BSQ rate for Cycle 12 is 0.56 USD per 1 BSQ

[MwithM]

At the start of cycle 11, the 90 days USD average price was 0.44. This is the recommended rate to be used for compensation on this cycle.

More info about the Compensation process at https://bisq.wiki/Compensation

[sqrrm]

I added a PR to calculate the volume weighted USD/BSQ price, which I think is the correct measure to use here. See screen shot. The average price over the last 90 days was USD0.56 per BSQ.

[MwithM]

I don't know exactly how this weighted average is calculated, but the sudden fall in BSQ price even using a 90 day average didn't look fair. 🤣
I'll wait a little more thumbs up to change the metrics for this and BSQ trading fee discount as well.

[sqrrm]

To explain a bit further why I did this. The current calculation is using the volume weighted average price of BSQ priced in BTC. This is a good measure for the average price of BSQ in BTC, but since we're not using BTC as our unit of account that doesn't really make sense. To just take the USD value of BTC at the end of the 90 days as the non weighted average price of BTC over that 90 day period is not a good measure of anything really.

My change is doing the same as we currently do, but measured in USD rather than BTC.

[ifarnung]

To clarify, if I understand correctly:

Current Method: 90 day Volume weighted BSQ price divided by Spot price of USD-BTC
Proprosed change: 90 day Volume weighted BSQ price divided by 90 day Volume weighted BSQ price

Seems like a totally reasonable change to make, as the whole policy of using an average is an arbitrary choice anyway. No measure is going to perfectly reflect the value one will capture for his/her BSQ, right? It's all just an estimate and stability of the policy probably being desirable for stakeholders. Would be fine to tweak the calculation if it's with the goal of moving to a more ossified policy and not just a way to bump up the low BSQ price for this month, right?

OFF TOPIC but related, personally, I would want to see using a 30 day average of the BSQ price and using the current price of BTC (ideally even closer to when CRs are due, but some lag is inevitable).

That .51 USD is pretty accurate to what you can exchange your BSQ for right now.

Using the longer time periods just introduces more lag. So in a declining price environment like we saw for BSQ in January, and BTC in March, people are generally being under compensated while those prices are falling which is kind of a double whammy. And in a rising price environment (like we see with BSQ today) people will be over compensated. Maybe we are seeing that phenomenon this month with the 90 day average of BSQ including the very low trades in January still, though they will roll off the average by next cycle. With stable prices, neither under nor over compensated, so no real issue.

[sqrrm]

@ifarnung Indeed, it's not clear that using an average over 90 days is the best method. This change was just to use a correct measure.

I support using a shorter look back period to calculate the price used for CRs. If we use any look back period it should be an average though in my opinion. Otherwise we might as well just use spot price at some point during the current cycle. For some kind of stability and to make it harder to manipulate I think it's better to use at least 30 days average.

[ifarnung]

Agreed, for BSQ an average is great as there are many outlier BSQ trade prices due to both the order book system of BISQ (any offer can be taken without respecting best bid or best offer) and the generally low liquidity of the BSQ token. For BTC it's not clear to me that an average is necessary as at any given time there are millions and millions of $ of liquidity at any given moment, as well as no worry of manipulation of the BTC price.

Thanks for taking the time to investigate the formula used and propose a fixed version! Let's see what others seem to think, if they care about this minutiae. 😂

[MwithM]

If nobody is against it, tomorrow I will edit the title of this issue to use the weighted average price over the last 90 days: BSQ rate for Cycle 12 is 0.56 USD per 1 BSQ

I'm in favour of using the 30 days instead of 90 as well, but as this is a bigger change, I'd rather implement it for Cycle 13. I think this also has implications for UI. The DAO wouldn't be using 90 day average for anything, so this metric could be removed and display just 30 days average.

[sqrrm]

Changing to using a 30 day average should be its own proposal. If you think it important you can create one, I would support it unless there are some very good arguments against.

[m52go]

Agree -- this change makes sense, so thanks for implementing. I'd also be in favor of a 30-day average.

[cbeams]

Just gave my 👍 here as well. Thanks @sqrrm for putting together the PR, and I agree with the change in the BSQ rate to reflect the weighted average.

I'm uncertain about the move from 90 day to 30 day. Need to think about it more.

[MwithM]

90 days volume weighted average was used.