This answer only addresses the second charts. I'll let Mathematician Matt Parker address Benford's LawMathematician Matt Parker address Benford's Law.
I can confirm [the result is] actually exactly what you'd expect, that's not out of order... And secondly Benford's Law is not a good test for election fraud. And I quote [from Benford's Law and the Detection of Election Law (2011)] "Benford's Law is problematic at best as a forensic tool when applied to elections".
To the graphs, the vertical scales are different. Narrow vertical scales make changes look larger. While wide vertical scales smooth out changes. Biden's graph is using a more narrow scale than Trump's.
I put them all together in one graph with the same scale and they don't look so different anymore.
I haven't verified the data from the original graph is correct. I had to eyeball the numbers from the graphs.
It is suspicious because someone had to choose to use different vertical axes for each graph. It looks like a case straight out of How To Lie With Statistics.