the value of a parameter lies within it.
It does get very technical because essentially you need to establish a probability value of p > .05 to be sure that the result you are looking at is statistically valid. In simple terms, this means that if you repeated a test over and over again, then for the result to be statistically significant the p value of >.05 means you would need to get the same result 95 times out of a 100.
That's how "probability" works in science.
Of course, football isn't science, but there's an important analogy to draw here. If you are going to try and establish a relationship between "cause" and "effect" that goes beyond "co-incidence", then you really do have to apply a bit more rigor to the data you are looking at. Otherwise, it's meaningless.
To suggest that having any one player, including Mr Done, not just taking part in a match, but actually "starting" a match has a direct effect on the outcome of those matches is a massive claim to make. It's huge and it couldn't be validated statistically. The reason why? There are too many other variables.
If you have a basic understanding of statistics you'd know that this could easily be coincidence. At best, it might be that there's some sort of "emerging trend" towards better results when he starts. (Apart from those first 5 games at the start of the season - of which we lost 4 - or should we take them out of our analysis in case they spoil the figures?).