Thread by @EvanHoffmann7

Hey farmers, in this post I'm going to break down one of the most consistent and comfy bear market farming strategies and how to safely hedge. This position uses the combo power of @GMX_IO and @AaveAave to capture consistent real yield by attaining a near delta-neutral position.

A brief background on GLP: GLP is a structured product, similar to tricrypto with more stable exposure and higher yield. GLP is the house liquidity for GMX and its components hold pretty constant. This makes it very hedgeable which is extremely attractive in these volatile times.

Protocols like @UmamiFinance are working to deliver sleek automated products offering highly competitive yield on stable through perfect GLP delta neutral vaults using various derivative products and synthetic shorts to counter GLP’s volatile components.

The problem is, it’s hard to be perfect. GLP’s internal balance of BTC, ETH, UNI, LINK, & stable fluctuate day to day. That along with DeFi option protocols having certain strategy and liquidity constraints makes it hard to execute on a mass scale.

So rather than be perfect, one can be near-perfect and execute this strategy themselves.

When setting up a neutral position oneself -- and because of GLP’s small variations -- the aim isn’t to be 100% neutral, but to be 90%+ neutral and add one’s own slight long or short bias.

Let’s walk through an example from the beginning:

Anon has 1.5 wBTC they’re holding long-term and 20k USDC that’s waiting to DCA in 2023.

Step 1: Deposit this collateral in a money market protocol like Aave.

Step 2: Find GLP’s composition by checking the GMX Dashboard.

app.gmx.io/#/dashboard

GLP is currently 30% ETH, 18% BTC, 1% LINK, 1% UNI, 50% Stables. For this strategy, the LINK and UNI can be thought of as ETH and bump ETH up to 32%.

Step 3: Time to match the weight of GLP to the total amount to be borrowed with some quick math

*assuming 1BTC = $20k and 1 ETH = $1300*

Collateral = 1.5 wBTC + 20K USDC = $50K collateral.

$50k * (2/3rd loan to collateral ratio *rate can vary*) = $33.3k value to borrow.

Finding how much ETH to borrow:
33.3k * (0.32 *32% of GLP*) = $10666 of ETH.
$10666/$1300 = 8.205 ETH to borrow.

Finding how much BTC to borrow:
33.3k * (0.18 *18% of GLP*) = $6000 of wBTC.
$6000/$20000 = 0.3 BTC to borrow.

33.3k * (0.5 *50% of GLP*) = $16666 USDC to borrow.

8.205 ETH + 0.3 BTC + 16.6k USDC -> $26.6k worth of GLP

Here’s the thing; I know that in a few weeks GLP’s internal balance will have made some small fluctuations. Maybe ETH dominance drops from 30% to 26% and Stables are up to 54% total weight.

There will be slight impermanent loss. It is so hard to be perfectly neutral, even rebalancing daily, the ratio is in a slow but constant dance as traders open, close, win and lose.

Step 4: Here’s where one determines their conscious bias. If one wants to farm GLP for say a 6 month period and also believes that the markets will slowly bleed or at best go sideways over the duration, one can add an extra ~2% on both the ETH and BTC borrows and ~4% less stable.

This will makes the GLP position nearly hedged with a slight bearish bias. Conversely, if one thinks the market has probably bottomed but still wants to be moderately hedged, they can put an extra 5-10% stable into their GLP buys and a little less ETH and BTC.

Step 5: Once one has their 3 weighted positions, it is time to make 3 GLP buys with that ETH, BTC, & stable. Strategy complete. There is no need to do anything once GLP is in a wallet, it is now yielding ETH and esGMX. Time to sit back, relax, and collect the Ethereum drip.

There are many ways to execute this strategy. For instance, one doesn’t need to borrow USDC if they have liquid stable, they can use some or all of that directly for GLP’s stable weight.

The reason why collateralizing can be helpful is so one can use their fully borrowed GLP position to pay itself off and slowly *risk on*. It’s all about how one wants to mentally model the position.

I like to think of this as a little flywheel, using debt against my long-term holdings to have GLP purchase itself over time.
Really, this is a way to earn approximately 15-20% APY on collateral (wBTC, stable, other) at a 2/3rds borrow ratio.

If Arbitrum GLP continues to earn ~30% APY and we only look at the volatile assets, it will take about 18 months or so to pay off all the borrowed ETH and BTC, or a little over half that time to pay off just the ETH portion.

Seven months into this strategy, I’m slowly closing in on paying the borrowed ETH off and then will begin working on the wBTC.

I should mention there are risks running this strategy as with anything in DeFi.

The main risk I have my sights on is the potential outperformance of traders vs. the house (GLP holders). The most likely way traders profit is if the market enters a mega up-only trend, and I am planning to cut GLP exposure prior to the next full-run bull market.

I keep my eye on all the fees and figures here: stats.gmx.io

Shout-out to @coindataschool, they've been kind enough to share some really amazing data analysis on GLP and GMX, you have to check it out if you're serious about the protocol. coindataschool.substack.com

I hope this has helped if you’ve been curious about hedging strategies, thinking about risk, or #GMX and #GLP tactics. Stay tuned for future DeFi shares!