Additional Markets and Pool Weight Adjustments



  • Additional Markets and Pool Weight Adjustments

    Intro

    It has been more than one month since the new Binance Smart Chain markets and pool weights were introduced via TIP-10. This proposal included several NEO pairs, however these have not yet been added due to NEO3 scheduling.

    About me

    I am a validator, trader and liquidity provider, but most importantly, I am a (bot) developer. Immediately after the the launch of demex, I began trading and have never stopped since then.
    Currently I run trading bots on Demex and produce a few million dollars of volume per month and I am in the top 10 of the trading leaderboard. Analysing and optimizing is part of my botting business/hobby.
    The changes, additions and adjustments I propose below will definitly not only benifit me, but also other traders. Which benefits the stakers in the end via increased volume, fees and platform awareness.

    It is important to me that I am transparent about my motives and the proposed changes.

    Missing Markets

    Currently these missing markets are waiting be launched with NEO3.

    • GAS/NNEO
    • NEX/NNEO

    Current Markets

    In the following section, I review the existing Demex markets and provide my view of why the markets are currently performing the way they do, with emphesis on arbitrage possibilities.

    ETH/USDC , SWTH/USDC and SWTH/ETH

    Three of the main markets since launch. The performance of all three has been good. People want to buy SWTH to stake or convert their staking profits into different assets.
    ETH prices have performed well in the last months and has therefore also been a good performer.

    Takin a closer look at these 3 markets, they build an triangular arbitrage oppurtunity. Due to this a huge amount of the past weeks volume was produced by my bot.
    Arbitrage means that if one of the three pairs changes price, the other two must also change. For example if ETH increases in price, the price in the SWTH/ETH has to decrease, otherwise the price for USDC/SWTH will not match the value in ETH.
    Such opportunity occurs all the time since the markets flucturates a lot.

    ETH/USDC , WBTC/USDC and ETH/WBTC

    These three markets create another triangular arbitrage opportunity, however the performance is not as good as USDC/ETH/SWTH. This is because BTC and ETH behave similarly, most commonly moving in the same direction, however sometimes one does perform better/worse.
    One reason these markets perform quite well is because the WBTC/USDC markets offers opportunities for cross exchange arbitrage. Deposits and withdrawals of WBTC, as for all ERC-20 tokens, are currently really expensive.

    BUSD/USDC

    Probobly one of the most important markets (for me) on Demex. This market is a bridge between all USDC and BUSD pairs and creates a lot of new triangular opportunites and generates a lot of demex volume.

    The best example is USDC/BUSD + BUSD/SWTH + USDC/SWTH. At the time of creating the BUSD markets switcheo had a listing event with Pancake Swap. This caused a huge price increase. Best thing was that deposits and withdrawals of BUSD and SWTH(BSC) are really really cheap and fast.
    Because of this triangular arbitrage the USDC/BUSD market crashed from 1 / 1 to 1 / 0.88, this means with 0.88 USDC it was possible to buy one full BUSD. But this is just one reason this market crashed. The second one is the AMM protocol from DEMEX.
    This protocol defines how the pools are transformed to orderbooks. We already saw adjustments on the protocol. But one main issue is that algorithm is not designed to work with 1 to 1 pools like BUSD/USDC or BTCB/WBTC. The large maker orders of these market are far from the match point, which require large slippage (for a stable coint), therefore they are not used.
    The main reason for this behaviour is to protect liquidity providers. In my opinion BUSD/USDC and BTCB/WBTC are every safe markets and therfore should use more of the pool volume at closer to the match point.

    NNEO/USDC NNEO/BUSD and CEL/USDC CEL/BUSD

    With the release of the new markets the pool weight of NNEO/USDC and CEL/USDC was removed. These pools are now almost empty and provide really bad liquidity. With the new BUSD/USDC market there is now triangular arbitrage opportunites which cannot be used due to this lack of liquidity.
    In my opinion these markets need new pool weights to restore some volume.

    NNEO/ETH

    A good market creates a triangular arbitrage opportunity, but with low liquidity NNEO/USDC limits volume.

