CMC Liquidity metric

[davidgdg]

So CMC has finally introduced liquidity as an exchange metric (better late than never). Liquidity has the small advantage over volume that it can't be faked.

The numbers are quite extraordinary and just go to show how spectacularly dishonest most self-reported exchange volumes are.  Basically what you want in an exchange is the lowest possible ratio of volume to liquidity (fat book, small spread). So let's see what the numbers now show.

Well lets start with today's number 1 exchange by volumep: TAGZ with a volume today of 4.3Bn USD. Well that's pretty amazing. Let's all go trade there. Except take a look at liquidity: 7m USD (ranked 29th). So that's a ratio of 614:1

Now let's look at the world number 65 by reported volume: the evil stealing scammers known as Bitfinex with a paltry volume of 147m. Except that their liquidity is 78m (2nd in the world).  So their ratio is  2:1 which seems to be by far the lowest in the world.

Go figure ….

[exstasie]

So CMC has finally introduced liquidity as an exchange metric (better late than never). Liquidity has the small advantage over volume that it can't be faked.

The numbers are quite extraordinary and just go to show how spectacularly dishonest most self-reported exchange volumes are.  Basically what you want in an exchange is the lowest possible ratio of volume to liquidity (fat book, small spread). So let's see what the numbers now show.

Well lets start with today's number 1 exchange by volumep: TAGZ with a volume today of 4.3Bn USD. Well that's pretty amazing. Let's all go trade there. Except take a look at liquidity: 7m USD (ranked 29th). So that's a ratio of 614:1

Now let's look at the world number 65 by reported volume: the evil stealing scammers known as Bitfinex with a paltry volume of 147m. Except that their liquidity is 78m (2nd in the world).  So their ratio is  2:1 which seems to be by far the lowest in the world.

Go figure …

Liquidity metrics can be gamed by spoofing and other algorithms.

Their ranking by liquidity is an improvement, but probably only a slight one. Does anyone in the room really believe HitBTC is the most liquid exchange in the world, ahead of Bitfinex, Binance, Kraken, and Coinbase?

[bhadz]

So CMC has finally introduced liquidity as an exchange metric (better late than never). Liquidity has the small advantage over volume that it can't be faked.

They can manipulate it, it's CMC and they have managed to manipulate data and even make an unknown exchange at the top.

Does anyone in the room really believe HitBTC is the most liquid exchange in the world, ahead of Bitfinex, Binance, Kraken, and Coinbase?

Hell yeah, I'm expecting Binance on that spot.

[davidgdg]

So CMC has finally introduced liquidity as an exchange metric (better late than never). Liquidity has the small advantage over volume that it can't be faked.

The numbers are quite extraordinary and just go to show how spectacularly dishonest most self-reported exchange volumes are.  Basically what you want in an exchange is the lowest possible ratio of volume to liquidity (fat book, small spread). So let's see what the numbers now show.

Well lets start with today's number 1 exchange by volumep: TAGZ with a volume today of 4.3Bn USD. Well that's pretty amazing. Let's all go trade there. Except take a look at liquidity: 7m USD (ranked 29th). So that's a ratio of 614:1

Now let's look at the world number 65 by reported volume: the evil stealing scammers known as Bitfinex with a paltry volume of 147m. Except that their liquidity is 78m (2nd in the world).  So their ratio is  2:1 which seems to be by far the lowest in the world.

Go figure …

Liquidity metrics can be gamed by spoofing and other algorithms.

Their ranking by liquidity is an improvement, but probably only a slight one. Does anyone in the room really believe HitBTC is the most liquid exchange in the world, ahead of Bitfinex, Binance, Kraken, and Coinbase?

Curious as to how liquidity could be faked? Would seem to be a very risky thing to do

[exstasie]

Liquidity metrics can be gamed by spoofing and other algorithms.

Curious as to how liquidity could be faked? Would seem to be a very risky thing to do

It's not risky when it's done by the exchanges themselves. It's not like we're talking about traders flashing real orders.

In most cases, these orders exist in the exchange API (which CMC is pulling) and visual order book, but aside from that, they are "fake" in every way. Traders can't actually buy/sell into them.

This practice is widespread.

[joniboini]

It looks like an alternative to Coingecko trust score but still looks unreliable atm. Especially with HitBTC on top of the list.

It would be better if they use unique visits, time spent on each page, etc as a metric to 'rank' the exchange. It might be manipulated too but it will cost them a lot more than feeding fake API data.

[RapTarX]

It would be better if they use unique visits, time spent on each page, etc as a metric to 'rank' the exchange. It might be manipulated too but it will cost them a lot more than feeding fake API data.

That's too much of work for a site like CMC. I guess a dedicated website only for monitorong exchanges would be good. If CMC so much data, that would be looking scrawled.
I didn't check this feature of CMC yet but with coingecko, this feature is a good one.

[sheenshane]

That was very sad to them, I used this site for a long time ago since I knew crypto and I start trading. I used this as one of my references.
It seems they are not trustworthy enough because of the fake data analysis.

But according to the chief strategy officer of CMC that was published and explained on the Cointelegraph.

“We believe our adaptive methodology will make our metric very difficult to ‘game’ as orders would need to be placed close to the mid-price, or risk being counter-productive to the Liquidity metric scoring.”
source:

Probably there's an issue of scoring volume and might be fixed very soon or just a manipulated one and the statement is one of their excuses.

[exstasie]

It looks like an alternative to Coingecko trust score but still looks unreliable atm. Especially with HitBTC on top of the list.

It would be better if they use unique visits, time spent on each page, etc as a metric to 'rank' the exchange. It might be manipulated too but it will cost them a lot more than feeding fake API data.

What CMC should do is pay a consultant like the Blockchain Transparency Institute to grade exchanges, and blacklist the most egregious offenders. Or if they wanted to do it cheaply, they could just integrate BTI's verification program into their ranking methodology.
https://www.bti.live/methodology/