Rami@rami_poker
One of the biggest breakthroughs I have had lately regarding farming is the FRAME of my risk and reward vs APR.
Instead of thining in APR terms, framing my farms in EV terms. The only thing that matters is your return:risk ratio.
I think it's best understood with an example:
Think of syrupUSDT looped on Fluid. What are my risks?
1-Sryup has exposure to mainly 3 things:
a) Sky (syrup deposits a bunch on sky)
b) Aave (sky deposits itself on aave) and
c) overcollateralized (mostly) BTC/XRP loans
2-Fluid tech
Now comes the tricky part, where u have to be honest with yourself. It's impossible to ascertain risk, but u need to give a good honest, conservative assessment (if u have an error in estimating risk and reward, it should always be on the side of caution).
You can look into places like opencover, where they sell you insurance, as a base framework to help you understand and estimate risk; but i think they have tons of mispriced covers. I will not help them do their job better by all means, so won't mention any particular risk, but most risk overall are overpriced (i mean they have to make money i guess), and a few (that i have purchased in the past) clearly underpriced by magnitudes. One of my favorites was earning 200-250% net APR on extra finance while purchasing smart contract risk insurance for 5% lol
At that time Extra finance was relatively novel and untested, so 5% was very crazy price (esp when u were getting 40-50 times that).
But lets go back to the syrup fluid position and my guesstimates of risk:
Aave risk: 1.25%
Sky: 1.75%
Overcollateralized loans: 2.5% (afaik those are not onchain liquidations but more like loans to institutions which take some time to settle and therefore the higher risk. could be wrong)
This leaves us a 5.5% risk (0,9875×0,9825×0,975) that something goes wrong and we lose a chunk of our money.
syrupUSDT pays 3.8% currently, so the bet has negative EV if we were to lose all our money. we wouldn't, but holding syrupUSDT overall doesnt have a great expected value.
We can loop it on fluid, though, adding another layer of risk.
Fluid risk: 3.5% (source: i made that up)
Now we go from 5.5% risk to 9% risk (0,945×0,965).
The payout of this play? 18-23%
It's not that insane (the ratio of reward:risk is still only just above 2), but definitely better than holding naked syrupUSDT.
The single most important factor for a farmer's portfolio is the weighted-EV (or reward:risk ratio) of his positions.
Now, you obviously need it to be above 1 to make money. Many people are so bad that their positions barely have a number above 1. Add execution price, retarded prices at which they buy their stablecoins, in/out costs, etc. and they're clearly losing money over the long term.
But if it's close to 1, you're basically making a tiny profit for exposing yourself to huge variance.
So you need to reduce it by increasing the number of positions.
But if you increase the number of positions too much you're going to hurt your EV (not to mention the time and energy required to handle them).
So there is always a tradeoff between diversification and EV.
Now, I wrote a few days ago about WHEN to play in poker. It mattered as much as WHERE to play. And got to the conclusion that my hourly rate as a professional poker player could vary between 2-3x and even 5-10x from playing at the best possible times of the week/year vs the worst.
And I was analyzing my portfolio in peak bullmarket (early-mid 2024) vs good bull market (late 2024-late 2025) vs shit market (right now), and it's curious that my APR from peak bullmarket to shitmarket was up to 10x what is now, and from bullmarket to now is around 3-4x.
Which more or less coincides with the ratios from peak times in poker:
Christmas, or a nice saturday night at 1am when all u have is drunk people as competition would be early 2024
Normal weekend midday would be 2025, when things are still very good
And morning-midday on a normal weekday would be the equivalent of now: more regs, less fish, so hourly rates is even 1/10th of peak time.
(Just to put some stupid example, if we take something as retarded as Openeden $EDEN, the risk of farming it (exposure to tbill issuing, pendle and morpho) were probably under 10%, so even one of the most retarded farms from 2025 offered a 6-7:1 ratio. whereas the best farms we have rn in defi dont get close to that)
I also think this change of frame in analyzing yields is the best you can do for yourself. Stop worrying about APR so much, stop worrying if your portfolio doens't make X APR, and start thinking more in terms of EV: a 12% APR but extremely safe farm is much better than a 50% APR but super risky one.
This also allows to frame polymarket bets (polymarket.com/?r=RAMIPOKER thank you) in the same way as any yield farm. I was mentioning the BIBI BOND recently on twitter, in which you could get 48% APR (4% for the whole month) betting that the Polish president of satan's chosen people (oops did i just say that) would remain in the presidency until the end of April. the most likely cause for him not to be president would be:
+death. he was already confirmed alive by independent foreign press, and the odds of him dying from health issues are probably under 1% for the month, more like 0.5%. iran's odds of getting him are non-zero but very close to 0 imo
+political death. almost 0 in times of 'war', and elections won't come for a while
So this farm offered a lot better risk reward than the fluid syrup one (but still worse than the openeden one from 2025):
if we estimate the odds of bibi the genocidooor not making it at around 1%, we have an incredible 4:1 ratio, almost twice as better as the 2:1 ratio of the syrup venus farm.
Overall, as side note, i think there is a ton of alpha in polymarkets right now. There are way more retards per $ of TVL there than in any other defi protocol. This won't happen forever, but probably until $POLY airdrop things will remain very good.
And also, as another side note, we also gotta price the inherent risks of crypto. There are risks, like idk blackswan for Ethereum network, USDT going down, some that we havent even thought about.... you name it.
You could phrase it by saying that there is a hidden, fixed cost of having your money exposed to crypto, think of it as a negative 1-2% tax on your money lol
Which means two things:
1-if EVs are more or less similar between two farms, we should probably go for the highest APR farm cause the hidden tax of crypto affects the highest APR farm less instenively. It offsets the hidden tax of being a crypto participant better
2-If a farm has very shit apr, let's say sky at 3,75%, even if i estimate its risk on 1.75% (so EV above 2$ return for every 1$ risked), you should probably stay away from them and at that point offramp if you can.
So i guess the conclusion is thast very low apr farms (think below 4%) are shit and we should avoid them like the plague and just look to take our money to tradfi
And TLDR just to sum it up:
weighted-EV or risk adjusted returns of your portfolio is the single most important metric as a crypto participant and how you should frame your farming. Volatility and risk of ruin are still important metrics and you should aim to minimize both without sacrificing sa too much EV (and time). portfolio APR is not a very important metric, at least compared to EV. there is a cost, a hidden tax of doing business in crypto.
Hope this is helpful!
