Neeraj K. Agrawal@NeerajKA· 11 days ago

RT @binji_x: dear ethereum community, we need to fix and prevent this in order to both scale to, and protect the world. the next wave of apps will ask for a selfie and a passport page before you can even press play. that is a privacy leak waiting to happen and it does not help the apps, because storing millions of passports is a liability. remember the recent tea app breach when ids leaked? spotify does not want that headline. zero knowledge proofs give us a cleaner path. with one tap your wallet can show spotify only one fact: i am over 18 > no birth date > no address > no id number, nothing else. the math proves the claim while every other field stays hidden. everyone wins > users keep their personal data at home > spotify and every other platform cut storage cost and breach risk > regulators still get the age gate they want this is already possible with some of the zk snark tools that exist; what we are missing is polished ux that meets other companies where they are and loud education. ethereum fixes this. the world just needs to see it. proofs not passports.

binji@binji_x· 11 days ago

dear ethereum community, we need to fix and prevent this in order to both scale to, and protect the world. the next wave of apps will ask for a selfie and a passport page before you can even press play. that is a privacy leak waiting to happen and it does not help the apps, because storing millions of passports is a liability. remember the recent tea app breach when ids leaked? spotify does not want that headline. zero knowledge proofs give us a cleaner path. with one tap your wallet can show spotify only one fact: i am over 18 > no birth date > no address > no id number, nothing else. the math proves the claim while every other field stays hidden. everyone wins > users keep their personal data at home > spotify and every other platform cut storage cost and breach risk > regulators still get the age gate they want this is already possible with some of the zk snark tools that exist; what we are missing is polished ux that meets other companies where they are and loud education. ethereum fixes this. the world just needs to see it. proofs not passports.

