# When does it make sense to single-stake a non-native token in a yield farm?

disclaimer: this post is a response to the discussion that followed the "If our liquidity is our voice..." post on reddit. I'll link to that original post so interested readers can glean helpful insight that I'm omitting.

##### TLDR: if you trust a farm to not rug pull, and they're offering a crazy high daily return on a non-native token pool, it may be profitable for you to stake there. But you need to be mindful of deposit-fees effectively locking you in for days before you even start to make interest. Otherwise, you'll want to consider whether there are any LPs with your non-native token that have a volume high enough to counter-balance IL.

So, as part of the discussion that's happened in this post I feel compelled to do a bit of a mathematical diversion to demonstrate compounding, as well as the actual value proposition for some of the pools these farms offer.

First, let's change the way we actually look at these financial products. Displaying an APY (or APR) is incredibly misleading for 2 reasons:

1. the farm most likely won't be around in 1 year, and even if it were

2. these rates certainly won't be stable for that year.

So, we're going to do all of our calculations based on a daily return (sometimes denoted as APRD). It's calculated by taking the APR and dividing it by 365. Now, I'm going to work through a scenario where you have \$IRON (or some other token) and you stake it in a farm's pool.

I found a farm that's offering an \$IRON pool with a daily return of 0.76% with a deposit fee of 4% (yikes!). Below is a screen shot of a simple spreadsheet which makes the assumptions that

• we're staking for 2 weeks,

• the APRD is stable during that period,

• the price of MATIC averages to around 1.45 during that period

I'd like to bring your attention to the peach colored line in the table. Since we get hit with a 4% deposit fee, we need to wait 6 days in this pool before our initial investment has been recouped. The remaining 8 days are "gravy".

The "optimal compounds per day" was found by optimizing the "net ROI" value highlighted in green. You need to be realistic when you play this game: harvest, swapping and depositing costs MATIC and therefore chips at your gains. If you compound too many times, you actually make less than if you were to just chill and actually sleep.

Let's say the entire idea of compounding every 3.5 hours isn't for you. What you probably want, then is to find a reputable AMM/DEX and see if they offer any LPs that have good enough volume to counter-balance IL. Everyone on DeFi needs to have a basic understanding of AMMs and IL, otherwise they're asking for losses, so I'm going to let you take a moment to go to youtube (or google) and look up "impermanent loss". When you're done, go to apy.vision and checkout their various pool statistics. Here's a screengrab:

Take note of the age of the pool, the 24H Volume, Reserve and 1y Fees/Reserves (this is effectively your expected return).

I took this screengrab for the MATIC pools, since I find farming on the ethereum network to be a rich man's game, and rich people don't usually read posts about ways to not lose their money. Let's talk about this:

1. We're all basically gamblers here, and the best gamblers have information on their side. The older a pool, the higher quality the information will be.

2. Volume is how we make money as Liquidity Providers (LPs). You get a cut of the fees dumped into the pool proportion to your contribution to the pool. So, if you are 0.001% of the pool, you get 0.001% of the fees. This is where volume is crucial. 0.001% of \$110,358.9 is \$1.10358. Realistically, if you're a retail trader you're not going to make up more than 1% of a high volume pool (and if you could, I don't think that's great risk management). If you decide you want dominate some more obscure pool, well if the volume is high enough, it might yield more than a more popular pool. Or, it might entitle you to, say 25% of \$0.03 which is garbage.

3. Impermanent Loss (IL) is very real and needs to be factored into your analysis. The reason we're going for high volume is because your LP fees need to be higher than your incurred IL. The important thing to remember with LPs paired against stable-coins: the IL is not pool specific. ETH/USDC, ETH/USDT, ETH/DAI, all of these pools are going to be hit with similar IL when ETH tears in either direction. This is why you don't want to be in some off-brand DEX somewhere, holding 25% of an LP that is commonly traded elsewhere. You're going to get burned by IL.

Bringing this full circle, you might notice that this option is not as profitable as the worked example above. Moreover, you might have noticed that plenty of farms and yield aggregators are offering nearly 1% daily returns on USDC-IRON LP tokens. In this case, I'd say that you should vet a project and if you trust them, probably stake the LP for yield aggregation. However, if we work through the same pool example, but now with ETH (I'm finding an APRD of about 0.19% via the same farm that was offering an awesome APRD on IRON):

Here, you'll see we need about 23 days to recoup our deposit fee. If you compare this to the USDC/WETH pool above, you'll see that while the highlighted return might feel meager, you'll start earning a return immediately and no one is going to ping you for withdrawing or depositing. So if you decide to liquidate your ETH position, you just leave the pool and do your business.

This post is getting long in the tooth. Leave your questions below and I'll try to answer them.