Typical player Strength here

I look at player distributions on line and it says for FIDE,  the median strength is 2000 and the distribution appears normal.

I play here and 1300 something puts me in the 85th percentile.  Surely this includes very low level players who played for a while and quit.

Is there any way of seeing my rating among active players?

Ironically, chess.com statistics ARE for "active players" (played within last 90 days I believe). median 2000 FIDE sounds really far off though. Are you sure you didn't misinterpret that statistic somehow? Granted, chess.com rating pool would be "softer" than FIDE or OTB organizations, but the "average" chess rating is around 1200 or so I believe (1200 ish  for chess.com, FIDE, USCF etc. although each varies slightly).

Don't know how accurate this is, but a simple Google search gives 1767 as "average" which is at least not 2000. 

https://www.reddit.com/r/chess/comments/44jtnr/fide_is_868_male_and_the_average_rating_of_male/ 

Understanding any average takes an understanding of how rating pools work though. 

I saw it in a graph depiction from a Google search. In any case, 1300 something 85th percentile is no where near 1700 something 50th percentile.

I understand how pools work.

I played at another site that allows non members to play and they get a 1500 rating. I think that site has inflated numbers accordingly.

85th percentile among active players lol. I may start getting a big head.

FIDE's median is skewed because they won't officially rate players who fail to score high enough to reach 1000 FIDE.

Exactly this.

It's impossible to have a FIDE rating below 1000. There are a bunch of players in my city who play FIDE tournaments regularly and are classified as unrated simply because their FIDE playing strength is somewhere around 700-999.

So if you're only counting "rated" FIDE players, then yeah, the median is going to be a really high number like 1800.

Telling people they are not worthy of a rating seems to be a bit arrogant.

Honestly the FIDE rating system is quite primitive.

No one below 1000 gets a rating. Your rating only updates once a month (it should update after every game, or at least after every tournament). The way they calculate it doesn't use the same glicko system that major online chess sites (such as Chess.com) use.

It's not an ideal system imo.

It's disgusting to be honest

Lol, when I was a kid, FIDE ratings started at 2200. They then reduced it to 2000, where it stayed for quite a while, before they extended it right down to patzerville.

What OTB rating should be considered more accurate?

This is the kind of arrogance that FIDE created.

Perhaps for the game, FIDE should have remained at 2200. I am sure the change was made to increase the FIDE financial situation.

No, I mean a good over the board rating for the common person.

No, I mean organization rating for the common player.

I was looking at their chart from 2004. 800 to 900 put a person in the 50th percentile.

http://www.uschess.org/archive/ratings/ratedist.php

Given that USCF ratings are generally 50 to 100 points higher than FIDE the average American adult FIDE is about 1425 to 1525 FIDE.

Found this thread with a Google search

https://www.chess.com/forum/view/general/fide-ratings-vs-chesscom-ratings-explored

It says the rating here for live chess is 100 to 150 points below the FIDE rating. So adding 125 and 75, you can reasonably estimate your USCF rating by adding 200 points to your live rating here.

My rating was not boosted.  As boosted ratings stop playing and new players join, the numbers will regress to what they used to be.

You obviously missed the point, while helping to demonstrate it.

Sometimes referred to as a Bactrian distribution.

The Bactrian Camel has two humps instead of one.