There are many methods to predict the market prices, and for digging-out the data to invest in. Since the shit-coin season is on a horizon, here is just one of infinite ways you can use to find where to plug your change-money.
First thing, remember, BTC currently dominates everything and everyone! The shit-coins are there just to pump the BTC and whatever you invest is at the high risk because of this kind of a market dominance. Therefore, investing in shitcoins is a placebo that follows a BTC trend.
Since it is a placebo, a dose of science and data-analysis may help choose the right sugar-pill.
If we could somehow take a look at the whole market and detect the outliers, with a few basic assumptions, we may be able to predict the trends.
ALRIGHT! lets do that, no programming skills required.
Ingredients
- Spreadsheet program and a text-editor
2. Data taken from: https://coinmarketcap.com/all/views/all/
3. Download a 30-day trial (or register it) http://52.45.171.32/download/eureqa-desktop-download/
Note: The 30-day trial can be used only once, so be careful where you install it, if you want to use another 30-days of a trial after the expiration. Eureqa is heavy on CPU, and to obtain the fast results, you may need a help of a GPU card.
STEP 1
Go to https://coinmarketcap.com/all/views/all/ and highlight the first 3000 (or less) coins from the list. The list is sorted by the rank. Paste the list in a spreadsheet program (this may take a couple of minutes to paste data). This is what the data will look like.
Now, you will need to massage this data a bit, to get a result that looks like this:
To remove the dollar-signs, commas, and other useless characters, you can use the search-replace in your own text-editing software (Notepad++, Geddit, or similar).
STEP 2
Install the Eureqa software, and copy-paste the data into it. The final result should look like this:
Please note that you don’t need the name of a coin, just numbers.
Now, you need to tell the program to skip the missing data. Go to a Process Data tab, and for each value, remove the missing data (don’t do anything else).
Alright, now move to a next tab called Define Search. There are three things you need to do here.
- Make the changes in the The Target Expression field to make sure you are solving the data for the rank.
2. Check (almost) all fields, see below which ones:
(all except the Integer Constant).
If you want to include the non-shitcoins in the search, do not check the fields under History (Delayed Variables, Simple….).
3. Choose the Hybrid correlation/error metric
PRESS PLAY!
STEP 3
Go to Results tab, sort the formulas by “Fit” and change to Residual Error Plot graph. If the dots are too big or the sound annoying, just go to settings and turn off the sound. Browse around and set the pixels of dots to a lower number.
STEP 4
After letting the program to run for a few hours, you will need the Residual Error Plot to see the outliers. Choose the dots, that stand-out, zoom-in by selecting the field around them.
From the graph above I have chosen the dot with the rank 1127, which translates to “TrueFeedBack”.
STEP 5
Invest! If the coin exists on the exchange you like (I only choose Binance or Hotbit), go there and place your order. TrueFeedBack sounds shitty enough, probably an Asian pump and dump scheme (only they can come up with such bad names). Checking… YIPE! team looks like a pump-and-dump clique, and they are all from Turkey. AWESOME! Lets put $10 on it and see what happens next.
BE EXTRA CAREFUL!
Purchasing coins/tokens is easy, selling them IS NOT!