    BTCB/WBTC

    Honestly this pair just has liquidity because of the pool weight and is a safe haven. Currently there is no good (triangular or cross exchange) arbitrage possible. We need more pairs with BTCB like BTCB/BUSD which would provide new triangular arbitrage opportunities with WBTC/USDC and BUSD/USDC.

    NEX/USDC

    Not bad at all, but only NEX/USDC is available at the moment. It is expensive to do arbitrage in my opinion.

    BNB/ETH

    The forgotten child. No liqudity, no pool weight and no arbitrage opportunity.

    Current Triangular Arbitrage Opportunities

    In summary, here are the existing on-exchange arbitrage opportunities

    BUSD1

    • BUSD1-CEL1-USDC1-BUSD1
    • BUSD1-NNEO2-USDC1-BUSD1
    • BUSD1-SWTH-USDC1-BUSD1
    • BUSD1-USDC1-CEL1-BUSD1
    • BUSD1-USDC1-NNEO2-BUSD1
    • BUSD1-USDC1-SWTH-BUSD1

    CEL1

    • CEL1-BUSD1-USDC1-CEL1
    • CEL1-USDC1-BUSD1-CEL1

    BNB1

    None

    BTCB

    None

    ETH1

    • ETH1-NNEO2-USDC1-ETH1
    • ETH1-SWTH-USDC1-ETH1
    • ETH1-USDC1-NNEO2-ETH1
    • ETH1-USDC1-SWTH-ETH1
    • ETH1-USDC1-WBTC1-ETH1
    • ETH1-WBTC1-USDC1-ETH1

    NNEO2

    • NNEO2-BUSD1-USDC1-NNEO2
    • NNEO2-ETH1-USDC1-NNEO2
    • NNEO2-USDC1-BUSD1-NNEO2
    • NNEO2-USDC1-ETH1-NNEO2

    SWTH

    • SWTH-BUSD1-USDC1-SWTH
    • SWTH-ETH1-USDC1-SWTH
    • SWTH-USDC1-BUSD1-SWTH
    • SWTH-USDC1-ETH1-SWTH

    USDC1

    • USDC1-BUSD1-CEL1-USDC1
    • USDC1-BUSD1-NNEO2-USDC1
    • USDC1-BUSD1-SWTH-USDC1
    • USDC1-CEL1-BUSD1-USDC1
    • USDC1-ETH1-NNEO2-USDC1
    • USDC1-ETH1-SWTH-USDC1
    • USDC1-ETH1-WBTC1-USDC1
    • USDC1-NNEO2-BUSD1-USDC1
    • USDC1-NNEO2-ETH1-USDC1
    • USDC1-SWTH-BUSD1-USDC1
    • USDC1-SWTH-ETH1-USDC1
    • USDC1-WBTC1-ETH1-USDC1

    WBTC1

    • WBTC1-ETH1-USDC1-WBTC1
    • WBTC1-USDC1-ETH1-WBTC1

    New Markets

    The following list represents pairs which could fill a gap and provide new opportunities.

    CEL/ETH

    WTF why is this pair not on demex yet?

    SWTH/NEO

    Main market from switcheo exchange, I miss this market );

    SWTH/BTCB

    First BTC pair with SWTH. I suggest BTCB because it is cheaper and allows directly arbitrage with Pancake Swap.

    BTCB/BUSD

    Equivalent to WBTC/USDC a new BSC pair with cheap deposit and withdrawal. This could help arbitraging the ERC pairs.

    BNB/BUSD

    This market could bring good volume too since Binance burns BNB on a regular basis.

    Proposal

    I propose to list the following markets: CEL/ETH and BNB/BUSD. Regarding SWTH/BTC, BTCB/BUSD and SWTH/NEO I would like to hear opinions.