OneKey@OneKeyHQ· 11 days ago

Becoming an LP on DEX: Why Does It Keep Making You Poorer? Did you know how many DEXs exist in crypto today? The answer is over 1,700 — far more than any other DeFi category. And what’s the total TVL of DEXs now? Around $2.3 billion. In comparison, lending protocols currently hold over $7.1 billion in TVL. Four years ago, DEXs peaked near $8 billion in TVL, while lending protocols had less than $5 billion. Clearly, being a DEX LP is no longer a “sure-win” strategy. You deposit two assets and provide liquidity for months, only to realize you’d have been better off just holding. The fees you earned are minimal, and impermanent loss (IL) just keeps stacking up. And don’t even mention Uniswap V3 — its active range setting is a high-level task most retail users can’t handle. You thought being an LP was about combining assets to earn trading fees — a seemingly safe, passive income. But once you understand the math and economics behind it, you may be shocked by the hidden costs of providing liquidity. This article dives into the heart of it. Being an LP Is Always a Tradeoff Between Risk and Return Providing liquidity isn’t a guaranteed way to make money. It’s essentially a permissionless collaboration: you put up assets, help facilitate trading, and in return, you earn certain rewards — but not without risk. While these are generally labeled as “rewards,” they actually come in various structures. So, what kinds of returns are we talking about? Broadly speaking, they fall into a few main categories: Trading fees: This is the most fundamental form of LP income, distributed proportionally among all liquidity providers in a pool based on trading volume. Every swap that goes through your pool contributes to this revenue. Protocol token incentives: Protocols often boost pool yields by distributing governance tokens. Through carefully designed emission curves or supply controls, they can artificially inflate returns to attract capital. Governance-based rewards and bribes: A more advanced layer of incentives can be seen in protocols like Curve and Velodrome, where veToken holders vote to direct emissions toward specific pools. As an LP, you can earn additional rewards by locking tokens and participating in this vote-directed flow of incentives. But what about the risks? Importantly, unless both assets in the LP are stablecoins, you’re inherently exposed to price risk. As market prices shift, your LP position rebalances accordingly, leading to changes in its overall value. It helps to think intuitively here: when you LP, you’re essentially market making. The take is — you’re doing it passively, which means your losses often come at the hands of more active players in the market. With So Many Rewards, Why Don’t LPs Make Money? The core reason behind LP underperformance is the market itself. Take a simple example: when ETH’s price drops, your LP position is effectively “buying the dip” — which sounds fine, until you realize it’s buying all the way down. As the value of your position depreciates, the supposed high APR becomes meaningless. Trading fees are uncertain; volatility is not. This is the origin of impermanent loss (IL). Impermanent loss is an inevitable cost for anyone providing liquidity on a DEX. It may sound abstract, so let’s use an example to illustrate how it works. 1. Initial state: Using the classic AMM formula x * y = k, suppose 1 ETH = 100 USDT. A user deposits 1 ETH and 100 USDT into the pool. The pool’s total size is 10 ETH and 1,000 USDT(total pool value: $2,000), meaning the user owns 10% of the pool. 2. Price change: ETH rises to 200 USDT. To maintain k constant, the pool rebalances by reducing ETH and increasing USDT. The pool now holds approximately 7.071 ETH and 1414.2 USDT. The total value is $2,828. Your 10% share is now worth about $282.8. 3. If you just held: If you held 1 ETH and 100 USDT instead of LPing, your portfolio would now be worth $300. 4. Impermanent loss: The IL here is 300 - 282.8 = 17.2 USDT. You didn’t lose money in absolute terms — your position grew from $200 to $282.8 — but you made less than you would have if you just held. That's an opportunity cost. So why call it “impermanent”? Because, if ETH returns to 100 USDT and you haven’t withdrawn yet, the loss disappears. IL only becomes realized when you withdraw at a price diverged from your entry. But IL isn’t the only problem. Even if prices return to the starting point, LPs can still lose out. Why? Because there are always sharper, faster players in the market. These active participants profit from volatility regardless of direction, whereas LPs — being passive — do not. This brings us to another core cost: Loss Versus Rebalancing (LVR). LVR is another key concept researchers use to assess LP cost. Let’s build on the IL example above to demonstrate how it works. 1. Initial state: As before, the user adds 1 ETH + 100 USDT into a DEX LP at 1 ETH = 100 USDT. 2. Price moves up and back down: ETH rises to 200 USDT, then falls back to 100. The pool rebalances throughout. In theory, the pool looks like it’s returned to its starting state. But not for you. 3. Rebalancing path comparison: After ETH hits 200 USDT, your LP position holds ~0.71 ETH and 141.5 USDT — meaning you sold ~0.29 ETH for 141.5 USDT total. But a rebalancing strategy that actively sells 0.29 ETH at 200 USDT would earn 58 USDT. So you’ve lost 16.5 USDT relative to the optimal rebalance. On the way back down: As ETH falls to 100, the LP effectively repurchases 0.29 ETH for 41.5 USDT. A smart rebalancer would buy the same 0.29 ETH at just 29 USDT. That’s another 12.5 USDT lost. 4. Total LVR: 16.5 + 12.5 = 29 USDT. This is the loss from not actively managing your position like a rebalancer — and instead relying on the DEX’s passive AMM mechanics. This illustrates that an arbitrageur, starting from the same position as an LP, can capture additional value simply by reacting faster to price changes between CEX and DEX. By continuously monitoring price movements and executing cross-venue trades, they are able to extract profit from the LP’s passive pricing exposure — even if market prices ultimately return to their original level. How to LP Smartly Now that we understand DEX LPing isn’t a “deposit-and-earn” model but rather a dynamic risk arena and a strategic composition game, the next question becomes clear: how do you become one of the few LPs who don’t lose money — or even turn a profit? Position Management: Modern DEX AMM designs have become increasingly sophisticated, but at a high level, they can still be categorized into two types: concentrated liquidity and non-concentrated liquidity. If you’re familiar with Uniswap, this roughly corresponds to the difference between V2 and V3. In concentrated liquidity models, you can deploy your funds within a specific price range. This can result in extremely high displayed APRs — but it also amplifies impermanent loss within that range. And once the price drifts too far outside your selected range, your liquidity effectively becomes idle — “just lying in the pool doing nothing,” generating no trading fees. That’s why choosing the right range is crucial for optimizing actual returns. For example, in Aerodrome’s WETH–USDC pool, different LP ranges produce wildly different APRs — from 300% down to 58%. This clearly illustrates how APR is highly sensitive to price volatility and range selection. Reward Structure: This is something we’ve already highlighted when discussing APR and APY. Whether it’s a lending or DEX protocol, understanding the mechanics behind yield — where the rewards come from and how they’re sustained — helps you make more rational LP decisions. High numbers often mask unsustainable incentives. Asset Quality: Price volatility isn’t always driven by normal market dynamics — sometimes it stems from problems with the asset itself. Take a USDT–USDC LP pair as an example. If USDT depegs due to a black swan event, traders will rush to dump their depreciating USDT into your pool, draining your valuable USDC. Arbitrageurs leave with profits; you, the LP, are left holding the bag. Mindset Shift: LPing on a DEX doesn’t always have to mean traditional market making. For example, memecoins often experience surging trading volume shortly after launch. If you believe a certain memecoin has already peaked in price but still sees high volume, it might be smart to LP — not just for exposure, but to earn fees that offset the volatility risk. Protocols like Meteora on Solana even allow single-sided LPing. In a SOL–USDC pool, for instance, you can choose to deposit only SOL. This setup essentially functions like a limit order, offering more flexible LP strategies tailored to active trading intentions. End For newcomers, being a DEX LP is undeniably a tough business. The mechanisms are complex, returns are volatile, and you face multiple layers of risk — from slippage to impermanent loss to MEV extraction. It’s far from the simple “deposit and earn” model many imagine. But precisely because it’s hard, LPing has never been treated as a “final form.” Instead, it’s become one of the most actively evolving areas in DeFi infrastructure. Today, we’re seeing more and more projects tackling these structural pain points head-on: Some combine LP with lending, allowing positions to earn both fees and interest to improve capital efficiency; Some design MEV-resistant AMMs that shield LPs from sandwich attacks and value extraction; Others explore automated impermanent loss hedging through structured strategies that protect LPs from price volatility. These solutions may still be in their early stages, but they all point to the same conclusion: DEX LPing is still evolving. Disclaimer: This content is for educational purposes only and does not constitute financial advice. DeFi protocols carry significant market and technical risks. Token prices and yields are highly volatile, and participating in DeFi may result in the loss of all invested capital. Always do your own research, understand the legal requirements in your jurisdiction, and evaluate risks carefully before getting involved.

BowTiedBull.eth - Read Pinned or NGMI@BowTiedBull· 11 days ago

Hilarious like math doesn't exist or something https://t.co/zW5SHfZhsz

Arpan Parikh, MD MBA 🧠@DrParikhMBA· 12 days ago

Nearly $1B has flowed into ambient AI scribes YTD 2025. I get the VC math. But here’s the provocative q: what measurable impact to *clinicians* and *patients* would we have seen if even a slice of that capital went to innovative care delivery, vs yet another scribe business?

