A Strategic Plan to Neutralize CEX Supply Overhang and Restore GMX Price Discovery

A Strategic Plan to Neutralize CEX Supply Overhang and Restore GMX Price Discovery

Summary

Despite BB&D successfully buying back over 2 million GMX — roughly equal to the total CEX + DEX circulating supply at the time of its inception — the token price has continued to decline steadily. This proposal presents evidence that structural manipulation via CEX mechanisms is the primary force suppressing GMX price, and outlines a comprehensive 5-point plan to reclaim pricing power by redirecting protocol resources toward a targeted CEX-focused strategy.


Context & Problem Statement

BB&D Has Achieved Its Mechanical Goal — But Not Its Price Goal

The Buyback & Distribute (BB&D) program has executed as designed. Over 2 million GMX have been repurchased from the open market. At the time BB&D was proposed, total GMX supply across CEXs and DEXs did not exceed 2 million tokens. By any reasonable expectation, this level of sustained buying pressure should have meaningfully reversed the downtrend.

Yet the price has declined in a near-linear fashion.

GMX Remains One of DeFi’s Highest-Yielding Protocols

Even at current depressed prices, GMX generates some of the highest real yield in DeFi relative to its FDV. From a fundamental standpoint, the current price action is disconnected from protocol performance.

The CEX Supply Manipulation Thesis

Through extensive conversations with industry OGs and project founders, a consistent pattern has emerged:

Once a token is listed on certain CEXs and initial trading activity cools, an overwhelming and seemingly inexplicable volume of sell-side pressure materializes. This pressure drains project treasuries and pushes tokens into sustained downtrends. Most projects, with limited reserves, eventually capitulate.

The mechanism works as follows:

  1. Privileged entities closely affiliated with certain CEXs have access to special accounts that can be credited with unlimited arbitrary token balances and/or capital — or, in more modern implementations, can borrow far in excess of actual exchange reserves (e.g., borrowing 800K–8M tokens against an 800K reserve).

  2. Price suppression cycle: These entities sell aggressively to push the price down, then buy back at lower levels to replenish positions. Community and project buybacks are absorbed by effectively unlimited sell-side liquidity.

  3. Death spiral induction: After sustained suppression, organic holders begin selling, accelerating the downward trend. The manipulating entity profits across the entire price decline — their gain is essentially the integral of price over time as the token trends toward zero.

GMX, despite its strong cash flows, is not immune to this dynamic. The difference is that GMX does have ongoing revenue — which means we have the resources to fight back.


Proposed Strategy: 5-Point Plan

1. Withdraw All On-Chain GMX Liquidity

Action: Remove all GMX token liquidity from on-chain pools and return it to the GMX Treasury.

Rationale: Much of the current on-chain liquidity is protocol-owned. By withdrawing it, we eliminate the LP-side sell pathway entirely. There is no strategic benefit to providing on-chain sell-side liquidity while the token is under active suppression. This forces price discovery to occur primarily on CEXs — where we will concentrate our offensive.

2. Redirect Buybacks to CEX (Binance TWAP)

Action: Transition from on-chain buybacks to TWAP buybacks executed via a GMX Labs-controlled Binance account. The operational flow:

  • Transfer funds to the Binance account weekly

  • Execute TWAP buyback over the week

  • Withdraw purchased GMX to Treasury

  • Repeat

Rationale: By buying directly on the CEX where suppression is occurring, we directly absorb the manipulated sell pressure and drain available supply from the exchange.

3. Temporarily Suspend Staking Rewards Distribution

Action: Pause all GMX staking reward distributions. All accrued rewards will be held in the GMX Treasury. Distribution resumes only when GMX price reaches its all-time high of $90.

Rationale: This creates a powerful forward-looking incentive. Rather than distributing rewards into a declining market (where they are likely sold), we accumulate a growing pool of pending rewards that becomes increasingly attractive as the price recovers. This transforms staking rewards from a source of sell pressure into a catalyst for price recovery.

4. Introduce Staking Power Mechanism for Reward Distribution

Action: When rewards are eventually distributed (upon reaching $90), they will be allocated proportionally based on cumulative Staking Power:

  • Staking Power = Sum of daily staking amount snapshots from plan inception to trigger date

  • Loyalty condition: If a staker’s amount at the time of distribution is less than 50% of their historical maximum staking amount, their Staking Power will be set to zero

Rationale: This rewards long-term, committed stakers and penalizes those who unstake opportunistically. The loyalty condition prevents users from unstaking during the recovery and then re-entering just before distribution.