    These new pool weights I suggest(if all markets get listed):

    Market Cur. Pool weight New Pool weight Liquidity Monthly Volume* **
    SWTH/ETH 4 - 20% 15 - 15% 18 - 18% $5.45m $9.73m
    WBTC/BTCB 1 - 5% 2 - 2% 1 - 1% $2.70m $217,800
    SWTH/USDC 4 - 20% 15 - 15% 18 - 18% $2.67m $9.18m
    WBTC/USDC 2 - 10% 7 - 7 % $2.10m $2.44m
    ETH/WBTC 1 - 5% 5 - 5% 4 - 4% $2.10m $2.2m
    USDC/BUSD 1 - 5% 3 - 3% $1.99m $4.4m
    ETH/USDC 2 - 10% 8 - 8% $1.65m $11.25m
    ETH/nNEO 1 - 5% 4 - 4% $1.22m $4.53m
    nNEO/BUSD 1 - 5% 4 - 4% $839,129 $7.11m
    SWTH/BUSD 2 - 10% 8 - 8% $623,983 $8.55m
    CEL/BUSD 1 - 5% 3 - 3% $552,435 $522,147
    nNEO/USDC 0 - 0% 4 - 4% 2 - 2% $59,160 $1.88m
    NEX/USDC 0 - 0% 0 - 0% $10,886 $13,892
    CEL/USDC 0 - 0% 3 - 3% 2 - 2% $4375 $43,045
    ETH/BNB 0 - 0% 8 - 8% 6 - 6% $0 $10,966
    BNB/BUSD - 8 - 8% 6 - 6% - -
    ETH/CEL - 3 - 3% 2 - 2% - -
    SWTH/BTCB - 3 - 3% - -
    BTCB/BUSD - 4 - 4%
    SWTH/nNEO - 3 - 3% - -

    *Liquidity taken from Demex - 04-27-2021
    ** Trading volume taken from swth.info - 04-27-2021. Volumes converted at current market prices. See below

    Coin USD
    CEL $6.15
    ETH $2,570
    BTC $55,000
    NEX $1.64
    NEO $93
    SWTH $0.0672
    BNB $570

    I tried to not make the changes too drastic, like removing all pool weight. I think the bigger pools are quite good and can share a bit of their weight. Some could now say that there are up to 25% decrease, yes but I used this decrease to boost mainly new SWTH markets or BSC markets. I also used 100 pool weights to adjust the weight at a 1% precision.

    As you may have noticed, I have not proposed new tokens, because I think we can still improve the liquidity a lot by improving the conditions.

    I hope for a civilized discussion about my proposed changes, I am open for feedback and changes as well.

    Devel

    • EDIT 1:
      -- Fixed wrong Market name.
      -- Removed 24h volume
      -- Added monthly volume converted at current market price.
      -- Added table with current market prices.
    • EDIT 2:
      -- Removed SWTH/nNEO and SWTH/BTCB(for now)
      -- Changed pool weights


  • Sounds great to me.



  • First of all, thanks for your contribution! You really put some effort into this!

    I really like this proposal. Expanding to more pairs will drive towards faster adoption. Furthermore, i think these changes do not shift the rewards too much and will most likely be received with open arms.

    One important thing is the stargate update. I think the best thing for the team to do is to make an estimation about how fast this upgrade will come to demex. If they see an implementation if the upcomming future (2-6 months) then it might be benificial to wait with this upgrade and do a complete overhaul on the system as a hole.



  • Thank your for your detailed proposal. I think implementing the whole proposal would be very good for Demex. If there are big concerns of having too thin liquidity when introducing too many markets, I would be happy, with a partial implementation as well.



  • We need SWTH/nNEO and Gas/nNEO back.. Lets get them back.



  • Great proposal, looking very good, with a clear (customer) reason and good argumentation. An extra thing to look at: having a GAS/Stable and a GAS/NEO pair could bring an extra arbitrage opportunity, while serving people with the NEO/GAS pair. We could all look for extra arbitrage opportunities. Im thinking about CEL and AVA as interesting choices.



  • What is the reason to list nGAS/nNEO and NEX/nNEO after NEO3 launch? In fact, should list these pairs immediately after Switcheo exchange is retired on 30th April unless of course custom NEX contract and nGAS wrapper is not ready. Other than that, no reason to wait for these pairs listing on Demex. It would take Flamingo or just one other exchange to take away NEX (nep5) volume if Demex keeps waiting till NEO3. So if the team is serious about migrating NEX/NEO Switcheo exchange volume to Demex, list this pair now instead of timing it with NEO3 for whatever logic other than it is not ready or technical issues. Otherwise forget NEX altogether on Demex. Not the end of the world.