MathWallet@MathWallet· 12 days ago

Math Wallet is going beyond ₿. https://t.co/z6YxVWdqjF

Rumpel Labs@RumpelLabs· 12 days ago

RT @phtevenstrong: I just bought some @0xHyperBeat points via @RumpelLabs's pHBEAT-1. They were at $0.20. That's an estimated FDV of $35M. @spectra_finance, e.g., values these points at $60M FDV. If/When Pendle comes to HyperEVM, we'll get even more confirming data points. Here's my math 🧵👇

Stephen | DeFi Dojo@phtevenstrong· 12 days ago

I just bought some @0xHyperBeat points via @RumpelLabs's pHBEAT-1. They were at $0.20. That's an estimated FDV of $35M. @spectra_finance, e.g., values these points at $60M FDV. If/When Pendle comes to HyperEVM, we'll get even more confirming data points. Here's my math 🧵👇

MathWallet@MathWallet· 12 days ago

MathWallet has listed the #BaseChain 's 'Hot Tokens' ! 🚀Explore new possibilities in the crypto world with @base @noicedotso @bankrbot @zora @KeetaNetwork ! $MATH https://t.co/stOVLPmI2g

Come-from-Beyond@c___f___b· 13 days ago

RT @FalaTerranova: Imagine a cryptocurrency designed to keep your transactions completely private, but now it's being controlled not by hackers, but by whoever offers more money to the participants. That's what's happening with @monero , and you need to understand why to protect your digital finances. Monero is a digital currency that prioritizes privacy, making it impossible to track who sends or receives funds. But now, it faces a different kind of attack. It's not a technical issue, like a virus or code flaw. Instead, it's an economic attack, where someone uses financial incentives to gain control over the network. The company behind this is called Qubic, and they're attracting miners – the people who validate transactions on Monero – with better payments. Qubic is managed by Sergey Ivancheglo, who helped create another cryptocurrency called IOTA. Their plan is simple: they want to control more than half of Monero's mining power, what we call 51% of the hashrate. They're not hiding it. They offer higher rewards, and miners join voluntarily. It's like they're buying the network's loyalty instead of forcing it. Let me explain how this works step by step. A miner connects to Qubic to mine Monero. Instead of getting paid in Monero, they receive Qubic tokens, called QUBIC. Qubic then sells the mined Monero on the market, uses the money to buy more QUBIC, and destroys some of those tokens, which drives up the price. With the higher price, miners profit more and attract even more participants. This way, Qubic's control grows day by day. There are no dirty tricks; it's just smart financial incentives. Why does this scare people? If Qubic reaches 51% of the hashrate, they can interfere with the network. They could ignore valid blocks from other miners, reject specific transactions, delay confirmations, or even change protocol rules. They're already suggesting people wait for 13 confirmations instead of the usual for Monero transactions, which shows the real risk. Imagine your transfers taking hours or getting blocked for no apparent reason. The cost to do this is surprisingly low. Monero spends about 130 thousand dollars per day on security, paying the miners. If someone offers just 10% more – around 7 to 10 thousand dollars daily – they can attract the majority. That's less than an ad on a crypto site, but enough to compromise the world's most private network. Monero is vulnerable for specific reasons. Its system allows mining with regular computers, using CPUs instead of expensive equipment. Miners switch pools easily chasing better profits. The block reward is fixed at 0.6 Monero, and network usage is moderate, with about 30 thousand transactions per day. Fees are low, which is good for users but leaves the economy weak. The cryptography is strong, but the economic incentives aren't. In the end, the network is for sale to the highest bidder. Qubic says they don't want to harm anyone. But intentions don't matter here. When power concentrates in few hands, risks of censorship or manipulation arise. This turns incentives into a new weapon against networks like this. The Monero community isn't standing still. Developers have alerted everyone and activated safety modes. Many recommend switching to decentralized pools, like P2Pool. Now, 13 confirmations have become standard for more protection. There are discussions about protocol updates, like hard forks, to distribute power better. And they're monitoring Qubic's growth closely. But many people still don't realize how serious it is. This marks a new phase for attacks on cryptocurrencies. It's no longer just about code flaws; it's about economics and social dynamics. Networks like Monero won't be destroyed by programmers, but by irresistible financial offers. It's a battle for attention, money, and processing power. What can you do now? Use wallets that prioritize privacy. Avoid centralized mining pools. If you mine, opt for P2Pool to keep things distributed. Ask exchanges to require at least 13 confirmations for Monero transactions. And spread the word – knowing about this helps the network survive. Monero represents one of the last refuges of privacy in the crypto world. If it's compromised this way, by buying incentives instead of breaking codes, the entire digital privacy structure gets shaken. Think about it: privacy isn't just a technical promise; it depends on a solid economic base. This situation teaches lessons for all networks that use proof of work, like Monero. Perfect math isn't enough. We need structures that resist financial pressures and encourage decentralization. Otherwise, the next "attack" will look like just a better deal. More reflections are emerging. Some compare this to past betrayals in the community, where getting too close led to disappointment. There are debates about whether Monero should stick with proof of work or explore options like proof of stake. In the end, it's a wake-up call: privacy networks need to evolve to defend themselves in a world where money speaks louder. And a curious note: figures like Sergey from Qubic remind us of others who nearly destroyed crypto ecosystems. Both cause chaos, one with money, the other with mining power. Same mess, different networks. In summary, Monero is under a real economic threat, but with quick actions, it can strengthen itself. Reflect on this: how do you protect something valuable like privacy when money tries to buy it? Stay alert, because this affects not just Monero, but the future of private cryptocurrencies.

AGI.Eth@ceobillionaire· 13 days ago

RT @iruletheworldmo: i’m not sure people grasp what super human level coding and math would mean. it’s very very close.

