We need to define a degree of "peggedness", maybe?

[cryptog]

The 24h volume indicated by coinmarket is 3,000 usd. So I would say that the data was not enough to determine an avg NBT/USD .

[Ben]

Why look at it in the single day timeframe? Take a look at the 90 day chart, and focus on the market cap value. With the exception of the spike when they were taking into account the reserve fund (which CMC has now taken out of the supply calculations), the value has been very stable.

[cryptog]

Why look at it in the single day timeframe?

Well because I think it is important to check whether or not nubit is stable over 24h.
Which makes me wonder how you set the price of nubit.
Is it the average price of the highest bids across exchanges weighted by volume?

[Ben]

It depends on the market that you are asking about:

NBT/USD
NBT/USD pairs are easy to set – 1 NBT = 1 USD (with a tiny spread to take into account exchange commissions)

NBT/{crypto}
NBT/crypto pairs are based on the current exchange rate of the crypto/USD. These markets (NBT/BTC, NBT/PPC) are updated 10s to 100s of times a day to keep in line with the most up to date pricing possible.

Prices are set based on a number of recognized rates, such as Bitstamp, BTC-e, CoinMarketCap. Primary feeds are checked against secondary feeds for discrepencies, so NuBot isn’t prone to “following the wrong price” if fed bad data.

NBT/{fiat}
NBT/{fiat} (non USD, right now basically EUR) are based on the daily exchange rate between FIAT/USD. The exchange rates don’t change frequently so NuBot is only programmed to attempt to get prices from the pricing API every 8 to 12 hours or so.

---

NuBits has been (and continues to be) stable over all timeframes. CoinMarketCap can only be trusted so far as its latest update. The current price on a given exchange may be different from the pricing you see, because a trade may not have been enacted using the newer values.

I don’t know what your screenshot shows (whether that is the “total” market view on the primary CMC screen or if it’s looking at a single exchange), but the reason you’re seeing $0.983982 has less to do with NBT’s stability, and more to do with the volatility of BTC and PPC.

Here’s an example:

If a trade just happened on BTER or CCEDK, and 100 NBT were traded for $100.02 worth of BTC (at a price of $370.00/BTC), you’d expect to see “$1.002/NBT” as the price. If the price of BTC updates to $365.00, CMC will update all other crypto prices (that came from X/BTC pairs) relative to that price change.

NuBot will have reset the walls to take into account the new pricing, so when the next trade happens, the new price will be reflected. Until then, however, it’s my understanding that you’ll see the price for NBT on CMC listed as “$0.988459”.

By necessity (they can only update after something happens), CMC is a lagging indicator of an exchange rate.

[tomjoad]

Nice reply @Ben. I’ve added a very slightly edited version of this to our FAQs. https://nubits.com/about/faqs#how-is-the-price-of-nubits-determined-on-coinmarketcap

[desrever]

I want to make a webpage with historical price of NuBits. What is the best metric to show value of 1 nbt? Do you think that using CMC will be misleading or bad advertising at times?

[cryptog]

@ben Tks a lot for your explanation. It makes a lot of sense.
The lag in data reflection is compatible totally with my own experience because a few minutes later it got “updated” and displayed a much larger volume with a peg into 0.99xxx

I don’t know what your screenshot shows (whether that is the “total” market view on the primary CMC screen or if it’s looking at a single exchange),

I think it was looking primarily at CCEDK NBT/BTC (meaning the other pairs were insignificant in terms of volume) with a volume of 3000 usd or so.

[Ben]

If CMC showed the value of NBT against the USD, instead of measuring it against BTC, I think it would be a lot better.

[cryptog]

NBT/USDNBT/USD pairs are easy to set – 1 NBT = 1 USD (with a tiny spread to take into account exchange commissions)

However,
on CCEDK NBT/USD,
on the buy side you have 0.998, on the sell side you 1.002. with a different quantity.
How do you get 1NBT = 1USD? Is it the value at the middle? But in this case I would think that you need to regard the quantity of bid/ask as a weight, probably, to compute a weighted average.

[tomjoad]

However, on CCEDK NBT/USD,on the buy side you have 0.998, on the sell side you 1.002. with a different quantity.How do you get 1NBT = 1USD?

That 0.002 USD discrepancy is the exchange fee Ben is referring to. The second paragraph of this section of the white paper explains it in a little more detail.

[cryptog]

@tomjoad So in a sense CMC knows about the inner workings of the API used by NuBot and will always display 1NBT = 1 USD at least based on the data feed of NBT/USD ?

For NBT/BTC the situation is different because you need to recompute based on BTC/USD.

By the way, on CMC, why is that there is no NBT/USD data feed right now?
NBT/BTC is by far the largest pair.
At the end of the day only NBT/USD counts and NBT/crypto does not help us advertise 1NBT=1USD since it causes a time lag in NBT/USD computation. For example right now it is 1NBT=0.977 USD.

I think we should be impeccable in the peg display, if possible.
We should probably create our own real time rate display Web service.
Is it on the agenda?

[tomjoad]

We should probably create our own real time rate display Web service.Is it on the agenda?

Custodians and NuBot operators are more qualified to answer your first comments, but yes we are working towards more live metrics on our website.

[Cybnate]

@tomjoad So in a sense CMC knows about the inner workings of the API used by NuBot and will always display 1NBT = 1 USD at least based on the data feed of NBT/USD ?

Nope, CMC doesn’t know anything about inner working or has any fixed pegging. They just show the price of the NBT/USD pair which is kept at 1 US$ by custodians. I’m pretty sure that when there are a number of buys at 1.002 and no sells, CMC would show 1.002 for the NBT/USD pair. On average it would show 1 USD. Also looking forward to the live metrics and statistics on our website.

[cryptog]

@Cybnate

but do you know how they set the avg when there is a buy wall at 0.998 and a sell wall at 1.002? Is it weighted in some ways?

[Ben]

Unless there is a measure used that I’m unaware of, CMC doesn’t look at the buy or sell walls, only at the trade data after a trade has occurred.

[Cybnate]

Memory says that this has been discussed before on either peercointalk forum or bitcointalk forum a while ago. But I can’t find it. You might try to ask them directly, they are running an ongoing support thread in the bitcointalk forum.

[cryptog]

@Cybnate tks for the pointer.