I think now is appropriate time to let @rspa jump back in (or any other party that has willing to design an run incentives program). We love competition guys.
2 Likes
I remember the rationale behind the $20M / 6 months structure from an older thread. It was based on historical data regarding fees to rewards ratio.
If a renewed program has the same or at least very similar structure as the past six months (sounds like it), I think it is very reasonable to expect that the results will be the same â aka not so great.
I am a trader and I go where I can make the most money. I donât see a big difference in terms of UX or price execution between the biggest platforms, so I go where I get the most rewards.
For the past year, maybe even right now, that is Hyperliquid, where you are probably making anywhere from 2x to 50x on fees spent (2023 more lucrative than now of course) based on pre-market points trading.
IIRC, the targeted ratio of rewards to fees here is way below 1.8.
In short, I am all for a renewal of an ongoing rewards program, but I think the $20M / 6 months structure is not competitive in todayâs landscape (numbers prove that) and wonât be able to attract NEW power users from other platforms.
I donât have a clear better alternative right now, but maybe something more extreme like a $10M / 1 month DYDX Unlimited launch push with a lot of experimentation (see all suggestions here) could help show a real significant change and then we can go from there and evaluate what should happen afterwards.
4 Likes
Itâs clear from the ongoing discussion that the current rewards program has not met the communityâs expectations. Continuing with a failed approach is simply throwing good money after bad. Itâs apparent that the main beneficiaries of this program are a few centralized entities, rather than the broader dYdX community.
The idea of running a third-party centralized rewards system doesnât align with the principles of decentralization that dYdX stands for. If @ChaosLabs were genuinely interested in enhancing the platform, they would focus on building an improved rewards system directly into the dYdX platform, rather than pushing for an external, centralized program.
We already have a native rewards formula that operates in a decentralized and autonomous way, rewarding users instantly. The logical step would be to refine and improve this existing system rather than introducing additional layers that primarily serve the interests of a few parties. Itâs time to address the real issues and stop turning a blind eye to the underlying conflicts of interest that are driving these proposals.
4 Likes
You will also offer to pay 110% of the commission on it, then the trading volume will increase. But this will be 99% fictitious trading.
In general, cancel this instant reward in DYDX tokens, because of this the token fell by 75% in 6 months.
Commission rewards can be reimbursed in vouchers that can be spent on commission. Cashback that can only be spent on commission.
And not spend millions of dydx tokens on instant rewards that are immediately dumped on the market.
- No one has proposed other incentive programs, or developed them, ready to launch by the end of September
- Offering a competition for a dozen traders with the provision of capital. This is zero effect on attracting a wide audience.
Until another program is ready to launch. This one can be extended. Waiting 3-5 months for a new one in the crypto market is suicide
- Incentive programs should be simplified as much as possible and colorfully (attractively) displayed in the trading interface.
- The trading competition by leagues should be left on a permanent basis. Expand the award places, for example, the top 100 in the bronze league. And show the results in the trading web interface. Moreover, all this is already in V3 DYDX.
1 Like
The reason is because the grants team here will only work with certain contributors. Usually ones that they are invested in. There have been plenty of others and some great teams try to do things here but they donât get the support they need so they exit left. If teams and contributors are not going to be supported then Trading Inc should be doing this stuff. They should have built the system properly from the beginning and they should be building the competitions, funded accounts system and so on.
1 Like
Thanks for thinking of us 
This thread comes to mind: x.com
Creating sticky users of the long tail is probably something that incentives are especially bad for. Either something truly fulfilling, interesting or special happens there, or users leave the second incentives dry up, as they always do.
We think ChaosLabs have done a good job of distributing the incentives so far. But as some community members pointed out in this thread it hasnât done much for the longer tail of trading assets.
BTW: The research here is mindblowing! Kudos to everyone involved!
4 Likes
Good thread. Thanks for sharing.
A proposal for 6 months extension is onchain. Is a strong NO from me. I still donât understand why it couldnât have been extended for just one or two seasons at most and then introduced a revamped program.
In its current form, this looks like Chaos Labs trying to secure their own payment. Such long-term commitments benefit only the vendors. Weâve already extended the grants program for more money and two years, and have we seen any positive changes that the community requested? No other team will be interested in proposing alternatives, and without healthy competition, this will all lead to the same mediocre results.
1 Like
Maybe some respected organizations like @gauntlet can offer an alternative. So we have a healthy competition
Thanks for the tag, @RealVovochka
As a quantitative research firm and former dYdX delegate, we agree that this programâs efficacy is unclear and should be further examined. Given the current community sentiment and the magnitude of this program, it seems prudent for the dYdX community to consider multiple options. The current structure potentially puts dYdX at a disadvantage relative to its peers.