Garry Tan@garrytan· 13 days ago

RT @incitafusio: Apparently if you are parent that opposes a graduation requirement for pilot ethnic studies program for SFUSD, you are "privilege parents" complaining and using your energy and resources to "destroy curriculum which will teach kids to create a more just and equitable world" I highly doubt this SFUSD parent has ever lived through or escaped communism, where millions of people have died for false equity in China, Vietnam, North Korea, former Soviet Union, Eastern Europe, etc. You want to help oppressed groups? Help kids have opportunities for upward mobility - give them solid reading, writing, math and communication skills, while building their resilience so they can overcome whatever disappointments and injustice they may face in life. Teach them to focus on the future that they can still impact, while not making the same mistakes of the past.

Terranova@FalaTerranova· 13 days ago

Imagine a cryptocurrency designed to keep your transactions completely private, but now it's being controlled not by hackers, but by whoever offers more money to the participants. That's what's happening with @monero , and you need to understand why to protect your digital finances. Monero is a digital currency that prioritizes privacy, making it impossible to track who sends or receives funds. But now, it faces a different kind of attack. It's not a technical issue, like a virus or code flaw. Instead, it's an economic attack, where someone uses financial incentives to gain control over the network. The company behind this is called Qubic, and they're attracting miners – the people who validate transactions on Monero – with better payments. Qubic is managed by Sergey Ivancheglo, who helped create another cryptocurrency called IOTA. Their plan is simple: they want to control more than half of Monero's mining power, what we call 51% of the hashrate. They're not hiding it. They offer higher rewards, and miners join voluntarily. It's like they're buying the network's loyalty instead of forcing it. Let me explain how this works step by step. A miner connects to Qubic to mine Monero. Instead of getting paid in Monero, they receive Qubic tokens, called QUBIC. Qubic then sells the mined Monero on the market, uses the money to buy more QUBIC, and destroys some of those tokens, which drives up the price. With the higher price, miners profit more and attract even more participants. This way, Qubic's control grows day by day. There are no dirty tricks; it's just smart financial incentives. Why does this scare people? If Qubic reaches 51% of the hashrate, they can interfere with the network. They could ignore valid blocks from other miners, reject specific transactions, delay confirmations, or even change protocol rules. They're already suggesting people wait for 13 confirmations instead of the usual for Monero transactions, which shows the real risk. Imagine your transfers taking hours or getting blocked for no apparent reason. The cost to do this is surprisingly low. Monero spends about 130 thousand dollars per day on security, paying the miners. If someone offers just 10% more – around 7 to 10 thousand dollars daily – they can attract the majority. That's less than an ad on a crypto site, but enough to compromise the world's most private network. Monero is vulnerable for specific reasons. Its system allows mining with regular computers, using CPUs instead of expensive equipment. Miners switch pools easily chasing better profits. The block reward is fixed at 0.6 Monero, and network usage is moderate, with about 30 thousand transactions per day. Fees are low, which is good for users but leaves the economy weak. The cryptography is strong, but the economic incentives aren't. In the end, the network is for sale to the highest bidder. Qubic says they don't want to harm anyone. But intentions don't matter here. When power concentrates in few hands, risks of censorship or manipulation arise. This turns incentives into a new weapon against networks like this. The Monero community isn't standing still. Developers have alerted everyone and activated safety modes. Many recommend switching to decentralized pools, like P2Pool. Now, 13 confirmations have become standard for more protection. There are discussions about protocol updates, like hard forks, to distribute power better. And they're monitoring Qubic's growth closely. But many people still don't realize how serious it is. This marks a new phase for attacks on cryptocurrencies. It's no longer just about code flaws; it's about economics and social dynamics. Networks like Monero won't be destroyed by programmers, but by irresistible financial offers. It's a battle for attention, money, and processing power. What can you do now? Use wallets that prioritize privacy. Avoid centralized mining pools. If you mine, opt for P2Pool to keep things distributed. Ask exchanges to require at least 13 confirmations for Monero transactions. And spread the word – knowing about this helps the network survive. Monero represents one of the last refuges of privacy in the crypto world. If it's compromised this way, by buying incentives instead of breaking codes, the entire digital privacy structure gets shaken. Think about it: privacy isn't just a technical promise; it depends on a solid economic base. This situation teaches lessons for all networks that use proof of work, like Monero. Perfect math isn't enough. We need structures that resist financial pressures and encourage decentralization. Otherwise, the next "attack" will look like just a better deal. More reflections are emerging. Some compare this to past betrayals in the community, where getting too close led to disappointment. There are debates about whether Monero should stick with proof of work or explore options like proof of stake. In the end, it's a wake-up call: privacy networks need to evolve to defend themselves in a world where money speaks louder. And a curious note: figures like Sergey from Qubic remind us of others who nearly destroyed crypto ecosystems. Both cause chaos, one with money, the other with mining power. Same mess, different networks. In summary, Monero is under a real economic threat, but with quick actions, it can strengthen itself. Reflect on this: how do you protect something valuable like privacy when money tries to buy it? Stay alert, because this affects not just Monero, but the future of private cryptocurrencies.

Liz4SF@incitafusio· 13 days ago

Apparently if you are parent that opposes a graduation requirement for pilot ethnic studies program for SFUSD, you are "privilege parents" complaining and using your energy and resources to "destroy curriculum which will teach kids to create a more just and equitable world" I highly doubt this SFUSD parent has ever lived through or escaped communism, where millions of people have died for false equity in China, Vietnam, North Korea, former Soviet Union, Eastern Europe, etc. You want to help oppressed groups? Help kids have opportunities for upward mobility - give them solid reading, writing, math and communication skills, while building their resilience so they can overcome whatever disappointments and injustice they may face in life. Teach them to focus on the future that they can still impact, while not making the same mistakes of the past.

