Pontem x Ditto Livestream Recap

Today we have the Ditto team with us  for our Waterside Wednesday live stream series.

Alejo: Welcome and thanks for joining us today! Go ahead and tell us a little about yourselves and what you’re building at Ditto.

Ditto: I’m one of the co-founders of Ditto, and I have a trading and MEV background. I then got some experience in EVM chains. The rest of my team has a lot of experience on Solana. Right now, we see the opportunity to do liquid staking better than it’s ever been done on other chains. I’d love to share with you and the community the vision we have for Ditto and the plan we have to get there.      

Alejo: I’m excited to hear about the nuances of staking on Aptos and also the vision for how Ditto wants to approach it.

Ditto: Let me first start with explaining what Ditto is. I want to answer some of the questions from the community because there’s a little bit of confusion. I noticed one question asking how Ditto and Pontem will be collaborating because they’re similar, but we’re actually completely different protocols.

We are a liquid staking product. At its core, what a protocol like Ditto does is make it so users can deposit Aptos with us, and we will stake their Aptos with validators who will produce Aptos blocks. For doing that, there is inflation that is currently set to 7% annually. This yield is a little bit different from most other sources of yield on the chain. Some will come from fees,  token commissions, but it’s very important to understand that the validator yields that you capture as an Aptos depositor in Ditto, those come from the chain itself and they come in the form of Aptos. That’s what our core product is. It’s called stAPT, so Staked Aptos.

Alejo: As I understand it, the minimum yield you’ll get from the Staked APTs is the validator rewards that are natively being given out by Aptos. But then you mentioned there might be some added fees potentially, I would love to hear more about that and how this solution differentiates from others.  

Ditto: This might be a good time to go through how the solution works. So the way stAPT works is 1 stAPT starts off being worth 1 APT and over time, it becomes worth more. Another question I saw that I can preemptively answer is: how do we ensure that it keeps its peg? On Aptos, you can go back from stAPT to APT without losing any APT.

You will, however, have to wait. So what that means is for users who hold stAPT, there are 3 ways to go back to APT. One way is using a DEX, like Liquidswap, and that answers the question of how we’re working together. We have a stable pool on Liquidswap right now where you can swap from APT to stAPT and  swap back. But in terms of unstaking, unstaking does exist on the Aptos chain. and tTe way that it works on the chain level is it’s a 30 day subscription period. It’s not a fixed unstake time. Say I stake with a validator on Day One and on Day 25 I decide I want to unstake. On Day 30, when my subscription ends, I get all my tokens back. So in this case, I only had to wait 5 days. Unstaking takes, at most, 30 days. Now let’s say we have 4 validators staggered weekly, now unstaking will take at most one week. What that means for our design is we have a delayed unstake and an immediate unstake. The delayed unstake option will be the one where users click unstake and they wait at most up to 30 days, but usually less, and with no fees you get your APT back. And the immediate unstake, which is currently live on our website, that’s the one where, of all the Aptos that we own, we actually keep a small percentage of that always liquid. And we do that so that the community users can use this to unstake instantly. That, of course, has a fee because it is instant, and there’s no guarantee that that pool will always be there because users could use up that pool. But as our subscriptions end, we’ll keep topping up that pool.            

Alejo: If I’m understanding it correctly, there’s a small liquidity pool that allows people to pay an extra fee to unstake immediately. So potentially, the fees could go up or down depending on the demand for this, right? Are the fees that people are paying then redistributed to the pool?

Ditto: That’s a good question. It’s exactly as you’ve described. So we’ve got this pool so that users can unstake instantly because we want to provide users with a better option than having to wait up to 30 days or swap on a DEX. However, the way we have designed it, depending on demand and supply, the fee does move. In terms of where we’re having that fee go, we’re considering having it go back into the index and trying to create incentives to hold onto the product.

Alejo: Cool, that makes sense. So technically some portion of the APT that’s staked is kept liquid. The overall fee would be a little bit under 7% if this feature isn’t being used, but it could be way over 7% if a lot of people are using this instant unstake. So it’s like when I use Coinbase and I make a withdrawal that I need right away, I’ll pay that 3% fee to have it immediately in my bank account? Am I understanding that correctly?

Ditto: Yes. The fee isn’t that high, but the idea is basically the same. Users that want to exit immediately without having to wait or go through a DEX pool, they get a better option. And the users that stay get a small prize. So it’s a win-win for everyone.