Gauntlet has extensive experience designing, implementing, and executing incentive programs across different DeFi protocols such as Uniswap, Immutable, Compound, and others. We additionally consider optimal incentive schemas in relation to trader UX and growth for perpetual exchanges like Jupiter and Perennial.
We welcome the opportunity to give the dYdX community an alternative program to consider. Weâre confident that we can design, implement, and execute a program that meets the needs of dYdX and the community.
Given the active vote, we look to the community to suggest next steps.
2 Likes
Thank you for your post @gauntlet.
Personally, I am specifically against the idea that a proposal responsible for 20+% of the available community treasury needs just 7 days to discuss. It was a premature decision to make this proposal live after just 7 days, two of which fall on a weekend during the most active vacation month.
Therefore, I believe it is in the long-term interests of dydx to vote against.
There are still a few weeks left until the end of Season 6 and I donât even see a problem with extending it, which is enough time to consider all alternative solutions and choose the best one
1 Like
I have an idea about the League. I find that when someone have already got high PNL, in order to keep the high rank they wonât like to trade anymore. Itâs not that great for League activity continue 6 weeks. Maybe we can change the Leagueâs period from 6 weeks to 1 day, and decrease the reward according.
2 Likes
Hi @shuo , completely second you on this, I had raised this point few months ago - Suggestions to Improve Upcoming Launch Incentives
1 Like
Unbelievable this has already gone to a vote. Itâs CLEAR that the program has failed to achieve itâs goals yet itâs being put to a vote to extend it. Strong NO from us. This program only benefits a few companies. Letâs hope those casting their votes can realise this before we have another chunk of money siphoned out of the community treasury that does nothing to benefit the community or the platform.
2 Likes
On behalf of the PRO Delegatorsâ validator, we want to share our review for this proposition and explain the motivations backing our vote.
First of all, the level of commitment to this thread is remarkable. We had to carefully review and check multiple posts before considering our decision, as many participants showed extreme care in crafting their feedback. We can only commend this dedication to making dYdX a better place for everyone.
Considering the detailed reviews provided and the numerous effectiveness problems raised by the community regarding this additional incentive program (letâs remember that volume-based activity already receives direct on-chain incentives, to which this program simply adds up), we have decided to cast a negative vote.
We believe the engaged discussions are proof that more careful planning is required before allocating a significant share of community funds to this programâs renewal. Nevertheless, we also want to emphasize our support for ChaosLabs and our willingness to extend incentives and further distribute the dYdX token supply to actual users of the chain. According to this vision, some people here have raised legitimate concerns over efficacy to target these real users, and provided well-crafted ideas for more effective solutions to alleviate the bots distributions. We hope to see an updated proposition to renew these incentives in the coming months and invite the proposer to take all the necessary time to assess and include more community feedback in their design before considering another proposition, which we will be happy to support.
To be clear, we do not oppose the program; we simply believe a more sophisticated version can emerge from this one after collecting and including the valuable community feedback.
Thanks for reading,
Govmos
8 Likes
Hi everyone, this is Paul from OpenBlock â we are an incentive modeling platform that powers 2B+ in annual incentive spend. Below, we share our thoughts on the proposal to encourage further discussion. If the community finds our insights valuable, weâre happy to take a more active role in designing or implementing the program.
Context
The proposal from Chaos Labs to relaunch the dYdX incentive program has sparked mixed reactions. While some support the continuation of the program, others criticize its current structure, noting that rewards are mostly captured by professional traders and bots, which skews the programâs benefits away from retail users. Critics suggest revising the program to better target retail traders, introduce gamification, and adjust the reward distribution to ensure fairer competition.
A proposal of $20 million in incentives for the next six months has been made, adding to the $30 million already spent over the previous six seasons. This has sparked mixed reactions within the community.
As part of this discussion, we are using dYdX as a case study to understand how incentives donât always work as intended, why this happens, and what can be done to address it. Weâll also present our view based on incentive programs we have designed for different Perps and how to enhance their effectiveness, ensuring they better align with their intended goals.
Assessment
What we know
In Perps, the distribution typically follows an exponential pattern (Pareto CDF in the following plot), meaning that the majority of trading volume is generated by a few number of addresses. This highlights the importance of understanding the behavior of power users and how it influences retail participation.
Source: OpenBlock
Hereâs a typical distribution of trades in Perps. These figures are based on simulated data, so you wonât see any outliers. However, they still provide insight into the general shape of the curves and illustrate how trade distribution tends to be skewed towards power users.
