Malt V2 Teaser + Update

Published in

Malt Protocol

8 min readAug 12, 2021

Greetings Malt Community,

Since our last announcement, we’ve been diligently preparing for the launch of Malt V2 and we are thrilled to be rolling it out soon. Through the experience and knowledge we gained during V1, we have fixed bugs, but more importantly, we have fundamentally built a better protocol. We’re excited to share with you a view behind the scenes into the work behind V2 and provide other updates on where Malt is going.

As this is going to be a lengthy post, here is a breakdown of what’s to follow:

Focal Vesting
Reward smoothing
Implied Collateral
Escape hatches on Arb tokens
Rebrand
Community Plan
AMA

Focal Vesting

Let’s start with the bugs from V1 and what we are doing to address them. As many of you already know, the bug that shut down V1 was in the liquidity mine contract. Specifically, the bug occurred in the way vested earnings were calculated. This calculation was complex and as with all complex things, that means the chances of making mistakes is high. Instead of simply fixing the bug and leaving all of the complexity (and the possibility of more subtle bugs existing), we have decided to remove the complexity and replace it with something simpler.

In Malt V2 rewards will still vest over a period of time, but the exact way the vesting occurs is changing. The new way of handling vesting makes the code significantly simpler and drastically reduces the chances of bugs,with little to no impact on the user experience.

The new system is something we are calling “Focal Vesting”. To understand why we are calling it that lets quickly run over how vesting worked in V1. In V1, rewards were generated by the stabilizer above peg and then allocated to each user pro-rata to their share of the bonded liquidity pool (LP). The allocation for each user then vested linearly over a 24 hour period.

The issue with this (and ultimately what caused the V1 bugs) is that rewards can come in at any point and then the vesting ends 24 hours after that. This meant that there are potentially many rewards all finishing their vesting at different points in time. To handle this, the protocol needed to keep track of exactly which “batch” each reward came from and exactly when that batch vests fully.

As an alternative for V2, we will utilize specific points in time called “Focal Points” where rewards will vest, regardless of when the rewards were actually created. If we set a focal point every 24 hours then the minimum vesting period on rewards is 24 hours and the maximum is 48 hours.

In practice, this means some rewards will vest over slightly different periods of time, but the differences are minimal and the impact to the user experience is zero. Zero impact to the users along with a significant reduction in code complexity is a massive win in any software system, especially one that controls significant amounts of money.

Reward smoothing

Rewards in V1 were great! However, one of the core problems with the V1 rewards system was that rewards only came in when the stabilizer produced them. In other words, if a lot of activity happened in one epoch, it would have a lot of rewards, and if the very next epoch had little activity, the rewards would be low. That creates an erratic Annualized Percentage Return (APR) figure, which as many pointed out, was at best confusing, and in many cases it created fear and uncertainty.

In V2 we had two options:

Simply average the epoch APRs over a longer period of time so the APR shown in the UI doesn’t fluctuate aggressively.
Change the way rewards are distributed to actually smooth out the APR.

Option 1 doesn’t really fix the problem, it just hides it. So we ruled that one out fairly quickly. That leaves the second option — changing the way rewards are distributed.

The direction we chose to go is option 1.5, which lands somewhere in between 1 and 2 (but mostly 2).

The protocol will keep track of the average APR of the total rewards generated and will endeavour to pay out some % of that APR every epoch (say 50%). Any capital over that throttled APR goes into an “Overflow” contract. The money in the overflow contract is used when rewards in a given epoch aren’t enough to reach the desired throttled APR.

Target APR is some amount less than the average profit

The overflow goes towards covering underflow later

This can be thought of as using excess profit during high activity epoch’s to sustain rewards through the low activity ones. The result is smoothing out erratic and unpredictable reward generation.

Benefits of this method include:

The overflow pool “charges up” during high demand periods which allows a guarantee of rewards during lower / no demand periods. This guarantee of reward capital will serve to create an additional favorable layer to the game theory of joining / leaving LP positions.
The overflow capital isn’t allocated to rewards yet, which builds into a new concept of implied collateral.

Implied Collateral

This is going to be one of the core ideas in Malt V2. Malt V1 had a Liquidity Extension which in some sense was collateral backing the coin. It was capital owned by the protocol that was used to help maintain peg. The additional capital to support the peg came from injections from protocol users via the auction mechanism.