Alejo: Very cool. And I imagine some of the issues that came up in Lido and ETH, staked ETH, were due to  there being no exit for unstaking ETH currently. Maybe that’s a super interesting dynamic that I don’t fully understand. Maybe you could share the similarities and differences between Ethereum and Aptos?

Ditto: Sure! So the result of that difference is that staked APT will keep the peg much better than staked ETH. I’ll explain it using an example. With staked ETH, you cannot unstake it back to ETH. The only way you can go back to ETH is by doing a swap. So what that means is, if staked ETH is at a discount to ETH on Curve or somewhere, of course you can try and buy that staked ETH and bring it back to ETH.

But the important thing to realize is that it’s an action which costs capital and you can’t close the loop. What that means is: you buy that stETH with ETH and now stETH is at parity with ETH. Now, if you want to exit that position, you have to settle. And that doesn’t work because you can’t close the loop. So the difference here is you actually can close the loop. With stAPT and APT, let’s say stAPT is at a discount to APT. What you can do is buy stETH at a discount, which is the same as the ETH example, but now the core difference is to go from stAPT to APT, you don’t have to sell, you can just unstake. So you can buy stAPT and unstake it and end up with APT. With stETH, you’d have to buy stETH and sell stETH, which would result in no net change in the price of stETH, which wouldn’t help improve the peg. With the stAPT case, you’ve got that buying pressure which buys stAPT and then you can immediately unstake and sell that APT. That means that that peg will come into line because upchargers and traders can close that loop. They can do that conversion using the unstake function.

Alejo: So if I’m understanding, 1 APT should always equal 1stAPT, and then any differences traders will likely arbitrage to make a small profit?

Ditto: I guess to be specific, in the short term, of course, price fluctuations are possible. But whenever APT subscription ends, there is that brief moment when the stake is ending and the unstake will take five minutes. In those moments, we would expect the stAPT price and the APT price to come in line because literally in five minutes it’s free money. If you buy any stAPT at a discount to APT right now, in five minutes you’ll be able to realize that profit. So the dynamic I would expect to occur is that as our subscriptions end on our validators, during those moments it will come in line. People will start to realize it comes in line once every week, two weeks, or every month depending on how many validators we have. Then as the market becomes efficient, people realize that’s a trade you can rely on because fundamentally it’s in the code.  

Alejo: Who are these people that arbitrage? Are they regular people or institutions?

Ditto: Honestly, anyone can take advantage of these opportunities. Some people are more experienced than others. Whether you’re talking about the large market makers or even just regular users, there’s nothing stopping you. The difference is it depends on the scale. If, fingers crossed, Ditto blows up in a good way and we have lots of staked APT, and the Liquidswap pool has a ton of liquidity, in this case, you would expect the institutions to come in. Simply because it’s the type of arbitrage opportunity that is very profitable for them at scale. At the scale we’re currently at, I wouldn’t expect institutions to be looking.

Alejo: I just thought of an idea: is this something that can be done on-chain with a vault, if people want to take on that five minute risk?

Ditto: Absolutely. We’re actually looking at this with one of the other protocols building on Aptos. I won’t reveal who that is, but that is an idea that has been thrown around. The risk is defined. Of course, there is risk. You know that at a certain time, there will be a profit, but there is still a chance the peg can move against you in the preceding days.  

Alejo: Nice! I’m excited to see that happen. This ecosystem’s only been alive for two weeks so it’s still very early. A lot of this infrastructure will need to be developed and it’s interesting seeing the differences in these on-chain products that give more democratized access through vaults, where people can just deposit into trading strategies versus some of the large market makers. Where do you see Ditto going in the future and how do you think it can differentiate long term? Maybe you could drop some alpha on your governance token?

Ditto: Sure! This would be a great time for me to emphasize our vision. I see the opportunity for Ditto to do staking better than how it’s been done before. People wonder how we can do this, and there are a few factors. The first I’ll mention is pretty key. Right now, we’re two weeks into the Aptos launch. Ditto was one of the very first protocols that launched on Aptos. A few hours after mainnet, we were the first liquid staking protocol on the chain, fully audited by the time we launched. This is different to how liquid staking has evolved elsewhere. If you look at Ethereum, Solana, and other chains, liquid staking comes a few months later. And when it does come, many of your primitives -- whether it’s DEXs, lending protocols, stablecoins -- already exist. It’s difficult to try and integrate it backwards into those protocols.