As pointed by @realvovochka, volumes and Daily Active Users (DAUs) have decreased over the last 6 months, despite the incentive program at play.
Source: @realvovochka
Besides, an increase in user growth doesnât automatically translate to strong user retention. Without solid user stickiness, the strong network effects seen now may weaken over time, making a user growth strategy ultimately ineffective (Zhou et al., 2024). As shown in the plot below, user retention has failed.
Source: @realvovochka
Perpetual trading platforms operate in an extremely competitive environment. The intense rivalry among Perps is exacerbated by âvampire attacksâ â where platforms attempt to siphon users from competitors through aggressive incentive programs and airdrops. This has led to users who arenât sticky, constantly chasing the next yield farming opportunity or reward. The real challenge is not just drawing users in, but finding ways to keep them engaged and loyal despite the constant pull of new, short-term incentives elsewhere.
What should be studied more deeply
Understanding network topology and density is essential for assessing a platformâs network effects. Denser and well-connected networks often lead to stronger user engagement and retention (Jackson, 2008). Analyzing the structure of the network can help identify key clusters and target strategies to enhance overall platform health and stickiness.
Another important point is to track who is selling tokens when allocations are distributed to identify short-term speculators. To encourage long-term commitment, we can consider implementing bonuses for users who hold their tokens over time, which can help stabilize the platform and reduce volatility.
Designing Incentives With Frictions
Before delving into designing the next incentive program, itâs crucial to achieve critical mass and maintain high liquidity while addressing several key challenges. Below are the main challenges and potential solutions that the program design should consider while modeling incentives:
- Gaming and exploitation: Openly shared incentive methodologies are susceptible to manipulation by sophisticated actors, leading to behaviors that can undermine the integrity of the program. If participants can easily predict how to maximize rewards, they might exploit the system in ways that were not intended, reducing the programâs effectiveness.
- Engagement of takers: Decoupling the roles of takers and makers to reduce risk for takers might inadvertently lead to reduced engagement. When takers face less risk, they may become less involved in trading activities, which can negatively impact overall market liquidity.
- Transparency vs. yield farming: Thereâs a fine balance between making the incentive structure clear and understandable for users and keeping certain details obfuscated to prevent gaming. While transparency can encourage participation, it also increases the risk of the program being exploited if the incentive mechanisms are too predictable.
Potential solutions are:
- Balancing clarity and obfuscation: The incentive program should provide clear guidance on the types of behaviors that are rewarded (e.g., consistent trading, providing liquidity) without revealing the exact algorithms or criteria. This approach helps maintain user engagement while reducing the risk of gaming. Regularly updating or rotating specific incentive criteria can further enhance unpredictability, protecting the systemâs integrity while keeping it accessible to users.
- Enhancing the flywheel effect: To strengthen the positive feedback loop between Market Makers (MMs) and traders, the program should carefully adjust the incentives-to-fees ratio for makers. By aligning these incentives with the fees paid by MMs, the program can ensure that MMs remain motivated to provide deep liquidity, which in turn attracts more traders and sustains high market activity.
- Optimizing liquidity sources: Identifying the best sources of liquidity and integrating them into the program design is crucial. The design should focus on ensuring that the market remains stable and attractive to both MMs and traders, which is key to maintaining a healthy and efficient trading environment.
Modeling Incentives with Economic Theory
There are multiple streams of economic research that can be leveraged to curb back up the daily active users on dYdX, stemming from mechanism design, contract theory, prospect theory, and two-sided markets. Our analysis is solely based on seminal contributions made in these respective fields. We are outlining here a few routes dYdX could follow based on our experience with Perps incentives programs. We surely lack some context, especially on the data side, to design a definitive incentive program. For simplicity, we assume a manageable compute cost maintaining the program.
We also focus on the retail side of the incentives in this article. Let Ui(t) the utility function of user i. Ui(t) is the constant relative risk aversion utility function, such that:
The parameter measures the degree of relative risk aversion that is implicit in the utility function. Hence, we can maximize utility under constraints, such as USDC deposits on dYdX.
The next question that comes to mind is what type of behaviors are to be maximized in a Perp and how can we achieve this? Here are a few incentives can be leveraged through a program:
Dynamic rewards based on engagement history: As seen in models of dynamic contracts (Hart and Holmström, 1987), where rewards are adjusted based on past performance, dYdX could implement a system where the more a trader engages with the platform, the more lucrative their rewards become over time, but with a requirement for consistent participation to maintain these rewards. This concept is grounded in the folk theorem, which suggests that long-term cooperation can be sustained if future rewards outweigh the short-term gains from deviation.
