Jump to content

An ODL update (Dec 2020)


Recommended Posts

  • Replies 64
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

I haven't posted an update on ODL traffic for some time, here's an update. Please note that I am only able to infer ODL traffic based on 'on-ledger' transactions - if there is ODL traffic that occurs

So ODL seems to have fallen off a cliff again. The last week - starting on wednesday, things seem to have taken a nosedive. One of the USD-MXN corridors/channels (orange) seems to have shut down and a

Excellent material @jbjnr - it's greatly appreciated !! You might have missed some of our past discussions in the "Bitso thread" regarding the dynamics in ODL in relation to market prices ? You

Posted Images

   This is just an observation and by no means have I gone into ODL to the depths that both the above have seem to have done. Periodical there seems to be testing between new pairs for a week or two that bumps the overall volume from currency that are not daily. This seems to bump the volume of ODL for a week and then it disappears for some time. If I remember right that happened with the Euro one and off for about two months before it was and has been in continuous use for a while now.  From that if we see the Thai Baht/USD along with AUD/PHP ramping up in the next month few months then we may be able to tell which corridors are coming online a few months in advance.

Link to post
Share on other sites
8 hours ago, JASCoder said:

You might have missed some of our past discussions in the "Bitso thread" regarding the dynamics in ODL in relation to market prices ?
Your charts and analysis appear to support my suspicion that ODL is operating under some constraints based on token volume not value.

In the sense that volume xrp = (volume $) / (price xrp) I can see how a volume constraint would exist, but I'm not sure why ripple would constrain ODL to operate based on volume of token rather than monetary value. Perhaps you could explain again what it is that you think you've observed as I might not quite understand you.

 

8 hours ago, JASCoder said:

You also mention the potential impacts from the recent introduction of more fiat pairs - like Bitso has recently done - this could be having some impact on the viewable metrics I'm now suspecting (see below).

That could be the explanation to the depressed volume this last week (see chart) of XRP->MXN trading. 

It could be. I don't see too much variability in the bitstamp->bitso channels  (what I see, is in the ODL plots above, which look reasonably well behaved) - if you are looking at 'trade volume' and not ODL volume, then it could be almost anything that is affecting the variability in your plots - do bitso provide historic data for (daily) trade in all their currency pairs? if they do, then it would be straightforward to do a cross correlation between each bitstamp->bitso ODL channel volume, with each bitso currency pair trade volume and then match up channels to currencies to potentially identify what's what. I don't really want to go down this path because I'm hoping that in a year or two, there will be many more ODL channels between dozens of exchanges and I don't want to monitor them all (don't really have time for that). (Note also that bitstamp has also added more currencies such as GBP and so there are many possibilities).

 

8 hours ago, JASCoder said:

Could the change we see below be from some XRP trading to USD now, maybe even into BTC then traded into MXN ? 
( The fee overhead to do that in-house at Bitso is near zero, as heavy trade-fee discounts would be in play )

I don't think ripple are going to make use of BTC on any of the exchanges. We saw Brad in the CNN interview the other day say that they could use other cryptos if liquidity was better but they haven't yet needed to - liquidity is everything and if they are already paying customers to grow xrp ODL channels, then it wouldn't make any sense right now to use another (or competing) token since it would just add liquidity to an alternative.

As for trade in USD at bitso - it would make sense for them to make use of this - my suspicion is that the return path from mxn->usd is the bottleneck and Bitso adding a usd trading option is aiming to grow that traffic organically directly on the exchange - probably, ripple and/or market-makers asked for it to be added and bitso responded. If traffic moves off ledger and goes direct to the exchange then you might be seeing trading that I can't pick up on ledger.

Off ledger trading bothers me, because if that becomes 'the norm' then, there's really no need for xrp at all, they can use BTC or anything else, even just fiat-fiat without a crypto. Lots of these emerging fintech forex trading competitors are essentially doing that.

Link to post
Share on other sites
8 hours ago, peanut56 said:

   This is just an observation and by no means have I gone into ODL to the depths that both the above have seem to have done. Periodical there seems to be testing between new pairs for a week or two that bumps the overall volume from currency that are not daily. This seems to bump the volume of ODL for a week and then it disappears for some time. If I remember right that happened with the Euro one and off for about two months before it was and has been in continuous use for a while now.  From that if we see the Thai Baht/USD along with AUD/PHP ramping up in the next month few months then we may be able to tell which corridors are coming online a few months in advance.

I tend to ignore any channels that appear and disappear - but if you think that some new ODL corridors are being tested, and you want me to plot out the activity, feel free to post wallet/tag addresses either here or in a DM and I'll tweak my scripts to have a look at them. If and when they go live (=permanent), I should see the traffic and my scripts will automatically include them anyway - but it may take weeks/months/years before that happens :(

Link to post
Share on other sites

Not sure how these analytics account for or express the difference in throughput between 1 $10,000 TX (a single ledger close) and 10,000 $1 TXs occurring, clearing via ILP, and Settling a batch via multiple or a single XRPL TX?

When moving from large $10k payments towards Streaming Payments, you would expect to service the same amount of liquidity, using less capital. Because inventory turnover cycles are seconds, not hours/days.

Link to post
Share on other sites

@jbjnr - thanks for the update and the in depth chart analysis! This is incredibly helpful. Great to see ODL volume picking up with smaller value transfers. @JASCoder - appreciate your charts as well!

Wanted to ask if you are both familiar with PlanB's stock to flow valuation model? If so, I was wondering if it would be possible to use a similar methodology and apply it towards XRP? My thinking was in order to more accurately do this for XRP we would need to include a forecast for burned XRP that's then factored into the current stock (i.e existing number of tokens in circulation). We would also need to make assumptions around the future flow (i.e. future supply increases) based on the average number of XRP distributed out of escrow. Based on my understanding of stock to flow, if we can find a way to account for burned XRP and future supply increase, then we should be able to apply the same valuation methodology to XRP. 

