Would it be the end of the world if the Knicks pick outside the top 2?

Now that we know Steve Mills will be representing the Knicks at the lottery (tonight!), it's time to start coming to terms with the idea that the team will not be picking first overall. There's an 80.1% chance New York doesn't pick first, so the basketball gods would need to work in our favor. Sadly, there's just no way this guy will appease the basketball gods. This one? Maybe, but I doubt he's the Steve Mills who'll be on the dais.

As I'm sure you all know by now, the Knicks might pick as low as fifth. Most fans (myself included) would be really disappointed if the Knicks pick any lower than second; after having sat through the crapfest of a season we just did, it would be pretty terrible if we aren't even rewarded with a top draft pick. Also, picking second is, logically, preferable to picking lower than that since the player pool is larger and more choices are always better assuming the team knows what it's doing. However, what I would like to posit is: it wouldn't be that bad if the Knicks picked third, fourth, or even fifth.

I can support that by taking a look at the history of the NBA Draft. Below, you'll see a list of every player picked between second and fifth from 1985 (the first year of the lottery) through 2010, along with the number of win shares amassed by each player in his career.

I left the first pick out of this analysis because: 1) Karl-Anthony Towns is almost the consensus #1 pick, but the next four are much more up in the air; and 2) historically, first picks have a much higher success rate than the remainder of the top 5.

Of course, it's important to note that correlation does not imply causation, and that even if every player picked fourth from 1985-2010 has to this point had a better career than every player picked third over that same period, it doesn't mean that the fourth pick in 2015 will be better than the third pick. Let me say that again, for emphasis: correlation doesn't imply causation and the 2nd pick this year won't automatically be better than the 5th pick just because Kevin Garnett was better than Antonio McDyess. Alright, since that's clear, let's get going.

I also calculated the total, average, median, maximum, and minimum win shares by pick. Note: for the 2nd pick, I left Len Bias (who unfortunately passed away before ever playing in an NBA game) out of the calculations.

Somewhat surprisingly, the highest average win shares belongs to the fifth pick even though the highest median belongs to the second pick. However, comparing the median for each pick to the average reveals quite a bit. Averages are affected much more by outliers (players who heavily outperform or underperform the expected number of win shares for that pick number, like Chris Paul or Adam Morrison) than medians. So, one very rudimentary way to tell how trustworthy the average is is to compare it to the median.

With the third pick, since the median is larger than the average, there are probably a few more third picks who have underperformed than who have overperformed. As for the fourth and (especially) fifth picks, since the average is much larger than the median, there are probably a lot more of those players who overperformed than underperformed.

Given that the averages for the third, fourth, and fifth picks were all affected by outliers, it's better to look at the medians for comparison. The median number of win shares decreases a bit with each successive pick, but not ridiculously so. Therefore, historically, there hasn't been much of a detriment in picking lower than second, even if you try to ignore players like Chris Paul or Kevin Garnett.

Even though none of this means that if the Knicks pick fifth the player they draft will be almost as good as the one drafted second, it's nice to know that history tells us picking anywhere in the top 5 has worked out pretty well. That being said, forget all this measured nonsense. The Knicks better pick first overall so they can draft Karl-Anthony Towns.