There’s a bit of an issue with the graphs in that article, as they are not directly comparable, and IMO, a conclusion is misleading.
In graph 1, Mean Absorbence on the Y axis shows that the regular SPF50 is in the order of 10 times more effective than DIY. The Y axis scale is log to the base 10, 1.00=1, 2.00=10, 3.00=1000 times more absorbence, or less transmittance.
This is a bit misleading: “As the chart shows, the homemade version has about half the absorbance of the commercial product.”
How many people know that absorbence is a log scale? I suspect most would think it was about half as good from that statement.
In the 2nd graph, it should be noted that SPF 10 is equivalent to an absorbence of 1.0, ie 90% blocked, 10% transmitted.
SPF 100 would be an absorbence of 2.0, or 99% blocked.
However, this is strongly dependent on how thickly the sunscreen is applied.