Modeling DeFi IV: Multi-Strategies

What is a multi-strat?

Hello, again, avid crypto enthusiasts. Today I have decided to write about multi-strategies and why they may be beneficial. I will discuss how they work, provide a simple example, and reveal a simplified model which demonstrates the power of automation.

Simply put, a multi-strat refers to ‘one asset, many strategies. Corval sums it up very well in his article on Multi-Strategy Vaults and the Crypto Space Program. Instead of parking your holdings in one strategy and leaving them to the whims of the market, place them somewhere that will continually move them around for you to find yield opportunities.

Example: Spookyswap’s xBOO

The easiest way to demonstrate multi-strats is by looking at the staking options for xBOO. As TVLs shift and token prices fluctuate, the yield available to xBOO holders continually varies. Looking beyond a user who wants to farm a particular asset, you must constantly monitor xBOO pool APRs and move funds around accordingly if you’re going to maximize your yield. Imagine if you have a significant amount of capital. Anywhere you deposit your xBOO is likely to dilute the yield. If you see APR dropping and move your funds, you will deplete the yield of the next pool. A balanced approach allows you to split up your xBOO and maximize yield from all pools. Hence, you would ideally incorporate multiple strategies to maximize your returns.

Example: USDC

The same goes for USDC. Suppose Reaper Farm has a USDC crypt that loops funds in Geist for a specific APR and a USDC crypt that loops funds in Tarot for a specific APR. With a high enough capital, your funds are potentially best placed in both crypts, balanced in a way that maximizes yield from both, without needlessly diluting either. Managing this on your own can be extremely convoluted and tedious.

Simple xBOO Demonstration

I am a fan of xBOO, and going to Spookyswap and seeing all the pools and APRs in one place may be helpful for readers throughout the following explanation.

For a certain capital (TVL), an investment (emission) will translate to a yield (APR). We have seen this before in my previous articles, and to make it even simpler for this demonstration, we will treat the term as one year.

APR(%) = Investment($)/Capital($)

To demonstrate, I will pose some hypothetical returns. If $1mil worth of xBOO accrues $200,000 value of LQDR in one year, the APR is 20%. If $1mil worth of xBOO accrues $250,000 value of xTAROT in one year, the APR is 25%. So if you have another $1mil worth of xBOO, which pool should you put it in? The xTAROT pool has a higher yield, right?

Before
Capital ($)Investment ($)APR (%)
LQDR100000020000020
xTAROT100000025000025
Before
After
Capital ($)Investment ($)APR (%)
LQDR100000020000020
xTAROT200000025000012.5
After

Not anymore. The capital has diluted the yield, and now the LQDR pool has better returns. But how much of your $1mil should you move across to the LQDR pool? This is the underlying question of multi-strategy crypts. How much capital to put in each strategy, in order to maximise yields (safely).

Finding the Balance

For this example, we can model returns against one variable; how much capital we invest in the LQDR pool ($0-1mil). Naturally, the rest of our capital will be in the xTAROT pool so that all of our xBOO is earning if we have $100K in the LQDR pool, $900K in the xTAROT pool, and so on. Each spread results in a different effective multi-strat APR.

We can see that to maximize yield in the circumstances described; we would want around $400K of xBOO farming LQDR and about $600K of xBOO farming xTAROT, giving us an effective APR of slightly over 15%.

Real Multi-strategy Crypts

Individually, you or I may not have enough capital to dilute the yields of any given strategy. Still, when everyone pools their funds together, yield aggregators like Reaper Farm certainly do. This is compounded by price action, and fluctuating TVLs will constantly vary APRs. That is where the power of a multi-strat comes in handy. The ability to shift funds across strategies to maximize effective yield on behalf of the user is a potent and capital-efficient tool.

Beyond that, the example I have modeled is two-dimensional. I only gave two pool (strategy) options, which meant there was one independent (capital allocation) and one dependent (effective APR) variable. When you add more pools, the combinations and permutations of capital allocation gain more dimensions, and modeling multi-strats requires complex arrays beyond this article’s scope. Imagine the intricacies of the codes behind the automation of multi-strats.

Of course, there are plenty of additional considerations. Setting limits on the percentage of capital allocated to an individual strategy within the crypt may help protect user funds from exploitation and limit yield from that strategy. Depositing your assets into a multi-strat is highly tax efficient because funds are moved around in the back-end, and users have no tax implications. Still, there is no passive income for yield farmers or exposure to farmed tokens. These are all good and bad, depending on your needs and priorities. My only hope is that you now better appreciate the capabilities and limitations of multi-strategy crypts to make better decisions for yourself.

Thank you

Please let me know if this piece was useful (or if you think I made any mistakes). The example I gave was relatively simple but the underlying concepts apply to complex strategies as well. If you have any ideas (DeFi broadly) you want modeled to help explain or compare then please let me know and I’ll see what I can do.

Happy investing,

kickflip

Scroll to Top