🍓🍓🍓@iruletheworldmo· 13 days ago

i’m not sure people grasp what super human level coding and math would mean. it’s very very close.

MathWallet@MathWallet· 13 days ago

MATH has used 20% of the income earned from Apr 1, 2025 to Jun 30, 2025 to repurchase and burn 118,521 $MATH . We have completed this round of MATH burn. You can check the burn records via the blockchain browser: https://t.co/WchBN7wqTp https://t.co/i1rV5lZ7y9

TinyTap EDU@TinyTapEDU· 13 days ago

The “summer slide” is real. Kids lose up to 40% of what they learned during the year! But thanks to TinyTap, we turned our summer into a learning adventure. They think it’s a game. I know it’s math, reading, and even emotion. Just sit back and watch them light up with confidence. That’s the kind of learning I’ll say yes to. What’s one topic you wish younger-you had explored, and what first step will you take today?

Bearly AI@bearlyai· 14 days ago

Google DeepMind’s Demis Hassabis thinks there is a 50% of AGI by 2030. Three ways to test for it: 1. Ensure general human-level intelligence across every domain (not “jagged intelligence”). 2. Back test to see if it can create novel insights (eg. give the model all human knowledge up to 1900 and see if it could produce Einstein’s theory of relativity). 3. Let top experts in 100s of field (eg. Terrence Tao in math) stress test for a month and decide if it reaches bar in specific domain. *** Full episode here: https://t.co/TJqlbS8Qxw

MathWallet@MathWallet· 14 days ago

🔥Latest Listed dApp MATH dApp Store: https://t.co/dntaVWyTqo @zkSwap_finance @polkaswap https://t.co/S1C4jXhwJu

Ben Smith@semaforben· 14 days ago

So sad to see that Tom Lehrer died. Musical and math prodigy and iconoclastic star of the conformist 1950s who totally rejected fame, one of the most interesing people I've ever written about https://t.co/DypWrNvkqt

Peter Brandt@PeterLBrandt· 14 days ago

RT @BindawaOfficial: This is a masterclass in capital efficiency and long-term thinking. Sim trading may build skill, but real trading builds wealth. When you understand the math, the choice is clear: control, freedom, tax advantages, and scalability all live on the real side. Respect to those who walk the walk.

ISMAIL BINDAWA 📊@BindawaOfficial· 14 days ago

This is a masterclass in capital efficiency and long-term thinking. Sim trading may build skill, but real trading builds wealth. When you understand the math, the choice is clear: control, freedom, tax advantages, and scalability all live on the real side. Respect to those who walk the walk.

Peter Brandt@PeterLBrandt· 14 days ago

Attention Sim for Profit traders This is not a slam X post, but one intended to educate People ask me why I do not trade Sim for Profit if I have skills as a trader Actually, the answer is really easy Stay with me -- you may need to do a little math yourself to understand the flow Let's say I trade a $1M futures account - real not Sim pretend Next, lets use an avg ROR of 40% (lower than my 40 yr avg) To rent a $1M Sim account would cost me 400 BPs per year With a real cash futures account I earn about 400 BPs of interest - right off the bat there is an 800 BP differential With actual futures account I have complete control over the treasury function of my capital, able to withdraw and deposit with complete discretion Also, with an actual account: --Complete control over trading frequency or lack of --No rules imposed by others restricting my strategy or tactics --Overnight swing trading is my style, not allowed by Sim firms --No chance my broker would go bust unless the CME went bust, and even then loss covered by insurance Profits of actual futures trading are subject to U.S. IRS 60/40 treatment. For me, a $400k gain in real futures incur a Form 1256 federal tax liability of $108k. In Sim payout at the ordinary rate, the tax would be $152k With U.S. tax differential and fee v. interest income considerations, my net net profit at 40% ROR in real futures is $372,000. With Sim for profit trading the exact same performance would yield net net profit of only $168,000 Would you rather work hard for $372k or $168k Thus, real futures trading, with all its other wonderful features, would profit a net net profit 2.2X that of Sim trading Doubtful? Work through the math yourself. Of course, all of this math depends on the fact I can trade profitably, which I have achieved in 37 out of 41 years Now, you are likely to say this is not a fair comparison with you because you do not have a $1MM account. Fair statement. But even if you can hobble together a $50k or $100k real account you can achieve much more long-term with real futures trading. Any further questions why I real trade vs. pretend trade? cc: @PeterLBrandt

LindyMan@PaulSkallas· 14 days ago

If english and history graduates commanded the same salary as math and science students you'd get a different type of student majoring in those subjects. The smart ones. That's essentially what happened in the medieval ages, the incentive structure rewarded classics and liberal arts. The priests had a lot of power. So did various courtesans in the royal court. Smart individuals flocked to monastic schools or early universities to study the "trivium" (grammar, logic, rhetoric) and "quadrivium" (arithmetic, geometry, music, astronomy), the core liberal arts. Figures like Thomas Aquinas or Peter Abelard were intellectual giants who thrived because theology/classics led to real power Intelligence follows incentives.

Filecoin@Filecoin· 14 days ago

Qwen3‑235B‑A22B‑Instruct‑2507 now runs fully local through @EternalAI_ and is stored on Filecoin. It skips thought chains, outputs direct responses, and performs well on reasoning, math, code, and multilingual tasks. The model supports a 256K context window.

Coin Bureau@coinbureau· 15 days ago

Trust math, not TradFi middlemen.