In V2, there will still be protocol owned Liquidity Extension, but there will be other pools of capital that will not necessarily be owned by the protocol (at least not for the express reason of collateralizing Malt), but can contribute towards an “Implied Collateral” value.

Overflow Pool

As mentioned above there is the overflow pool which under normal stable conditions will be used to smooth out rewards when rewards during a given epoch don’t meet the desired APR. If needed, this pool can also be used to backstop peg.

Auction Pool

Another new contract called the “Auction Pool” will also add to the implied collateral. This new Auction Pool is filled using a small % of the above peg profits. The pool is owned by the bonded LP’s, pro-rata their share. However, the rewards only unlock once that capital has been used to defend peg. This capital is collateral in the sense that it is being used to defend peg, however, it is still user owned speculative capital. That capital being used for two distinct, but aligned goals bringing new efficiency to Malt.

Swing Trading Contract

Another new contract introduced in V2 is the Swing Trading contract. This is a contract that will buy Malt under peg and sell it above peg. It is doing so on behalf of the protocol itself. The profit generated by this process can be used in any way the protocol deems useful.

This contract will be filled using rewards that are forfeited by users who unbond. So every time a user unbonds and forfeits their unvested rewards, they are adding to the implied collateral of the protocol.

Auction Order Book

This is an area of active research for us, and we are not 100% sure whether this will make it into V2 or V2.x. However, we will mention it here anyway. This will allow users to pre-pledge to auctions in price ranges of their choosing. This capital is again speculative user capital that is trying to generate profit, but it is also serving the purpose of defending peg. Again, this can be seen as an implicit form of collateral.

Taking all of the explicit collateral (i.e.the Liquidity Extension), plus the implied collateral (i.e.the Auction Pool and the Auction Order Book) and the bits in the middle like the swing trading contract, we end up with a total implied collateral value.

Over time we can optimize the protocol parameters such that the size of each of these pools can allow us to make some level of guarantee that is close to that of fully collateralized stablecoins without the protocol owning all of that collateral. Making use of speculative user capital as a form of incentivized collateral brings a new dimension and capital efficiency to the algorithmic stablecoin space.

Escape hatches on Arb tokens

One of the largest issues we saw with the Arbitrage Auctions in V1 was the risk profile of owning Arb Tokens. Simply making the Arb Tokens more profitable (by offering a premium) missed the fact that the risk profile on them is profoundly different to simply swing trading.

In essence, the arb tokens in V1 are an infinite time horizon binary call option. The binary nature (100% capital loss or profit of the premium) of the tokens is not a desirable risk profile.

For V2, we are exploring ways to offer “escape hatches” from the Arb Tokens to dampen the binary nature of the tokens. We want to:

Allow users to get out at a loss if price drops below their purchase price and they are not comfortable holding the Arb Tokens anymore.
Allow users to get out at break even or a small profit if the Malt market price is above where they bought the Arb Tokens, but we are still not back to peg yet.

These desired outcomes have to be carefully weighed up against the impact they will have on peg recovery. For example, we want to avoid incentivizing swing trading the Arb auctions to the detriment of peg recovery (this kind of swing trading behaviour already existed on the DEX during V1).

These are open problems we are still working hard towards solving along with other concepts around the peg mechanism and strengthening the concept of implied collateral.

Rebrand

We are also iterating on ideas around a rebrand and a redesign of the UI. We have some concepts already and are working on some others.

If you are a graphic designer, illustrator, UX designer or UI designer and would like to help out let us know at hiring@malt.money or message myself (0xScotch) or someone else on the team.

Community Plans

Community is a big part of our vision for Malt. We want Malt to be the best community in Defi. The best place to hang out. The best place to learn and ask questions about Defi. The best place to meet like minded people and make awesome new friends.

To this end, we want to:

Run contests / competitions
Incentivize doing great things in our community
Have regular community activities like game nights, movie nights, book clubs, etc
Have learning resources to help people get the most out of their defi journey
Much more.

If you have ideas for awesome things we could do for our community don’t hesitate to message us on Discord, Telegram or Twitter.

AMA

To provide a more freeform discussion platform, I will hold another community AMA call on Discord at 5pm UTC on Tuesday, August 17, 2021.

During the call, attendees will be able to ask me anything you wish about Malt V2, the community, longer term ideas / roadmap and anything else you can think of.

Looking forward to hearing from everyone and sharing more about the roll out of Malt V2!

0xScotch