The reason why we launched Ditto immediately is that my vision is for all DeFi on Aptos to go through staked Aptos. I don’t want anyone holding APT. I only want people holding stAPT. The reason for that is I believe that staked APT is practically better than APT. There are two detractors. One, of course, is that you cannot pay gas fees in stAPT, you have to use APT but it’s very cheap on this chain. The other is, I acknowledge that there is a layer of smart contract risk. We are audited through OtterSec, who are the top Move auditing firm, and we are very confident in our abilities. But having said that, I acknowledge that we are still new. Apart from that, we think that staked APT is better in every application. As a user, you’re helping the network and you’re earning yield. Staked APT can also be used anywhere on DEXs, and it can be a trading token. I imagine that instead of having APT pools everywhere, we have stAPT pools everywhere. Similarly, you can mint stablecoins against it. That’s been a large part of our focus too, reaching out to protocols and working on these partnerships and integrations. I won’t reveal any new ones right now. A few have been released on our Twitter, and there are definitely more to come. But expect to be able to use stAPT everywhere.

Alejo: Is there an analog of any other ecosystem where the staked derivative is used more than average?

Ditto: I’d say Lido has done a decent job, where staked ETH are used a lot to mint DAI. But we don’t use staked ETH everywhere. One of those reasons is you can’t make staked ETH trade at parity to ETH because of what we talked about earlier. But because you can unstaked stAPT to APT, that peg is much stronger. It also makes them a much more fungible product because you just go from one to the other. That, plus the fact that we exist so early in the ecosystem, I think that’s what changed the opportunity for us. We see liquid staking protocols on other chains trying to do the same thing, and they’ve had some measure of success, but we believe, with the opportunity we have, that we can do better.

Alejo: And this is a very composable Lego brick. For example, if I’m providing liquidity to a pool, let’s use Dogecoin in this example, I could be doing DOGE-APT because I want some exposure on APT. What you’re saying is if I use stAPT, I could do the same thing except I would also be getting 7% on my APT?

Ditto: Exactly! You would be doing the same thing except you would be earning a little bit more money, which always helps.        

Alejo: What I can imagine is some of the automation around routing, even if people have APT or some other pair, as long as there’s enough liquidity on the stAPT against DOGE and other native assets like USDC, ETH, etc., then someone will find a route.

Ditto: Exactly. And on that, we have worked with Aptos routers and we are integrated. But in terms of the routers, for the APT-stAPT swap, they can go through the DEXs. That’s also why we have the immediate unstake, so the routers can use that option instantly. Overall, our end goal is we want stAPT to have more liquidity and be used. That’s what we’re pushing for, and that’s the vision.

Alejo: The one place where I’ve seen the staked derivatives also taking a large part of the ecosystem is Polkadot and Kusama, where staking is a requirement to get those chains live. So for those that aren’t familiar with Polkadot and Kusama, they’re somewhat similar to Cosmos, where it’s independent app chains, but they’re sharing security. As a filter, almost like spam protection of launching these, you need to stake either yourself or through crowdloans -- which are community stakes in exchange for rewards -- the native DOT or KSM tokens, and then the parachains are able to exist for 4 years on Polkadot and 2 years on Kusama. In that ecosystem, I think something like 40% or 50% of the tokens are staked, which is kind of a crazy amount. Liquid staking derivatives also seem to play a big role there, given that there’s actually not a lot of unstaked tokens. It’ll be interesting to see how this all develops. One question asked by someone from the community: How do you see this ecosystem developing with more centralized ecosystems like Coinbase? What are the advantages and disadvantages?

Ditto: That’s an interesting question. This could be a good opportunity for me to talk about how we’ve done our validators at Ditto. So decentralization versus centralization, it’s a very hot topic. My opinion is, I believe decentralization is important and that’s why we’re in this space. It’s called DeFi. We’re building these decentralized, trustless applications because we feel it’s important. Having said that, I think whenever you’re trying to do anything valuable and new, you can’t get there instantly. The way I think about decentralization as a whole is it’s the end goal we’re trying to get to. But it’s one of those things where you need to make sure to take the appropriate steps towards it.