໊@ppyowna· 15 days ago

i hate the fact that math and science smart students are considered smarter than english and history smart students

GM VIETNAM (🌞,🇻🇳) | Vietnam Blockchain Week@gmvn_official· 16 days ago

Let’s welcome Trong Nguyen, Ph.D., Head of Research at KyberSwap, to GM Vietnam 2025! With a Ph.D. in Applied Math (Machine Learning) and deep expertise in cryptography, blockchain scalability, and security, he leads cutting-edge research that powers the next generation of decentralized finance. 🎤 Catch his insights live at GM Vietnam 2025! Register now: https://t.co/qIVWExh6Q0

MathWallet@MathWallet· 16 days ago

🔥 Time to save MORE! @BNBCHAIN ’s 0-fee stablecoin campaign is still ON 🎯 $Math

Miles Deutscher@milesdeutscher· 16 days ago

The most important skill (in the AI age) isn't math, writing, coding or physics. It's your capacity to learn fast. Your ability to iterate, adapt and act quicker than others.

AGI.Eth@ceobillionaire· 17 days ago

RT @lmthang: Right before #imo2025, together with colleagues from Mountain View, NYC, Singapore, etc, we all gathered at @GoogleDeepMind headquarter in London for our final push for IMO. I believe that week was when all magic happened! We put all individual recipes (that we figured out before) together and did a yolo run (with the compute that I had to beg various groups to loan) to train our most advanced Gemini model. We finished training 2 days before IMO :D That model achieved SOTA results, not just for math, but coding along with other reasoning tasks, unbelievable! That week was also when we figured all details to scale our Deep Think mode, alongside with other enhanced inference strategies. We finalized our runbooks that predetermined the configs to be used during the IMO days (so there will be no human intervention at all). And the rest is history. Thanks @CadeMetz and @nytimes for featuring that historic moment https://t.co/b2TnTByzJ3 The photo below was taken in London during that magical week. Tagging various team members in random order who worked extremely for that moment @YiTayML @theophaneweber @DawsenHwang @JiemingMao @jon_lee0 @zichengxu42 @vinayramasesh @mirrokni @blackhc @gjb_ai @g01na2 @LeiYu63 @NateKushman @jj_at_brown @quocleix @freeetext together with many others that I don't yet have X handles.

Morning Brew ☕️@MorningBrew· 17 days ago

1/ Wall Street's math wizards are scratching their heads as quant hedge funds bleed cash in a summer slump that defies explanation. Let's get into it. https://t.co/Mm96OwlCxj

Morning Brew ☕️@MorningBrew· 17 days ago

Incredible television: Powell checks Trump's math on Fed renovations right in front of reporters Could cut the tension with a knife https://t.co/jB5u5wo0UQ

Thang Luong@lmthang· 17 days ago

Right before #imo2025, together with colleagues from Mountain View, NYC, Singapore, etc, we all gathered at @GoogleDeepMind headquarter in London for our final push for IMO. I believe that week was when all magic happened! We put all individual recipes (that we figured out before) together and did a yolo run (with the compute that I had to beg various groups to loan) to train our most advanced Gemini model. We finished training 2 days before IMO :D That model achieved SOTA results, not just for math, but coding along with other reasoning tasks, unbelievable! That week was also when we figured all details to scale our Deep Think mode, alongside with other enhanced inference strategies. We finalized our runbooks that predetermined the configs to be used during the IMO days (so there will be no human intervention at all). And the rest is history. Thanks @CadeMetz and @nytimes for featuring that historic moment https://t.co/b2TnTByzJ3 The photo below was taken in London during that magical week. Tagging various team members in random order who worked extremely for that moment @YiTayML @theophaneweber @DawsenHwang @JiemingMao @jon_lee0 @zichengxu42 @vinayramasesh @mirrokni @blackhc @gjb_ai @g01na2 @LeiYu63 @NateKushman @jj_at_brown @quocleix @freeetext together with many others that I don't yet have X handles.

YO@yield· 17 days ago

A very compelling case for checking out our @pendle_fi YT-yoETH market. The math looks good. We certainly didn't hire intern for his looks.

MathWallet@MathWallet· 17 days ago

💳 UNLOCK CryptoPayments with MathPay! Now LIVE exclusively on @BNBCHAIN : ✅ Accept USDT, USDC, USD1 instantly ✅ Elevate checkout experience: Fast, smooth & secure 👇 Empower your business today: https://t.co/gKPWU4CajZ $MATH https://t.co/iAgE3tPoH8

TinyTap EDU@TinyTapEDU· 17 days ago

Screen time that actually teaches? 📚✨ Turns out we’re not the only ones excited about it! Simple Flow dropped a 1-minute review of TinyTap and how our games turn phonics, math, and social‑skills practice into pure play! https://t.co/YVjAk5Dag2

Justin Wu π@hackapreneur· 17 days ago

Anon Satoshi made everyone believe. This is the lore. No face, no ego, no founder to worship or cancel. Just code, math, and trustless truth. That’s why #Bitcoin became valuable. If anything it is the most OG meme coin to counter inflation and the dollar. Community owned & run. What do you think? 🤔

Arthur B.@ArthurB· 18 days ago

For what it's worth, I started a small quant co a few years ago, that will still happily hire IMO/IOI medalists and other math/cs wizards, as the singularity isn't quite upon us just yet.

Greg Osuri 🇺🇸 deAI Summer 2025@gregosuri· 18 days ago

Breakthrough #1: Low-Communication Optimizers Instead of "talk every step," new math lets GPUs think locally, then share occasionally. Bandwidth required drops 100-500× with <1% accuracy loss.

