Carryover: There is ~230 NBT remaining from the last grant
Grant Amount: 4000 x 5.4% + 4000 x 7.2% = 216 + 288 - 230 = 274 NBT
NBTPool Fee: 8000 x 1% = 80 NBT
Total Grant Requested: 354 NBT
=##=##=##=##=##=## Custodian Hash ends with this line ##=##=##=##=##=##=
Verify. Use everything between and including the <custodianhash></custodianhash> tags.
Changes from Month 1
Month 1 saw more activity on the NBT/BTC pair, so I wanted to increase the limits. However, I feel that we should not increase the BTC pair too much based upon the goals of focusing on fiat.
That said, there were some “issues” with the USD pair. There is activity and the USD side is full, but the NBT said is always empty. It appears that whenever NBT are available, they get bought immediately. It is tough to re-balance on Cryptsy because their USD/BTC conversion rate is always at a major premium, so traders take a loss trying to re-balance. I would welcome any suggestions from the community on how to make this pair stronger.
To address this and to “comply” with the current trend of pools, I have reduced the rates - 0.24% BTC and 0.18% USD. BUT I propose an increase in the tolerance on the USD pair to 3%.
Typo where you said both pools are nbt/USD. As far as rebalancing, instead of dropping your nbt in the bot, put it up at a big spread like 5% and see if it still gets bought. If it does, I bet you could make serious dough arbitraging.
NBT is being used as an exist of local illiquid USD on Cryptsy. If you can withdraw USD on Cryptsy at a better cost and use it to buy NBT either directly on NBT/USD on ccedk or via btc, then NBT is fulfilling its function as a fiat substitute on exchanges. I support Nagalim’s suggestion to set a bigger spread in this effort, as a ALP user or as an individual trader.
I think you shouldn’t increase the limit on NBT/BTC pair that much. I think 1500 on both sides is enough, total 3,000 NBT. Let LPs compete. The current trading on Cryptsy is very low and I don’t see a need to provide more liquidity there. The NBT/USD pair is valuable in keeping the peg, so I’m fine with that.
Ok and given the discussion about lowering liquidity costs. Did you notice that both NuPool and LiquidBits lowered their rates for LPs on all pairs? May I suggest you go with that too, not sure why Cryptsy should have higher rates than e.g. Poloniex, but maybe you do have a good rational for that?
I can see lowering to 0.25% for BTC.
I would lower the USD to 0.18% BUT raise the tolerance to 3% simply due to the cost of doing business on the fiat pairs on cryptsy.
Updated the main post. Time to vote.
The ONLY reason this is for 30 days is because I want to change to @Nagalim’s fixed rate pool. And I’m hoping one month will give enough time to shake out the bugs.
A 3% tolerance on that pair is basically a 6% SAF. It’s an interesting perspective to look at SAF as a means of offsetting rebalance costs, as it was originally meant for volatility costs. I’d question the effectiveness of such a method as only one trade might happen between rebalancing, whereas volatility can cause many trades to happen at a spread, thereby amplifying the effect of SAF.
I ask you to consider the end user. Would you want to trade nbt on cryptsy if you knew you’d have to pay 3% to sell for USD, then another big % to withdraw?
If you want to offset the 0.5% nbt withdrawal fee with a ~0.7% tolerance, I’d support that.
Sorry I didn’t catch it before you hashed it. 3% is a lot, you’d be swimming in uncharted territory. Not that that’s a bad thing, necessarily…
USD withdrawal is a different beast on Cryptsy. It’s not the simple 0.5%…I can speak from experience. Once you have USD you’re stuck unless you do one of 3 things.
Change to BTC, which carries a 3%+ cost due to the illiquid premium market
Withdraw to another e-wallet that you have to setup which is more hassle and a 1% fee. Not to mention the wire fees after that.
Withdraw to a physical check which takes 3-5 business days to get to you
SO, USD on cryptsy is a hassle. Passing the 3% to the end user saves Nu network from the cost. Plus 3% is comparable to the BTC/USD premium that they are already paying. I’d be willing to lower it to 2% though, but the motion is out???
But both the LP and the end user have to pay the USD withdrawal, and that is not really unique to Cryptsy. We are trying to provide a service such that users with USD don’t have to pay that very fee, but can instead trade for USD and withdraw that. The LPs are paid by Nu enough to offset the withdrawal fee if they are trying to rebalance. Maybe not, like I said it’s an interesting perspective and I’m not completely convinced it’s wrong, I just question the efficiency of it. You can try to pass this grant as is for a test run if you like. Besides, maximizing the offset is not necessarily the best LP strategy and you may find your LPs using offsets significantly less than the allowed tolerance. It’s an interesting experiment, for sure.
an experiment indeed. the usd in cryptsy has more difficulties since you have to be authenticated with a picture and official documents. Cryptsy lacks the different authentication levels that other exchanges have!
p.s. i see a “big” rise in BTC’s value but NU’s buy side is not shrinking, i wonder!
I can tell you there are only 2 providing USD liquidity - and I am one. And there is no NBT. I’m not compensated by Nu enough to re-balance as it stands, which is why everything is setting in USD. It would take ~15 days to make 3% to cover the cost of making the USD -> BTC -> NBT re-balance. Or wait the 5-7 days for a check and then use coinbase BTC -> NBT.
I see, basically every alt-coin on cryptsy is trading for more than the coinmarketcap average. Maybe you should consider modifying the price feeds to have a local offset, such that NBT is worth $1.01 instead of $1, then have a smaller tolerance. This is equivalent to an asymmetric tolerance. It would require a change in both the client and the server and would result in something like a buy wall at $1 and a sell wall at $1.02. This would not require any change to the BTC pool, only the USD pool.
The fundamental issue here is that if NBT had existed when cryptsy started operation it wouldn’t have gotten this bad. But as is, you’re basically attempting to provide an arbitrage gateway for the entire exchange.
I can hard code some changes to the client, but it would break the other pools unless I do some labor intensive wiring out of parameters. I can easily make you a branch of my github specifically for NuRiver NBT/USD though.
I suggest we hold a bit supporting liquidity operations on cryptsy given their unclear situation.
Plus we have already a good liquidity scattered over several exchanges, I am feeling:{“blocks”:561218,“buy”:113501.0792,“sell”:119643.7158}
EDIT: Well the more liquidity the better although we should limit the costs.
Supporting this as it is a relatively small grant and still partially experimental given the issues we have seen on Cryptsy.
I think it is good to have good, but small liquidity spread over many exchanges over having 1 or 2 major ones.