5. Deploy Treasury Buy Wall at $5

Action: The GMX Treasury will deploy a buy wall of 1 million GMX tokens at $5 (approximately $5M), expandable to 2 million tokens ($10M) if necessary.

Rationale: This serves as a hard floor and a safety net for concerned community members. It also represents an extremely favorable entry for the Treasury — any GMX acquired at $5 is an outstanding use of protocol funds. We explicitly welcome FUD-driven selling into this wall; it is a net benefit for the protocol’s long-term positioning. Even in the most extreme scenario where GMX reaches $5 and half of all stakers leave, the core protocol (trading and LP operations) remains entirely unaffected, as the token economy is already structurally independent from protocol operations.


Risk Assessment & Worst-Case Analysis

Worst case: GMX price drops to $5 or temporarily approaches zero, and up to 50% of stakers exit.

Impact on protocol operations: Minimal. GMX’s trading engine and LP system operate independently of the token. Traders and LPs are not affected by token price movements.

Upside of worst case: Treasury acquires GMX at historically low prices, and once the positive spiral begins, the accumulated pending staking rewards become an enormously powerful incentive for re-entry and new staking.


Timeline & Feasibility

  • Development time required: Less than 1 day

  • Strategic window: The current low price is precisely when this plan should be executed. At $90 or even $30, the risk of aggressive action would have been prohibitive. At current levels, there is very little left to lose and everything to gain.

  • Projected impact: At current buyback rates, the entire Binance GMX reserve can be depleted within approximately 2 months, effectively stripping the CEX of pricing power.


Conclusion

GMX is in a unique position among suppressed tokens: we have real, sustainable cash flow and a battle-tested protocol. The problem is not fundamentals — it is structural manipulation of supply on CEXs. This proposal attacks the problem at its source by concentrating all protocol resources on draining CEX supply, removing on-chain sell-side liquidity, and creating asymmetric incentives for long-term holders.

Heavy times call for bold measures. Let’s reclaim GMX.

Q

5 Likes

I’m thinking here Q, its going to take some time to digest but overall I like some of the plans being laid out here (remove the lines from AI though :slight_smile: ).

Imran

the dilution of staking yields from removing GMX tokens from CEX would be significant, so removal of staking yields I guess, is a stone killing two stones :laughing:

I’m supporting this. Just want to know how sure we are that CEX is suppressing our price.

we almost embarked on a “detective agency work” to check onchain and CEX activities to look at why token price is low, but we didn’t do it, as we still lay the blame on lower trader activities and new shiny tokens (other perps)

Lastly, I hope we don’t focus too much on price such that we create exit liquidity for large token holders. Treasury fund is huge ammunition for us to build better :slight_smile:

This is a brilliant idea, a final game. Group A believes the price will return to 90, while Group B doesn’t. Group B will sell all GMX at the midpoint price and forfeit all GMX rewards. Group A will hold all GMX and receive all GMX rewards. As the price gradually rises to 90, BB&D will continuously buy back GMX and lock up GMX holdings. However, there’s only one problem: once the price reaches 90 and GMX rewards are claimed, the GMX price will plummet. Everyone will short sell to hedge, could which pose a huge risk to GLP?

4 Likes

There will be selling pressure at 90 — that is entirely predictable. We could choose, for example, to switch to a linear release after reaching the trigger point, but I don’t believe that is necessary at all. This process could take three months, maybe one year, or even longer. Those who are able to hold on for that long and have sufficient staking power are very likely the most loyal users of GMX. The selling pressure is fully manageable.

2 Likes

Haven’t sold much except those I need for some crisis in my life, and definitely would not sell now.

I am supporting this.

I do want to point out that as loyalists, and as people who work with extremely talented people in this protocol, we are biased and too optimistic.

Waiting for Sir and Mdms to point out our flaws in thinking

1 Like

thank you for creating this proposal

i agree with the thinking here, and also agree there is not much to lose at this point

it is difficult to fully determine the causes of the price action these recent years, there can be many different explanations and contributing factors, attempting this can help bring some clarity

at the same time, i believe there are some considerations that may require adjustments regarding the plan:

  1. if on-chain liquidity is removed and tokens are bought back from CEXes and withdrawn, the end result seems to be that there would not be any liquidity both on-chain or on CEXes
    1. so i am wondering if it would make sense to still buyback on CEXes and withdraw, but reduce instead of eliminate on-chain liquidity
      1. A possible way to do this would be to reduce the Uniswap liquidity, while maintaining the liquidity in the GMX markets on GMX V2
    2. another consideration for this is because, supposing this leads to GMX token price increasing to e.g. $90 as targeted, there needs to be liquidity beneath $90 to make this a “real” price, otherwise it would be a price on liquidity that may be too thin
  2. i believe there is a requirement from some CEXes to maintain a minimal amount of liquidity to continue being listed, so that should be factored in as well, i don’t think a very large amount of GMX is needed for this, so this plan can be executed while fulfilling the requirements (i’m assuming GMX being delisted, while somewhat aligned with the plan, is not really a desired effect at this point)
    1. this also helps ensure the GMX markets on GMX V2 continue to function well, since these use the oracle price from CEXes
4 Likes
  1. Regarding the liquidity withdrawal issue, the current proposal is to withdraw all protocol-provided liquidity back to the Treasury. As far as I understand, most of the liquidity on Uniswap is provided by the Treasury, while all liquidity on GMX V2 is provided by users. Therefore, even if we withdraw liquidity, it will not affect the portion provided by users. In effect, this achieves the same outcome as preserving GMX V2, as you mentioned.
  2. Reaching 90 USD will be a sufficiently long process. Along the way, users will inevitably sell, and there will be full turnover at each price level before moving to the next. This TWAP-style purchasing approach is extremely gradual, meaning that at any given time and price, the level is well-supported. It is very different from deploying the entire treasury in a one-time buy, which could push the price to 90 USD instantly and then collapse back to the original level even lower. Therefore, we will oppose any aggressive large one-time purchases. For large-scale buying, we would only consider building low-price buy walls. Typically, this would not even require actual capital deployment, but if those price levels are reached, the treasury would realize significant gains.
  3. We do not intend for GMX to be delisted from CEXs. Given that it has not been delisted even now, maintaining trading activity through buybacks makes delisting even less likely. By choosing to operate on CEXs, the goal is to absorb all available supply and potentially even trigger a short squeeze. However, if GMX were ever to be unlisted from CEXs (which would only occur as a proactive choice in the future, when there is effectively no supply left to buy), the Treasury could then gradually redeploy liquidity back on-chain.
  4. So even if the endpoint is the same, this process would be smoother and easier for the community to accept. The ultimate objective is to regain pricing power. As long as pricing power resides on CEXs, we operate there. If pricing power returns on-chain, the Treasury can reallocate liquidity back on-chain and dynamically manage it to achieve the optimal outcome.
4 Likes

The selling pressure on GMX — could it be caused by the continuous maturation and subsequent selling of esGMX?

From https://dune.com/saulius/gmx-analytics, We can see that the total staked amount for GMX + esGMX is 8.08 million (as shown on Dune Analytics, e.g., ~8,089,121 total GMX + esGMX staked in recent data).
In contrast, the official GMX website/stats page shows only ~6.79 million for pure staked GMX, creating a gap of about 1.29 million tokens.

I don’t know if the curve statistics above are accurate.
I think the data from Dune still has a reasonable level of credibility worth referencing.
Moreover, the official GMX team should be able to know exactly how much esGMX will mature within the next year, so they should be able to estimate whether it is one of the main source of selling pressure.

thank you for the replies to this, i’m still concerned if there would be sufficient liquidity

e.g. for the GMX [GMX-USDC] pool on V2, there is about $300k in total of liquidity, which doesn’t seem to be very much

though i believe for a lot of tokens with larger market caps, the liquidity may not be high as well, so perhaps that amount is ok

but maybe it could also be considered to deposit some of the funds that would be withdrawn from the protocol owned Uniswap liquidity into the GMX V2 pool, e.g. $300k worth

the curve of the virtual order book on GMX V2 is determined by the price impact factors set by Chaos Labs, which in turn is dependent on the depth of liquidity on CEXes, so a larger GMX V2 pool does not directly lead to orders on the sell side, but having more liquidity especially protocol owned may help to give confidence

a consideration though is that if the price of GMX V2 increases, the GMX side of the pool would increase in value which leads to arbitrage to balance the GMX and USDC sides of the pool

so just sharing the thoughts and considerations, i don’t have too strong an opinion for this aspect

1 Like

We can see that the total staked amount for GMX + esGMX is 8.08 million

if the esGMX is staked then it is not being vested

for esGMX being vested, the amount that should be considered would be the amount in the Vester contracts