Ryan Selkis (d/acc) 🇺🇸@Selkis_2028· 18 days ago

RT @RealTheoWold: Microsoft: Record profits Record market cap Layoffs 9k domestic tech workers Meanwhile…Microsoft applies for 4,752 H-1B visas. @JDVance : “This math doesn’t add up.“

Decentraland@decentraland· 18 days ago

Time to flex those quick reflexes 🔥 Try your hand at Word Search, Math Kittens, and Whack-a-Frog, now ready to play in Decentraland https://t.co/RTZwVu3jn5

Theo Wold@RealTheoWold· 18 days ago

Microsoft: Record profits Record market cap Layoffs 9k domestic tech workers Meanwhile…Microsoft applies for 4,752 H-1B visas. @JDVance : “This math doesn’t add up.“

Cullen Roche@cullenroche· 18 days ago

Back of the napkin math: If the FFR was cut to 1.5% then 10 year term premium is typically about 1% which implies 2.5% 10 year. Average weighted maturity of govt debt is 6.5 years so avg rate of debt likely lands near 2.25%. Debt held by public is $30T so that delta is about $425B in annual interest expense. That's being VERY generous since a 1.5% overnight rate is absurd in an environment with 2.5% core inflation. We likely stoke inflation with such a move and have to raise rates....

Arthur B.@ArthurB· 19 days ago

RT @_Dave__White_: the openai IMO news hit me pretty heavy this weekend i'm still in the acute phase of the impact, i think i consider myself a professional mathematician (a characterization some actual professional mathematicians might take issue with, but my party my rules) and i don't think i can answer a single imo question ok, yes, imo is its own little athletic subsection of math for which i have not trained, etc. etc., but. if i meet someone in the wild who has an IMO gold, i immediately update to "this person is much better at math than i am" now a bunch of robots can do it. as someone who has a lot of their identity and their actual life built around "is good at math," it's a gut punch. it's a kind of dying. like, one day you discover you can talk to dogs. it's fun and interesting so you do it more, learning the intricacies of their language and their deepest customs. you learn other people are surprised by what you can do. you have never quite fit in, but you learn people appreciate your ability and want you around to help them. the dogs appreciate you too, the only biped who really gets it. you assemble for yourself a kind of belonging. then one day you wake up and the universal dog translator is for sale at walmart for $4.99 the IMO result isn't news, exactly. in fact, if you look at the METR agent task length over time plot, i think agents being able to solve ~ 1.5 hour problems is coming right on time. so in some way we should not be surprised. and indeed, it appears multiple companies have achieved the same result. it's just... the rising tide rising as fast as it has been rising of course, grief for my personal identity as a mathematician (and/or productive member of society) is the smallest part of this story multiply that grief out by *every* mathematician, by every coder, maybe every knowledge worker, every artist... over the next few years... it's a slightly bigger story and of course, beyond that, there is the fear of actual death, which perhaps i'll go into more later. this package -- grief for relevance, grief for life, grief for what i have known -- isn't unique to the ai age or anything like that. i think it is a standard thing as one appreaches end of career or end of life. it just might be that that is coming a bit sooner for many of us, all at once. i wonder if we are ready

tunez (evm/acc) 🤖@cryptunez· 19 days ago

Mark Zuckerberg dropped out of college I dropped out of middle school You do the math

WealthSquad Chris 🌴@CJ_Johnson17th· 19 days ago

RT @yrn_jo: The amount of people who don’t understand Bitcoin or basic math is alarming lol. Shit coins are playing catch up that’s not something to celebrate….

Mansa 👑@yrn_jo· 19 days ago

The amount of people who don’t understand Bitcoin or basic math is alarming lol. Shit coins are playing catch up that’s not something to celebrate….

Harvest@harvest_finance· 19 days ago

honest take from Harvest's Twitter admin over the past few years, we've transformed Harvest into a yield powerhouse, with only the old-school tractor remaining on the surface • Pristine security track record for over 4 years now • Top yield sources on @base • New product line: Autopilots - high-frequency reallocation vaults in tandem with math wizards of @ipor_io • $10M to $45M TVL in the past year • Unmatched yield tracking precision - past, current, future data and outlooks • Lifetime yield accrued by your wallet, down to a single decimal, reflected by most major fiat currencies - USD, CHF, EUR, GBP, YEN • Clear and transparent benchmarks of all our products - no performance guesswork • Home to one of DeFi's longest-operating vaults • In-depth vaults audit in 2025 with @HalbornSecurity • Near 24/7 support from real humans side note: since the introduction of precise earnings info boxes and tabs in the Harvest App about a year ago, we have literally stopped receiving user tickets in that regard Harvest users know precisely what they get to earn, at what rate, and how frequently the newly accrued yield rolls into their wallet still here, still building for better DeFi as a DAO onwards

Arrow Markets | $BTC & $ETH Options🔺@ArrowMarkets· 19 days ago

RT @ArrowMarkets: What does a flat 20% rebate actually mean when you trade options? Is it enough to flip the math in your favor? The answer is an astounding yes. Let’s break it down with real examples 🧵👇 https://t.co/4JazLakuDr

