You're not impressing anyone trying to say that you are some master data wiz, yet you fail to provide any analysis but IF you did (which you are not going to do), it would lead to some "magic" and "paradox" and the result will be totally different. I do work extensively and teach business analytics (and R in particular) but that is besides the point. Don't try to use the "I am so good" label on yourself, of some magic result.
On specifics,
1) Obviously, you avoid discussion on totally incorrect statements like "sample size is really small," when the data used actually uses the entire population since 2010 (the Calipari era), every single player drafted, and raw data presented.
2) The analysis is truly overwhelming (and pretty good actually). It does give Kentucky (and others) a huge advantage in metrics such as 'total draft score", which specifically accounts not only for drafted position, but also for the recruits' HS rankings (using 247). Furthermore, it gives Kentucky (and others) a huge advantage in "net score," which also adjusts for "busts", a statistical metric of the top-25 players who went undrafted. So it also accounts for the classic argument (often on this board as well), that some players were busts, high-ranked recruits who did not pan out.
So don't point to a "mythical and fictitious" analysis that would point to a different results (but you are not going to do) or try to impress us with personal statements of how good you are. If you want to present your own statistical analysis to support your "paradox" or contrary result, then do it. For someone as good as your are, and with all the raw data actually at hand (you should thank the person giving you the raw data in spreadsheets), that would be a piece of cake. But my guess is that your big data wizardry will just stop short at yet another post of how good you are, how good your own analysis would have been, and how different your results would be -- IF you did it.