  1. Affiliate Vester (ARB) Address: 0x7c100c0F...A81df49B2 | Arbitrum One : 25k esGMX
  2. GMX Vester (ARB) Address: 0x199070DD...26B363004 | Arbitrum One : 60k esGMX
  3. GLP Vester (ARB) Address: 0xA75287d2...33972042E | Arbitrum One : 4k esGMX
  4. Affiliate Vester (AVAX): https://snowtrace.io/address/0x754eC029EF9926184b4CFDeA7756FbBAE7f326f7 : 316 esGMX
  5. GMX Vester (AVAX) https://snowtrace.io/address/0x472361d3cA5F49c8E633FB50385BfaD1e018b445 : 10k esGMX
  6. GLP Vester (AVAX) https://snowtrace.io/address/0x62331A7Bd1dfB3A7642B7db50B5509E57CA3154A : 215 esGMX

so in total about 95k esGMX, since esGMX has not been distributed to GMX and GLP for i believe more than 2 years now, likely most should have already been vested, so majority of already vested tokens may be dormant and just not withdraw yet

I am very surprised to see such a radical and aggressive proposal. It’s strange that so many people support it, yet my less aggressive proposal (to remove as much GMX from circulation as possible) still hasn’t received enough votes.

In short, I think the proposal is completely unrealistic and will certainly not work. There are countless reasons why the $90 price target will not be reached. In fact, given the current protocol activity, I would bet that achieving a price of even $30 seems unlikely. The reason is simple: many current 8M stakers will sell as soon as the price increases 2-3x. There is no way to prevent that. Even if onchain liquidity is withdrawn, I can still create a new LP position and put all my tokens singlesided for sale. Then arbitrage between DEX and CEX will take over.

While we are searching for other ways to support the GMX price, I ask everyone to reconsider and vote for my current proposal (still 1 day to go). This is an intermediate, temporary measure designed to achieve a similar target.

2 Likes

thanks for your detail investigation.

~ 95k esGMX have been deposited in the vesting vaults, so only around 260 GMX are released daily as rewards. It’s not a big deal. However, the much bigger impact and selling pressure will be after the vesting cycle ends and the reserved staked GMX is unblocked: for just 95k esGMX, the current locked reserve is ~800k staked GMX! Of course, not everyone will sell, but a significant proportion certainly will.

1 Like

Glad to see the giga brains working & leading.

Generally support the plan.

In the end, every stock has only one true buyer, the company itself.“

3 Likes

I understand the rationale behind a buy wall at these depressed levels to accrue gmx in the Treasury. Suspending staking rewards fully, I understand, too, but it is quite heavy-handed. A few important points I would consider while further working out this idea:

  • The GMX token being available on different venues is important exposure, particularly once the price trend reverses. Let’s avoid creating a situation where users who want to buy GMX effectively can’t. For example, because swap aggregators and in-wallet swap features can’t find meaningful liquidity for the token, because the liq is now all on GMX itself. (GMX’s swap markets are only integrated by OpenOcean)

  • Buybacks are highly effective when the token price is depressed, but exponentially lose effectiveness as the token price goes up. In this context, I would consider the prior discussion about buybacks happening only at certain key levels, or being based on HTF moving averages, etc.

  • Any set percentage (like 50%) to qualify for the reward distribution based on Staking Power is gameable, it seems to me.

  • As for suspending staking reward distributions until GMX hits its ATH of $90:

This ties in with a crucial question that, beneath the surface, seems to divide the existing community of tokenholders: should GMX primarily be a high-yield-earning token with a comparatively low value, or a strong value-accruing token that offers no yield?

3 Likes

Agree with Jonezee here, this proposal is extremely heavy handed.

Understand price is low, but this is trying to force ppl to sell to treasury for cheap maybe they are long term holders, which might backfire in the long run.

Really recommend to rationalize, focus on expanding trader base. If your trader who holds gmx sells due to this they might just leave the platform… dont force them to sell, u can set the buy wall naturally if it goes down it fills dont remove dex LP as that is a revenue stream

If this proposal goes through i can find myself supporting the staking pause but not the LP removal on chain, its negative impact to remove the DEX pools. Sincerely hope you reconsider on this aspect and everything else on this proposal i can support.

2 Likes

I guess the question here is not if prices are manipulated, but if a token with constant buying pressure in the form of BB&D works better in a low or high liquidity environment. If the latter we should actively reduce liquidity, if the former we should increase liquidity.

PS: What would be the intent of Binance to keep pushing prices lower after the first trading boom is gone?

its not the case imho.

Proposal assume that only protocol owned liquidity will be removed from onchain pools. There is still other onchain LP players.

So I am pretty sure that would be enough to support organic retail demand.