Liquidity provider profitability experiment

Some people have suggested that liquidity providers lose money because speculators push them into NuBits when Bitcoin is about to rise and into Bitcoin when Bitcoin is about to drop. I doubt anything like that occurs simply because no one can predict how the Bitcoin price will change.

It is an important question. If speculators can bleed value from the network by taking advantage of liquidity providers, that is a serious issue for the network. If evidence against the notion that liquidity providers are victims of speculators can be uncovered and communicated, then the network can get on with the business of providing cheap liquidity.

I have designed an experiment where I have provided $20,000 in NuBits and Bitcoin. This breaks down to 9953 NBT and 23.820 BTC. The BTC have an initial valuation of $421.79 each, bring the starting net asset value to exactly $20,000. The funds will remain in use at Poloniex by NuBot in connection with NuPool for an extended period. I will watch and report how the net asset value changes. There are three variants of the net asset value that I will report in this thread, perhaps on an approximately weekly basis:

  1. Simple NAV: The USD value in the Poloniex account, where NBT is valued at $1 and Bitcoin is valued at the price used by NuBot.

  2. Bitcoin Price Adjusted NAV: The same as the Simple NAV, but adjusted for changes in the Bitcoin price. If the Bitcoin price were $431.79, then with a starting price of 421.79, $10 times the 23.82 BTC would be subtracted from this NAV value.

  3. Bitcoin Price Adjusted With Pool Payments NAV: This will be the same as the Bitcoin Price Adjusted NAV, but with NuPool payments added in.

I accepted all default NuBot settings except the following:

“txFee”: 0.15,
“bookSellwall”: 50000.0,
“bookBuywall”: 50000.0,
“bookSellOffset”: 0.0035,
“bookBuyOffset”: 0.0035,

This creates a single buy wall and a single sell wall with exactly a 1% spread, not including exchange transaction fees.

While I don’t know how the Simple NAV will unfold, my hypothesis is the Bitcoin Price Adjusted NAV will grow from profits from the 1% spread (which will be reduced to 0.85% by the Poloniex fee). I expect the Bitcoin Price Adjusted With Pool Payments NAV to grow even more.

The experiment started a little less than a day ago, around the beginning of April 11 GMT. I have presented it before any results are available so that the methodology could be reviewed and critiqued, which may result in the experiment design being altered at this early date.

Many thanks to @woolly_sammoth, who provided extensive technical assistance in getting NuBot set up.


This is wonderful, thank you.


Thank you @muchogusto for undertaking this generous experiment. It will benefit Nu both during its running and with the results it provides.


Thank you. It would be interesting to see the result of this.

Here are some early results after only 48 hours:

Simple NAV: 20,175, a 0.87% profit

Bitcoin Price Adjusted NAV: with current price of 426.35, the adjustment is -109, so 20,066, a 0.33% profit

Bitcoin Price Adjusted With Pool Payments NAV: NuPool rewards total 18, so 20,084, a 0.42% profit. Annualised and compounded rate of return is 115%.


cool, can you show out the BTC balance and NBT balance?

Might I ask whether your BTC balance went up in those 48 hours?

are you expecting btc/usd to be stable for the whole month?

I will going forward. Thanks for the request.[quote=“masterOfDisaster, post:8, topic:3770, full:true”]
Might I ask whether your BTC balance went up in those 48 hours?

It did, up to about $15,000 as I remember.

Not sure I understand the meaning behind this question, but I do detect a little sarcasm. Two of the NAVs reported are adjusted for changes in Bitcoin price.

Finally, sorry my NuBot program is down right now. My server rebooted unexpectedly and I can’t seem to get it going again. Have asked woolly_sammoth for help.

I got it up and running again.

1 Like

I think cryptog is asking about how you will normalize out any long term bitcoin trends over the course of the experiment. I think you’ve already done that with your NAVs. I don’t think it was meant to be facetious.

Sorry for being unclear. This is what I wanted to know.
And no, i am not sarcastic.

This experiment seems to be interesting in the mid/long run, I’ll be following this…

Latest results:

NBT Balance: 20,188
BTC Balance: 0.000145

Simple NAV: 20,188, a 0.94% profit

Bitcoin Price Adjusted NAV: with current price of 427.69, the adjustment is -141, so 20,048, a 0.24% profit

Bitcoin Price Adjusted With Pool Payments NAV: NuPool rewards total 72, so 20,120, a 0.60% profit. Annualised and compounded rate of return is 40%.


Results as of 25/04/2016:

NBT Balance: 11,691
BTC Balance: 18.37

Simple NAV: 20,219, a 1.01% profit

Bitcoin Price Adjusted NAV: with current price of 464.22, the adjustment is -1011, so 19,208, a 3.96% loss

Bitcoin Price Adjusted With Pool Payments NAV: NuPool rewards total 124, so 19,332, a 3.34% loss. Annualised and compounded rate of loss is 60%.

Notes: Very little trading has gone on, though trading did bring a small profit. Most profit has come from NuPool compensation. The account was all in NuBits for most of the last week when Bitcoin rose fast. While the simple NAV is up the Bitcoin price adjusted calculations indicate a loss when compared to simply holding NuBits and Bitcoin in equal quantities.


You were unlucky to perform the experiment in a bullish BTC period!

In a bearish period @muchogusto might be as unlucky as now - then he’d sit on BTC in times of dwindling BTC rates…

After all that’s what we expect from this experiment:
a liquidity provider bears the risk of a partial loss, because the funds are used from traders hedging BTC volatility and furthermore takes the risk of a total loss by exchange defaults/thefts.

Let’s see whether this experiment can meet the expectations with regard to volatility.
We can learn from it how high the offset must be to compensate the loss from volatility/hedging.
While this will not be true for all situations, it’s better than the guesstimates we have now.

Thank you, @muchogusto for doing this!


So, the LPC cannot make money unless supported by a grant, justified by a perpetual NBT market demand


If you add the costs from exchange defaults/thefts to the spread, you can operate at break even or make a again if the offset is even higher.