  • SWTH/nNEO on Demex is not required at the moment as Flamingo is already there for that with good liquidity. Switcheo exchange users who like to trade SWTH/NEO can trade there. As for liquidity provider, why will someone lock SWTH/nNEO on Demex unless you can beat the Flamingo APY of 48.98% for this pair.



  • Reducing weight % for SWTH/ETH, SWTH/USDC, SWTH/BUSD needs to be well thought and proceeded with caution. There is Pancakeswap already offering a better APR than Demex for SWTH/BNB. Not even talking about APR % putting that lp token in Beefy, Autofarm, Swamp.

    You don't want a situation where above 3 pairs APR goes low that people start removing liquidity as soon as their 1 month lock is over. Especially it is an insult to SWTH/ETH liquidity providers which is the highest at $5.7m liquidity and only 21.39% APR as of this writing. For comparison, just staking SWTH APR on TradeHub is 34%.

    If it ain't broke, don't fix it. Complacency should not set in thinking that SWTH/ETH and SWTH/USDC liquidity is good right now so tinker with lowering weight %. In fact, SWTH/BUSD liquidity reduced by about 800k-1m USD recently and hence that pool APR increased providing an opportunity.



  • Yes BTCB/WBTC is a safe haven. The only use of BTCB at the moment on Demex is for users to onboard cheap BTC. Deposit BTCB with minimal network fee and sell to WBTC to invest in WBTC liquidity pools or trade WBTC pairs. But that purpose is also somehow reduced as big quantities cannot be traded closer to 1:1 as the algo does not consider stablecoin pools (BTCB/WBTC and USDC/BUSD) as unique and just treats it as any other A and B token and hence the orders placed in the order book are according to the formula instead of closer as possible to 1:1 with higher liquidity/quantity.

    So to make more use of the BTCB/WBTC liquidity pool, more BTCB pairs listing is required. Otherwise abandon this pair altogether. Abandoning makes more sense.



  • I don't see BNB/BUSD bringing any volume. Pancakeswap is already there for that. In fact for any bep20/bep20 pair listing on Demex, it is better to integrate pcs order books into Demex rather than building our own bsc pools and launching new pairs. Users will just swap on pcs or the second or third of BSC dexes instead of using Demex. As for liquidity providers for such pairs, if you don't beat or come closer to providing APR % of similar pairs on other dexes, they are not going to come to you.



  • Regarding CEL/ETH, there is no guarantee of success here. If CEL/BUSD did not work, very less chance of CEL/ETH bringing volumes unless you can offer more than 150% APR minimum on this pair. With CEL/BUSD, one side is bep20 so that removes high fees issue of deposit and withdrawal. With CEL/ETH both sides are erc20 and there is a high fees issue of deposit and withdrawal because of Ethereum network.

    Let me tell you why CEL/ETH worked and did good volumes on Switcheo exchange:

    -ETH was mostly between $200 and $400 that time. Also went lower to 100's range for months. https://medium.com/switcheo/switcheo-lists-celsius-cel-b28c2856820c
    -ETH gas network used to be less than 20 gwei most of the time. Even during odd times it momentarily jumped, ETH price was low to justify the odd high network fee.
    -Switcheo had an option to charge users withdrawal fee in ETH instead of only CEL.
    -There was no deposit fees on Switcheo.
    -Mainly, Switcheo and Idex were the only exchanges with CEL listed. Yes, Uniswap did not have CEL listed.

    Now why CEL/ETH won't work on Demex:

    -ETH price and network fee is not the same now as it used to be during Switcheo days. It has become expensive to deposit/withdraw any erc-20 token on Demex because of high ETH price and network fee (you can't expect a user to stay inside L2 Demex. a user deposits/withdraws a token multiple times to/from exchange for various reasons).
    -Try explaining the average Joe that he will pay $30-60 for each deposit and withdrawal from Demex to personal wallet depending upon Ethereum conditions.
    -You might say but Demex trading is a few dollars cheaper than Uniswap, yes it is. But Uniswap users save the hassle of depositing and withdrawing from the exchange every time. Just swap and boom, right into MetaMask.
    -Now there is Uniswap, Quickswap (dexes), FTX, Liquid (cexes) doing major volume for CEL.



