@PArike I send you a small problem or discrepancy, it's very strange.
Currency PHL using your exponential formula gives a more or less close value, but it is dominated by auto exchanges and it does not work for me, but good. But when I choose the complete formula, I mean nethash, blocktime, reward ...... The result changes an awful lot to worse. I have checked the data several times, can not give a result that is 1/3 of the result with the exponential formula. To see if something is wrong, I think the blocking time may be wrong in the formula, or it is simply something else. I leave you the full currency data so you can configure it.
PHL X16r block reward 2.5 Blocktime 150 seconds
Difficulty
http://explore.placeh.io:8080/ext/summary token json data.[0].difficulty
Price
https://api.crex24.com/v2/public/tickers?instrument=PHL-BTC token json [?(@.instrument=='PHL-BTC')].bid
Nethash
http://explore.placeh.io:8080/api/getnetworkhashpsWith the exponential formula in the same machine 0.0014 6x 1080ti, with the complete formula 0.0058 the difference is huge, I do not understand it, it should not be like that, something fails.
I have done the measurement with exponential and the correct profit would be 1.25 but only when the auto exchanges (zpool, zergpool etc ...) do not give war.
It's not a coin I'm going to use, but this huge difference has caught my attention that something is not right. The formula with all the data can not be so far away. Somehow I do not trust the blocktime that is 150 seconds 2.5 minutes, but it is not a non-metric value.
I await your conclusions.
First a general answer. The formula for calculating coins per day using Difficulty-Reward-ExpFactor works for almost all coins and the formula using NetHash works for many coins as well. There will however always be coins with some unique properties where either one or none of these formulas can predict the number of coins per day. That's not an indication that the formulas are wrong, just that they may not be usable in a few cases.
Coins based on CuckaRoo29 / CuckaToo31 / CuckooCycle coins are examples where you never will get any correct values out of the formulas.
I just wanted to highlight this in general as I assume you will play around with quite a number of coins to do these calculations, and eventually you will find coins where they do something unique and the numbers will look strange.
In this case I entered the sources of information as you did, and from what I can see the Difficulty and NetHash forumula gives almost the same result. It was only 1 - 2 percent difference between them. However, in my case I used Block Time 60 seconds (1 minute) as this is what both CoinCalculators.io and CoinToMine reports. If you entered Block Time 150 seconds, it probably explain the difference you got.