Every day i set up and run numerical calculations to predict the performance of my ideas. Some of these calculations can take as little as an hour but some can take as much as a month. In average they take at least a few days so it is extremely beneficial if i can predict the outcome of a calculation after just a few hours as that way i can save days of calculation time.

For those of you who do not know, numerical calculations consist from a set of equations which loop until a stable solution is obtained.
We start with a set of input values for which that equations finds a set of solutions; this set of solutions is called an intermittent result. The intermittent results form the input values for repeating that set of equations. This is called an ‘iteration’ and a full numerical calculation can consist of many 1000 iterations.

Now let’s take a look at 2 graphs:
https://ipfs.pics/ipfs/QmX7sgEpaZHzZ7UF96jCcWVRBamuDHrkhft4dvmTcSFr9m

The top graph represents values of a calculation i have been doing today and the bottom graph represents value of STEEM.
See any similarities? I do too.
The top graph represents values of torque for a part i am working on; the values are plotted for each iteration and then connected. Similarly the bottom graph has values of STEEM plotted for every day.

It is a coincidence that the two graphs look similar but it is not a coincident that their trends (below in red) look similar:
https://ipfs.pics/ipfs/QmVVCDL9nuRv4jKfe1kCPam16MxdUsm1CHj51Viuu7sECX

Be it in a complex computational simulation or in a stock/currency market we always will see fluctuations in observed values. In short term they seem wildly unpredictable but once you have examined enough of them you realize that in the long term there are only a few trends. 
The overall long term trend can be a simple linear up-slope or down-slope, it can be a U-curve (Gaussian curve) turned up or down, or it can be an exponential up/down slope. At the moment it looks like STEEMIT has a Gaussian-like curve, but bare in mind we are seeing only a medium-term trend (we have only a few months worth of data).

Can we guess the long term trend from this?
Yes, with a reasonable certainty but always keep in mind nothing in life is 100% certain.

Ok, approaching this from an engineering standpoint we must first consider if the values can fall out of ‘’bounds’’. The bottom bound is 0 as if STEEM reaches negative values it is unlikely to survive; the top bound is infinity. Infinite value does by default not exist in our market so this bound is secure. The bottom bound is more tricky though - if investors bail out the platform could experience a slow and painful death but its owners would still have the time to look for alternatives. Since STEEMIT has so many users now and it has gathered considerable financial backing i think it is safe to say it will not die out. In the worst case scenario STEEMIT will just be bought by a bigger online player.

For me this is good enough to put in the money i can afford to lose. But not just yet as i am waiting for STEEM to fall under $0.1.
Of course i cannot be sure this will happen, but judging on what i have seen in 2008 stock market crash and the gold price variations in the last 15 years i think STEEMIT’s long term trend will be roughly like this:
https://ipfs.pics/ipfs/QmbMKJwEfRMN7Yw7TuLGAPeYTqwytfqjHoDE2FNHe2YdNz

Nearly any (natural or unnatural) system undergoes a periodic fluctuating pattern like this when a disturbance occurs in the system. This fluctuation gradually dissipates and the system eventually stabilizes. 
In case of STEEM, starting and releasing STEEMIT into the public was the ‘’disturbance’’ and the system is now going through its natural process of stabilizing. Looking short-term one can easily mistake the plummeting steem value for a sign of Armageddon, but i think the ‘’bounds’’ are defined well enough for STEEMIT to survive these fluctuations and that STEEMIT has enough energy (backers + users)  that it can stabilize with a gently raising linear slope.

If you are asking why would STEEM not just ‘’stabilize’’ at $0.0, let me explain it this way:
*Imagine STEEM as a rubber bouncy ball. 
Imagine the financial value of STEEMIT as the floor.
You drop the ball from a certain height (this is the initial financial value with which STEEMIT was founded) thus giving it potential energy. As gravity is pulling the ball down its potential energy is turning into kinetic energy. When the ball hits the floor it bounces back up. The higher it bounces the more potential energy it has and this energy attracts investors. Investors raise  the financial value of STEEMIT and thus raise the ‘’floor’’. So the next time the ball bounces it will not bounce off ground 0, but off step 1 (then step 2, then 3, etc). Of course at each step something is lost to friction (people cashing out) thus the ball gains less and less height till it eventually comes to a halt on step # X.*
https://ipfs.pics/ipfs/QmTVQQMRBT5B1pB62vQiTZCGjUxvgN3FWnsZKLLEwY741d

At that point the founders/owners will need to find a way to monetize the platform (maybe by selling ad-space) and thus get the ball bouncing again. But that is probably at least 1 year away, so for now i will not yet speculate how it will be done.