  • As for your proposal, can experiment with listing CEL/ETH (so the fascination is over and myth is proved that CEL/ETH did wonders for Switcheo exchange and will repeat the same performance on Demex) by reducing pool weight little from pairs other than 3 SWTH pairs. Reducing from BTCB/WBTC makes the most sense.

    BNB/BUSD is not going to bring any volume. Binance, pcs and other bsc dexes is already for this.

    *It would have been nice if you would have mentioned that if only CEL/ETH and BNB/BUSD is listed, what weights to give them and reduce from what pairs. But you have posted a complete overhaul of all LP in the process.



  • @nwo said in Additional Markets and Pool Weight Adjustments:

    What is the reason to list nGAS/nNEO and NEX/nNEO after NEO3 launch? In fact, should list these pairs immediately after Switcheo exchange is retired on 30th April unless of course custom NEX contract and nGAS wrapper is not ready. Other than that, no reason to wait for these pairs listing on Demex. It would take Flamingo or just one other exchange to take away NEX (nep5) volume if Demex keeps waiting till NEO3. So if the team is serious about migrating NEX/NEO Switcheo exchange volume to Demex, list this pair now instead of timing it with NEO3 for whatever logic other than it is not ready or technical issues. Otherwise forget NEX altogether on Demex. Not the end of the world.

    Currently we have nNEO a wrapped NEO from Flamingo. Problem is if we get native NEO this token is not equal to nNEO. Technically these are two different tokens. If we list GAS/nNEO and NEX/nNEO but want to switch to real NEO the pair needs to be removed and replaced with the correct one.

    @nwo said in Additional Markets and Pool Weight Adjustments:

    SWTH/nNEO on Demex is not required at the moment as Flamingo is already there for that with good liquidity. Switcheo exchange users who like to trade SWTH/NEO can trade there. As for liquidity provider, why will someone lock SWTH/nNEO on Demex unless you can beat the Flamingo APY of 48.98% for this pair.

    I saw this argument from you many times. If you continue this why not sunset Demex, other exchanges offer better ARPs and more Liquidity. SWTH/NEO is the main pair form Switcheo why shouldn't we list it?

    @nwo said in Additional Markets and Pool Weight Adjustments:

    Reducing weight % for SWTH/ETH, SWTH/USDC, SWTH/BUSD needs to be well thought and proceeded with caution. There is Pancakeswap already offering a better APR than Demex for SWTH/BNB. Not even talking about APR % putting that lp token in Beefy, Autofarm, Swamp.
    You don't want a situation where above 3 pairs APR goes low that people start removing liquidity as soon as their 1 month lock is over. Especially it is an insult to SWTH/ETH liquidity providers which is the highest at $5.7m liquidity and only 21.39% APR as of this writing. For comparison, just staking SWTH APR on TradeHub is 34%.
    If it ain't broke, don't fix it. Complacency should not set in thinking that SWTH/ETH and SWTH/USDC liquidity is good right now so tinker with lowering weight %. In fact, SWTH/BUSD liquidity reduced by about 800k-1m USD recently and hence that pool APR increased providing an opportunity.

    I think I will rework the pool weight so the change is not to high. 20% -> 17/18%. The missing Liquidity form NEO/BUSD can already be seen in daily volume. From $100k to $30-40k. Speaking of ARP this increased, and soon we should see entering liquidity again.

    @nwo said in Additional Markets and Pool Weight Adjustments:

    I don't see BNB/BUSD bringing any volume. Pancakeswap is already there for that. In fact for any bep20/bep20 pair listing on Demex, it is better to integrate pcs order books into Demex rather than building our own bsc pools and launching new pairs. Users will just swap on pcs or the second or third of BSC dexes instead of using Demex. As for liquidity providers for such pairs, if you don't beat or come closer to providing APR % of similar pairs on other dexes, they are not going to come to you.