What I mean by that is that protocols don’t start off decentralized. You have a team of, like, six guys come together to build something, so protocols start off completely centralized. But the vision is there to have decentralization. My thought is: decentralization is something you want to slowly give to the community as the chain, community, and team mature. I feel that taking baby steps is the way you build that sustainable, mature governance. Definitely admitting everything straight away, being like, “Ok, cool, we're going to launch governance on Day One.” I don’t think that’s the right way. It opens you up to attacks by actors which may not be aligned with the long term goals of the community and the protocol.

And in terms of how that relates to our validators, so we are working with 4 large, centralized validator companies. They’re all on dozens of chains and control hundreds of millions of stake. They have very high uptime and their run times are amazing. People may come to us and may criticize us for having a pretty centralized validator set. What we say is, “Yes, that’s correct.” Why? Because Aptos has not been around a very long time, and new chains can have issues. Fingers crossed we have none. When things are new, you want to make sure you’ve got people with that experience. If there are any issues at all, we’ve got full confidence in our validators that they’ll have great performance and uptime. And that’s better for everyone because when the validator does better, you actually get more rewards. If your validator has downtime, of course that’s bad because you’re not earning any rewards. We’ve done that so this way, you can rest assured that we’ve got the best validators working with us. They know what they’re doing.

Long term, is that the goal? Also no. Long term, we do want things to become more decentralized. Our overall goal is to slowly bring about decentralization and have validators come out from the community. I think that’s the approach to decentralization. At its core, you need a product that works and is reliable. That’s something I wouldn’t want to compromise on. Then as things mature and people understand that you started off with a centralized validator set, which gives us the best product, eventually we will open it up to the community and have that decentralization as things mature, as we become better at this. There will still be rigorous guidelines and processes for the community validators that the community itself will create and decide on. But we will open up that option down the line.                                

Alejo: This is super interesting and in my opinion, decentralization is a spectrum and there are actually tradeoffs to it. Like trying to organize a bunch of people on the internet to come around one decision. Just imagine trying to have people pick what parameters should be in this very specific DeFi protocol. Sounds very very hard.

However, we have these key primitives that allow us to vote and transact value on the internet. In my opinion, it’s a tool, and then how we use it and leverage it is up to us to create the best thing for the users. There’s inherent central points of failure, but also there are added benefits. At the end of the day, the market will self-select and settle into what’s best for the users. It’s awesome that you guys are building this and thinking about decentralization and the future and even starting out with liquidity mining programs to start getting those governance tokens out into the world so people can participate in this protocol. There’s been a lot of FUD around Aptos staking. There’s a thread that went viral not too long ago by Paul Fidika who is building Open Rails. I wanted to get your take on it. Let me read the tweets.    

Ditto: I remember the exact thread you’re talking about. What I would say is, I’m not the CEO, I’m not the core developer, so on some of the technical details, I would need to defer to my CEO. I think the core thing on that thread was it came across as it was from someone who potentially had a negative experience and who is quite upset and ranting. In terms of the points on that, a number of them were incorrect, or they were adjusted after the thread was posted. Some points he made I wouldn’t be able to comment on without preparing something with my CEO, unfortunately. Staking on Aptos is different from other chains, but staking is different every time. I do think Aptos has made some compromises in order to facilitate an extremely high transaction per second and very short uptime. I think that’s just how these things are. I think sometimes, you need to make compromises to differentiate yourself from the competition. I think what Aptos is aiming for is to be a very high throughput chain, and they’ve made moves I think will give them an edge there. Whatever changes you make will involve some kind of compromise.

Alejo: Exactly. I’m definitely not an expert in this. It would just generally be my take that in order to have high security, high uptime, high reliability, we make these decisions architecturally to make sure high quality validators exist as opposed to low quality ones that would potentially take down the network. I generally agree that, early on, we should be prioritizing uptime, reliability, security, so these decisions make sense to me personally, but I do want to see potentially some experimentation and movement into trying to lower the minimum threshold. It would be interesting to see potentially how that could be done. Generally I think the right choices were made. We have a network that works, which is the most important thing. It can manage the capacity. So far, we’ve seen it with pretty high throughput so it’s exciting to see that it actually works and over time we can move more toward decentralization.

Thank you Ditto for joining us! If you would like to see the full live stream, check it out on YouTube. Follow Ditto on their Twitter and Discord for more updates about their product. To stay up to date with the latest updates and content from Pontem, follow up on our Twitter, and join our Discord and Telegram chat. We hope to see you back for our Wednesday and Friday live streams!