[bisq-network/compensation] For Cycle 18 (#687)

[chimp1984]

Ah ok. I have not read the initial proposal before...

So after doing a few more calculations I think the problem is that we have both the USD/BTC volatility as well as the BTC/BSQ volatility as part of the model and we apply the 30 or 90 days average on both BSQ and BTC.

The calculation with 30 days average and excluded outliers (1% threshold - see https://github.com/bisq-network/bisq/pull/4706) would be:
BTC: 14705,88 * 0,00005652 = 0,8311763376 BTC
0,8311763376 BTC is 10861 USD @13068 BTC/USD
A average rate BTC/USD rate of 12031 would result in 10 000 USD. I have not run a calculation to check what was the 30/90 days average Btc price (from Bisq trades, which might be a bit different than global market rate), but it feels realistic as prices have just recently spiked to 13k.

<img width="1194" alt="1perc" src="https://user-images.githubusercontent.com/54558767/97127846-bd6ac380-1708-11eb-807e-499599c7db25.png">

I don't know how to deal best with the situation. The terms of the proposal/contract are matching the way it was calculated. I think the uncertainties of volatility on BTC and BSQ have been under-estimated but are a fact which we have to deal with. 
The chosen model distributes the volatility risk between the guarantor and @oscarguindzberg . If we would change it to convert the requested USD directly to BTC it would be only on the shoulders of the guarantor. 
Simple though example: Imagine BTC goes down to 3000 USD. The guarantor would lose then more than 3 BTC for a single months payment (and Bisq would likely not get segwit completed). On the other side if BTC goes up to 30 K @oscarguindzberg would only get 0.33 BTC and if BTC settles back down to 10k afterwards his monthly payment was much lower as expected. Of course we cannot avoid volatility but with the current model we smoothen out the risk over a period of time. It makes it harder to understand intuitively but I think its more fair and safe.

@oscarguindzberg What do you think? If you agree can you adjust your calculation with the adjusted rate where we filtered out the outliers which are suspect to price manipulation?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/bisq-network/compensation/issues/687#issuecomment-716273024
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bisq.network/pipermail/bisq-github/attachments/20201025/997978db/attachment.html>