Sabitlenmiş Tweet
Ramil Amirov
108 posts
Ramil Amirov
@mcp0x
perp research @chaoslabs defi hip hop 📚 fully personal pov on crypto and stuff
Katılım Kasım 2023
131 Takip Edilen206 Takipçiler
thx!
from what I've seen, there have been mentions of this number
x.com/annanay/status…
and here - "Some exchanges (Binance, Bybit) have positive baseline funding rates of 11% annualized, meaning if funding is within a certain range it snaps back to 11% by default. Those exchanges make up over 50% of open interest. We can see below the impact positive baseline funding has to the distribution of funding on Binance and Bybit in particular"
x.com/ConorRyder/sta…
can't say that these guys are taking other side, but they do acknowledge this issue
regarding the literature, the best one I found was the paper by Ackerer, Hugonnier and Jermann (nber.org/papers/w32936), they derive the correct FR formula theoretically
ak0@annanay
To address some follow up directly: My guess is that Bitmex picked I =1bp and H=8 hours since the ‘fair’ I when they listed their XBTUSDT perp was around 10.95%, I=1bp was the lowest they could think of, and H = 8 hours made it 10.95%. These choices now continue to plague the industry 😅 If you enjoyed this, I’ll be highlighting some other design choices we’ve made for our tradfi-native perps.
English
hey ramil, nice work and interesting read!
do you know if this has been discussed before in the crypto literature, or even on ct? i am curious whether people are taking the other side, because otherwise this looks pretty self-evident to me.
unless, of course, venues are intentionally trying to nudge the market in one direction over the other 🤔
English
Longs on @binance and @HyperliquidX quietly overpaid $480 million in funding last year on $BTC and $ETH markets alone
Almost every major perp exchange copy-pasted BitMEX’s 2017 emergency setting… and "forgot" to touch it for nine straight years
Binance carved out an exception for their own token.
Here is why🧵
English
first, great idea! very curious to see the outcome of this experiment
second, I went through the submission form, and while the idea (add small agent to turn submission form to the dialog) was a fairly obvious one, I hadn't seen the actual implementation, so h/t to @cyberfund for creativity
godspeed
Konstantin Lomashuk cyber/acc@Lomashuk
When I was 22, my first CEO had one rule: Every time someone said “this is impossible,” they paid $5. Small thing. Rewired me for life. You start seeing how often people kill ideas before they try. How most limits live in the head, not in reality. “Impossible” usually just means “I don’t know how” or “I’m afraid.” I think about this every single day now. Because we’re already inside the singularity. Most people don’t feel it yet - the surface still looks normal. Underneath, everything is being rewritten. For the first time in history, one founder can ship what used to take a team of 50. Small teams are eating giants. And we’re still at the very beginning. The real bottleneck isn’t intelligence anymore. It’s speed of adaptation. Most founders already know what to do. They just can’t operate at AI-native speed. No focus. No rituals. Still running on pre-AI muscle. Still calling things impossible that already aren’t. The Monastery That’s why we built the digital Monastery. An environment for founders who want to fully adapt to this new reality. Extreme focus. High intensity. Small groups. Almost unhealthy commitment. People who compress years into months. People who operate with leverage that didn’t exist before. Not for everyone. By design. This is the best time and the last time to build. Best - because AI amplifies ambitious people and the world hasn’t caught up. Last - because intelligence is becoming a commodity. Soon everyone has the same tools. A short window before the world reorganizes. A window where small AI-native teams can create asymmetries that won’t exist again. The future belongs to those who move first. Do the impossible. While it still looks impossible.
English
come on, guys; start a heated argument with me, tell me where I'm wrong
Ramil Amirov@mcp0x
Longs on @binance and @HyperliquidX quietly overpaid $480 million in funding last year on $BTC and $ETH markets alone Almost every major perp exchange copy-pasted BitMEX’s 2017 emergency setting… and "forgot" to touch it for nine straight years Binance carved out an exception for their own token. Here is why🧵
English
English
Full paper: papers.ssrn.com/sol3/papers.cf…
$480M in 2025 on just two assets at two venues. The total across the entire industry over the last decade is in the tens of billions.
If you're a long trader, you've been paying part of it.
English
A few venues are starting to fix this for RWAs.
@Lighter_xyz differentiates by asset class: separate IR for FX, equities, commodities.
@tradexyz applies a x0.5 multiplier across all markets, explicitly citing SOFR in their docs.
@QFEX dropped both the constant and the clamp entirely. Formula is just the market premium.
None get it fully right. But they're trying. Most aren't.
And the deeper problem: even if an exchange tried to fix this, it wouldn't work. They'd just lose volume and feed arbitrageurs exploiting the funding gap between @binance and the fixed venue.