Link to post
Share on other sites
21 hours ago, Mpolnet said:

Wanted to ask if you are both familiar with PlanB's stock to flow valuation model? If so, I was wondering if it would be possible to use a similar methodology and apply it towards XRP? My thinking was in order to more accurately do this for XRP we would need to include a forecast for burned XRP that's then factored into the current stock (i.e existing number of tokens in circulation). We would also need to make assumptions around the future flow (i.e. future supply increases) based on the average number of XRP distributed out of escrow. Based on my understanding of stock to flow, if we can find a way to account for burned XRP and future supply increase, then we should be able to apply the same valuation methodology to XRP. 

I googled the PlanB stock to flow model and I'm not terribly impressed by it. Certainly a high creation rate (flow) would imply a lower price, but that's about it.

Since inception of the ledger, about 10 million xrp have been burned - Jed sold almost 3 times that amount this morning in an hour and a half. Fees burned are negligible at the current time, maybe one day in the future when traffic is massive, it might have a noticeable effect.

New tokens entering circulation from founders wallets and everyone who bought at sub cent levels can be modelled to some extent, but don't forget, when Jed sells 30million tokens at 50c every day this week, the people buying them aren't desperate to sell them for less than that - in some sense those sales are providing a price support level. The flaw in the stock flow model is that it looks at supply, but not demand. Demand is highly non-linear and not easy to predict - the xrp price ultimately will be driven by the demand that is being created by all the activities going on like ODL, the flare network one day -  and the speculative traders.

Having said the above - I have looked at the xrp token distribution amount as a parameter in a model I'm working on, but price in my model is determined by ODL utility and is very low currently (hence why I've said a few times on this forum, that when ODL reaches 100million+ daily, then it might have an upwards effect on price).

Link to post
Share on other sites
1 hour ago, jbjnr said:

I googled the PlanB stock to flow model and I'm not terribly impressed by it. Certainly a high creation rate (flow) would imply a lower price, but that's about it.

Since inception of the ledger, about 10 million xrp have been burned - Jed sold almost 3 times that amount this morning in an hour and a half. Fees burned are negligible at the current time, maybe one day in the future when traffic is massive, it might have a noticeable effect.

New tokens entering circulation from founders wallets and everyone who bought at sub cent levels can be modelled to some extent, but don't forget, when Jed sells 30million tokens at 50c every day this week, the people buying them aren't desperate to sell them for less than that - in some sense those sales are providing a price support level. The flaw in the stock flow model is that it looks at supply, but not demand. Demand is highly non-linear and not easy to predict - the xrp price ultimately will be driven by the demand that is being created by all the activities going on like ODL, the flare network one day -  and the speculative traders.

Having said the above - I have looked at the xrp token distribution amount as a parameter in a model I'm working on, but price in my model is determined by ODL utility and is very low currently (hence why I've said a few times on this forum, that when ODL reaches 100million+ daily, then it might have an upwards effect on price).

Thanks for the detailed thoughts on your views of the stock to flow model! You make some valid points - I like your thoughts on Jed's selling pressure creating a potential support line. Would be interested to see your valuation model on XRP based on demand if you're able to share :D

Link to post
Share on other sites
1 hour ago, jbjnr said:

So ODL seems to have fallen off a cliff again. The last week - starting on wednesday, things seem to have taken a nosedive. One of the USD-MXN corridors/channels (orange) seems to have shut down and although one of the others (blue USD-MXN) is carrying on, it hasn't picked up the slack.

The (green) Australian AUD-USD channel also seems to have stopped, as well as the (red) USD-PHP channel.

(I heard a podcast recently, where David Schwartz was lamenting about regulations and saying that despite trying to be compliant, they were still being told they hadn't done whatever was expected by the authorities - this drop is a bit disturbing, I hope it is just a temporary glitch of some kind and not something more serious. Hopefully, they've just changed some destination tags and my data mining isn't picking it up yet. Maybe it's just moved off leger and onto an exchange. Sadly, we just don't know yet ....

image.thumb.png.2c4de9653bb86e71413beb137b734bee.png

same data - but zoomed in a bit more to make the bars more easy to inspect.

image.thumb.png.6397c5990cc95450adf809f8432a0f23.png

Thanks. Please keep us apprised.

Link to post
Share on other sites
55 minutes ago, Archbob said:

It seems it'll take years just to get back to where we were in May of this year.

Probably a bit too early to draw that conclusion definitively. What reasons could there be for the drop?

  • Taken down some corridors due to incoming regulations - very scary (but why leave some USD-MXN traffic up if the other had to go?)
  • Technical changes to the software in preparation for new operations - possible, (but again, why leave some up and running)
  • Liquidity problems - highly unlikely to be caused by Flare airdrop, but maybe market makers have pulled a plug? (Volatility issues over the last week making it too expensive to operate?)
  • End of a trial period with several customers who started at the same time? Also seems unlikely -  @Nicholas (Paging Nicolas Steiger) for a possible explanation, (though I'm not expecting a reply).
  • Technical change to traffic to move it off ledger and less traceable - scary - (don't really need xrp any more for that).
  • a.n.other explanation?
Link to post
Share on other sites

I just posted to the "Bitso thread" a few morning observations - maybe it's time to migrate posting there to this - more ODL in-general - thread ? 

Quote

Bitso's XRP->MXN trade volumes dropped off precipitously two weeks ago.
This past weekend's volume was very low, except for hour 22 on Sunday, or 1m exactly.
This morning, at hour 9 volume resumed an average of 400k per hour.

 

Edited by JASCoder
spelling
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.