Maybe I missed something, but this method seems sketchy. Merely by counting the number and type of troops at a battle, the author then assigns a % chance for a win on each side. Since Napoleon won numerous battles, and there was much more data for his battles than others like Alexander the Great, the author concludes he was the best general ever. And yet all he has done is look at troop numbers. What about the succession of battles in linear time? The morale of the troops on both sides (side note: Napoleon thought morale was by far the most important characteristic of an army)? The conduct of lower ranking officers? The technology that is brought to bear?
These unaddressed factors may clarify the analysis of General Lee, for instance, who was facing extremely incompetent opponents for the majority of the war, and by the end he was then facing more advanced weaponry, better supplied opponents, and another competent general in Grant. I am not on the Lee bandwagon by any stretch, I think he made some risky moves that paid off, but he also made an absolutely unforced error at Pickett's charge which completely lost him the troop number advantages gained at other battles. And if you're facing McClellan who is incapable of attacking you for months despite superior forces and odds, then you don't truly deserve heaps of accolades.
These are all crucial questions that lie far beyond the author's simplistic approach. The data simply isn't enough to warrant the conclusions, nor does the author arrive at them in a statistically sound manner.
These unaddressed factors may clarify the analysis of General Lee, for instance, who was facing extremely incompetent opponents for the majority of the war, and by the end he was then facing more advanced weaponry, better supplied opponents, and another competent general in Grant. I am not on the Lee bandwagon by any stretch, I think he made some risky moves that paid off, but he also made an absolutely unforced error at Pickett's charge which completely lost him the troop number advantages gained at other battles. And if you're facing McClellan who is incapable of attacking you for months despite superior forces and odds, then you don't truly deserve heaps of accolades.
These are all crucial questions that lie far beyond the author's simplistic approach. The data simply isn't enough to warrant the conclusions, nor does the author arrive at them in a statistically sound manner.