    Definitly there will be volume if we add pool weight. Even if there is not that much liquidity locked we still do volume. We dont need $10m liquidity pools to generate volume. $1m pool is enough to generate $10m volume per month. See NEO/BUSD.

    @nwo said in Additional Markets and Pool Weight Adjustments:

    As for your proposal, can experiment with listing CEL/ETH (so the fascination is over and myth is proved that CEL/ETH did wonders for Switcheo exchange and will repeat the same performance on Demex) by reducing pool weight little from pairs other than 3 SWTH pairs. Reducing from BTCB/WBTC makes the most sense.

    The team decided to make a proposal for CEL/ETH so it is listed anyway. Next is creating a pool(without pool weight) and then we can see if liquidity comes. Maybe I will provide a few $ to bootstrap the pool.

    @nwo said in Additional Markets and Pool Weight Adjustments:

    *It would have been nice if you would have mentioned that if only CEL/ETH and BNB/BUSD is listed, what weights to give them and reduce from what pairs. But you have posted a complete overhaul of all LP in the process.

    Because my proposal targets to list all pairs I propose and rebalance all pool weights a bit. In my opinion a pool weight of 1% could bring atleast a minimum off liquidity to a pool to start trading. If the pool does not do enough volume with 1-2% pool weight it is still possible to remove the pool weight.

    Last but not least I want to thank you for the dicussion(so far).

    I will look at the CEL/ETH market and pool and will comeback to adjust the pool weights.

    As for WBTC/BTCB and USDC/BUSD I think I will drop the pool weight to 2%. WBTC/BTCB has way to much locked liquidity but generates to less volume. USDC/BUSD is very important but I saw already a lot of people placing limit orders -> 50% of the orders are from users.
    users: http://54.255.5.46:5001/get_orders?market=busd1_usdc1&initiator=user&order_status=open
    amm: http://54.255.5.46:5001/get_orders?market=busd1_usdc1&initiator=amm&order_status=open



  • I updated the pool weights and removed SWTH/nNEO(lets wait for NEO3) and SWTH/BTCB(too much pairs at once).

    Accoding to the feedback I increased the pool weight for SWTH/ETH and SWTH/USDC to 18%. I lowered WBTC/BTCB to 1%, this pool is currently "abused" for free money.



  • @Devel484 said in Additional Markets and Pool Weight Adjustments:

    I updated the pool weights and removed SWTH/nNEO(lets wait for NEO3) and SWTH/BTCB(too much pairs at once).

    Accoding to the feedback I increased the pool weight for SWTH/ETH and SWTH/USDC to 18%. I lowered WBTC/BTCB to 1%, this pool is currently "abused" for free money.

    "I saw this argument from you many times. If you continue this why not sunset Demex, other exchanges offer better ARPs and more Liquidity. SWTH/NEO is the main pair form Switcheo why shouldn't we list it?"

    Good joke. Curious to know where and how many times you saw this argument from me.

    And you went from telling me that how SWTH/nNEO was needed on Demex to lets wait for NEO3, Haha. And for SWTH/BTCB - too much pairs at once. But while reducing weights of 3 SWTH pairs and other too much pairs at once (10 plus) no issue eh?

    List SWTH/nNEO, SWTH/BTCB. Even better add SWTH/BNB and SWTH/WBTC also.

    Will bring millions of volumes like the recently listed pairs of BTCB/BUSD, BNB/BUSD, CEL/ETH.



  • A bit late to the party, I would suggest that changes be incremental rather than a sweeping change.

    A sweeping change proposal could potentially backfire as:

    1. This is complex and affects a lot of people. Holders might not fully understand the impact until it hits.
    2. A sweeping change will also cause a lot of confusion about the old/new apys for various pools.
    3. Holders who are locked have no ability to unlock if it negatively impacts them.

    Therefore I suggest that the approach should be about proposals that change one thing at a time until we reach equilibrium or have nothing left to change.

    It's similar to the hill-climbing algorithm where we can reach a locally optimum point by simply making incremental decisions that are beneficial to us.


Log in to reply