English
"Need obey" post compaction damn somehow this feels strange to read
English
spent some time reviewing @qfex
first of all, I want to say that these guys actually gave thought to their design, and this deserves huge respect, considering that funding rate design is something the perp industry decided to abandon
let's start with how they improved their design, despite @annanay's initial opinion that having a clamp in the funding rate formula helps by "this way short-term deviations around the fair will not contribute". although this is true, these deviations represent the fair rate, so why would you want to make it harder for the market to show its mood? in @qfex's version of the FR formula from February 26, as you can observe in the attached post, the clamp was present, but now it's dropped, which supports my claim that the clamp is a redundant part of FR design.
with this figured out, we can move on to some questions:
1. IR = 0
although @annanay acknowledges the consequences of setting an understated IR: "Note we kept interest I=0. Picking a value for I has little effect on the market, other than forcing the perp to trade away from the index if the fair market interest rate does not equal I. Since the rate changes over time, the exchange has no business deciding the rate. This is replicated across tradfi exchanges - eg, CME doesn't hardcode the S&P500 dividend yield into the ES mark formula," I can't agree with it. keeping IR = 0 is still picking some value, and zero is not correct in any case. with this treatment, you could choose basically any (incorrect) value for IR, since the only thing it influences is the deviation of perp price from spot price.
2. "At @QFEX, recognizing that tradfi funding should be lower than crypto"
I can't agree with this point, since the data and logic don't support his claim. from the data I analyzed, the fair IR for BTC is ~4.3%, for ETH ~2.1%, for commodities ~3.6%, for EUR/USD ~1.7%, for US equities ~0–3.6%. so, as you can see, there is no gap between crypto and tradfi assets; moreover, if we examine some "high-earning" crypto assets (like SOL, with higher APR for staking), then we should observe negative IR. however, this claim could be true in some sense, if we take crypto volatility into account (frequent price changes => frequent trade direction change => frequent premium change => volatile funding), though looking at the volatility of stocks and commodities this year, I could imagine that this doesn't make a strong argument :)
anyway, these guys made a huge improvement from the initial formula, so hat tip to them!
for vis: @hftgod @prnvbn @annanay @joshuawharton
ak0@annanay
The last explainer on the funding formula you'll ever need to read. Part of the privilege of designing for the tradfi-native space is getting to examine crypto assumptions and seeing whether they still make sense for tradfi perps. A ‘standard’ perps formula, eg: F = clamp (average_premium, 5ps, default_interest=1bp)/ 8, paid every hour. is probably the least understood thing in crypto, which is bizarre given the volumes that now flow through perps. The aim is to create a formula that allows: >the market to realize its ‘fair’ funding rate in a stable way; >the perp mark price to track the underlier closely. If you don’t know what the ‘fair’ funding is, there are good explainers online, but broadly the perps should pay out (riskfree_rate - dividend_yield)% annualized [for equities], or around 0-3.5% for most US stocks. If flow is very one-sided, the perp price should deviate from the fair, allowing participants to earn risk-free money as it does so, and incentivizing more capital deployment which should move it back in line. Let’s parameterize the formula, and decide how to pick the params: F = clamp(average_premium, C, I) / D, paid every H hours If we pick H=8 and D=1 like Bitmex, say, then an average premium of just 1bp would lead to a whopping 10.95% annualized rate! Given 1bp is around a tick, we can see that this is really unstable. We can counteract this by: >having a ‘clamp’, C. This way short-term deviations around the fair will not contribute. >picking a big value of D. However, if it’s too big, then the perp price will not match the index price for the market to realize its ‘fair’ funding. At @QFEX, recognizing that tradfi funding should be lower than crypto, we chose as attached. Note we kept interest I=0. Picking a value for I has little effect on the market, other than forcing the perp to trade away from the index if the fair market interest rate does not equal I. Since the rate changes over time, the exchange has no business deciding the rate. This is replicated across tradfi exchanges - eg, CME doesn’t hardcode the S&P500 dividend yield into the ES mark formula.
English
@0xturbanurban thanks for the kind words! yep, RWA perps are understudied as well, very interesting to see how various venues are gonna handle this
English
@mcp0x The full piece is quite well written. An interesting dimension worth considering: perps where mark price is a weighted basket referencing dated futures that have to be rolled (eg oil)...how you manage the roll (especially in steep backwardation) leads to some interesting trades.
English
kk, while @hyperliquidx community keeps being silent, we can move on to the next protocol - @ethena
ethena didn't do anything wrong actually, but they are one of the biggest beneficiaries of the hardcoded interest rate;
to be brief, the idea of the ethena is to long spot ETH/BTC, short the perp, collect funding
and honestly, there is nothing wrong with this, I even believe that ethena single-handedly improved market structure, by creating huge pressure on funding rates
however, if we deconstruct funding rate into parts, we would see that it consists of premium and interest rate. in the healthy market conditions mean premium over a long period should tend to zero, so the main source of income for sUSDe is interest rate, but it's elevated!
therefore, I have a couple of questions for @ethena team and their supporters:
1. how much of sUSDe's APY would you attribute to the hardcoded interest rate vs genuine long demand?
2. does Ethena have a view on whether/when exchanges will recalibrate?
3. if they did, what would the protocol APY look like at a fair funding regime?
@gdog97_ @litocoen @ConorRyder @n2ckchong @sshxbt @hosseeb
Ramil Amirov@mcp0x
English