Loss aversion-based rewards: Implement a system where traders accumulate potential rewards that are âlockedâ and can only be unlocked through continuous engagement. If traders stop engaging with dYdX, they risk losing these accumulated rewards. This is grounded in the seminal research made by Kahneman and Tversky (1979).
Collaborative trading pools: Enable traders to form groups or pools where their combined trading activity can unlock higher rewards for all members. This leverages the network effect (Katz & Shapiro, 1985; Tirole & Rochet, 2003), as traders have a vested interest in encouraging each other to stay active.
Vest-and-lock rewards: Implement a vesting period for rewards where benefits accumulate over time but are only accessible after a sustained period of trading. If traders withdraw early, they forfeit a portion of their rewards, encouraging them to remain active to maximize their returns. This is grounded in the work of OâDonoghue & Rabin (1999), studying decision-making processes between short and long term preferences. Traders who are able to wait on their reward are more keen to stick to dYdX and not only here for farming.
Important note: these programs should be primarily data-driven so as to verify and closely monitor the accuracy of our model. Since we do not have such granular data for dYdX, we have kept the theoretical part for educational purposes.
Hence, the overarching utility function could be defined by these parameters, incentivizing long-term committed traders. Nonetheless, the allocation should be tuned to be competitive wrt other Perp protocols.
Concluding Remarks
When budgeting for user acquisition, itâs crucial to consider both the cost and the quality of users gained. Jupiterâs airdrop, with a user acquisition cost of nearly $1,000 per user, created buzz and rapid onboarding but didnât ensure long-term engagement. Such high costs often attract yield farmers, leading to a temporary increase in users without sustained growth. Therefore, while airdrops can quickly boost user numbers, they must be carefully designed to attract users who will remain active and contribute to the platformâs ecosystem.
In order to attract the right users, it is important to conduct thorough wash trading and Sybil analysis. These attacks are significant concerns in the context of incentive programs and user acquisition, though they are outside the immediate scope of this discussion.
While these issues are not directly addressed in the current analysis, they are critical topics that should be explored in further research. At OpenBlock, we have designed algorithms to address both of these issues, and would be happy to assist with the implementation of a long-term incentive program for dYdX. Understanding and mitigating these risks is essential for ensuring the long-term integrity and success of any incentive program.
References.
Hart, O., & Holmström, B. (1987). The Theory of Contracts. In Advances in Economic Theory.
Jackson, M. O. (2008). Social and Economic Networks. Princeton University Press.
Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica, 47(2), 263-291.
Katz, M. L., & Shapiro, C. (1985). Network Externalities, Competition, and Compatibility. The American Economic Review, 75(3), 424-440.
OâDonoghue, T., & Rabin, M. (1999). Doing It Now or Later. The American Economic Review, 89(1), 103-124.
Rochet, J.-C., & Tirole, J. (2003). Platform Competition in Two-Sided Markets. Journal of the European Economic Association, 1(4), 990-1029.
Zhou, Z., Zhang, L., and Van Alstyne, M. (2024). How users drive value in two-sided markets: platform designs that matter. MIS Quarterly, 48(1), pp.1.
7 Likes
The issues here are far deeper than just the incentives alone. Itâs poor leadership, self interests at play, grants teams with a lack of understanding as to what traders need. And the list goes on.
I donât believe this statement is fair; the foundation has done extremely important work to get us to where we are. These conversations, in my opinion, often require time to respond thoughtfully and with more data. In this regard, I think itâs important to maintain a respectful tone, understanding that the interest is shared. In any case, I believe the interesting thing about DAOs is the ability of their token holders to support those initiatives or teams that can best contribute to the project.
3 Likes
Thank you very much for your contribution; thereâs a lot of information to digest. The reflections on incentive programs are interesting. However, I believe itâs important to take into account the business dynamics involved in derivatives trading. We might think that this is a traditional business where we should try to retain our customers by encouraging them to use our product more, but I believe these comparisons do not apply in this case. In derivatives trading platforms, itâs common not to achieve customer retention in most cases, especially with retail clients, because itâs very difficult for them to acquire the necessary experience to be sustainable in the long term. In this sense, I think these retention statistics are unrealistic, and in reality, derivatives brokers are continuously attracting new customers to try to compensate for the loss of clients. There are certainly ways to change this model, but not by subsidizing customersâ losses with our own tokens, especially when the profits are being extracted by other professional traders who are indeed profitable. For this reason, I believe we should perhaps focus more effort on attracting more customers while also striving to make clients more profitable through trading, not through subsidies.
3 Likes
I donât believe I said the foundation hasnât.
1 Like