Dave White@_Dave__White_· 20 days ago

the openai IMO news hit me pretty heavy this weekend i'm still in the acute phase of the impact, i think i consider myself a professional mathematician (a characterization some actual professional mathematicians might take issue with, but my party my rules) and i don't think i can answer a single imo question ok, yes, imo is its own little athletic subsection of math for which i have not trained, etc. etc., but. if i meet someone in the wild who has an IMO gold, i immediately update to "this person is much better at math than i am" now a bunch of robots can do it. as someone who has a lot of their identity and their actual life built around "is good at math," it's a gut punch. it's a kind of dying. like, one day you discover you can talk to dogs. it's fun and interesting so you do it more, learning the intricacies of their language and their deepest customs. you learn other people are surprised by what you can do. you have never quite fit in, but you learn people appreciate your ability and want you around to help them. the dogs appreciate you too, the only biped who really gets it. you assemble for yourself a kind of belonging. then one day you wake up and the universal dog translator is for sale at walmart for $4.99 the IMO result isn't news, exactly. in fact, if you look at the METR agent task length over time plot, i think agents being able to solve ~ 1.5 hour problems is coming right on time. so in some way we should not be surprised. and indeed, it appears multiple companies have achieved the same result. it's just... the rising tide rising as fast as it has been rising of course, grief for my personal identity as a mathematician (and/or productive member of society) is the smallest part of this story multiply that grief out by *every* mathematician, by every coder, maybe every knowledge worker, every artist... over the next few years... it's a slightly bigger story and of course, beyond that, there is the fear of actual death, which perhaps i'll go into more later. this package -- grief for relevance, grief for life, grief for what i have known -- isn't unique to the ai age or anything like that. i think it is a standard thing as one appreaches end of career or end of life. it just might be that that is coming a bit sooner for many of us, all at once. i wonder if we are ready

VentureBeat@VentureBeat· 20 days ago

Google DeepMind makes AI history with gold medal win at world's toughest math competition https://t.co/aursFjoEK7

Garry Tan@garrytan· 20 days ago

KQED's poor coverage of CA's new math framework was a miscarriage of journalism—refusal to engage experts like UC Berkeley EECS chair Jelani Nelson endangers student futures Evidence & transparency were ignored, harming kids statewide KQED's newsroom needs reform.

MathWallet@MathWallet· 21 days ago

🔥Latest Listed dApp MATH dApp Store: https://t.co/dntaVWzrfW @weRoamxyz @TopCutFinance @kinetiq_xyz @prjx_hl @Ostrich_HQ @HarvestFlow_io @BlackholeDex @paretocredit @cobaltx_io @xStocksFi @FullSailFi https://t.co/W3HYrEiAmX

SingularityNET@SingularityNET· 21 days ago

RT @bengoertzel: I pretty much agree w/ Gary's take on this. Seems very impressive to have done this well on Math Olympiad stuff w/o reference to an external math engine, Internet lookups etc. However without knowing more about what tricks were used (or not used) to get the results, it's not clear HOW impressive exactly this is... It's also not clear whether this indicates any kind of progress toward AGI. My gut feel is probably not but depending on what actually is going on under the hood, it's certainly *possible*...

Arrow Markets | $BTC & $ETH Options🔺@ArrowMarkets· 21 days ago

5/ 💡 Whether you’re hedging, swinging, or farming vol— 20% rebate means your downside is softer, and your upside gets a tailwind. Boost Rewards = flipping the math in your favor. 📅 Phase 1 starts July 21. Reward pool is capped so first come, first stacked .🏹🔥

Arrow Markets | $BTC & $ETH Options🔺@ArrowMarkets· 21 days ago

What does a flat 20% rebate actually mean when you trade options? Is it enough to flip the math in your favor? The answer is an astounding yes. Let’s break it down with real examples 🧵👇 https://t.co/4JazLakuDr

Ben Goertzel@bengoertzel· 21 days ago

I pretty much agree w/ Gary's take on this. Seems very impressive to have done this well on Math Olympiad stuff w/o reference to an external math engine, Internet lookups etc. However without knowing more about what tricks were used (or not used) to get the results, it's not clear HOW impressive exactly this is... It's also not clear whether this indicates any kind of progress toward AGI. My gut feel is probably not but depending on what actually is going on under the hood, it's certainly *possible*...

Ernest Ryu@ErnestRyu· 22 days ago

7. However, LLMs will become exceedingly powerful for problems that *someone* knows how to solve (in-distribution, in training data). In math research, you combine existing techniques with new creative ideas. LLMs will significantly accelerate the former part. (7/10)

Gary Marcus@GaryMarcus· 22 days ago

Hot take on OpenAI’s IMO gold What does it mean? I don’t know (yet). The fact that no tools, coding or internet was used is genuinely impressive. That said, my overall impression is that OpenAI has told us the result, but not how it was achieved. That leaves me with many questions, assuming that the result — which has not been independently validated by the IMO – is legitimate. • Architecture and training. The model that got the Gold did much better than any publicly available model (Though DeepMind is rumors to also have a new, similar result). Why is the new system better? How does it differ from previous models in terms of architecture, training, data augmentation, etc? It is described only as a new experimental model using “new experimental general-purpose techniques” that applies lots of inference time, which tells us relatively little. • Generality. How *general* is the new result? How well does it work on other kinds of math problems? On scientific questions in general? On things that have nothing to do with science or math? What is the scope? • Does this solution come with costs along other dimensions? (e.g., o3 hallucinates more than o1)? For example, it is already clear from the github that the output language is substantially less idiomatic than we see in typical LLMs (“Need sharpness: show can't be less.”; “Need show also all terms multiple of 3”, “So sufficiency done”). • Compatibility: Can this new approach be used with tools and coding and web search, where those are needed in other domains? (The weird language already suggests some concerns; one would not use the same system to write news articles). • Comprehensibility of output and validity of test score. A CS student in an online forum reported “[The] OpenAI proofs are very unreadable. I can't understand them. So the score that the AI gets depends on how charitable the grader is.” • Utility, even in math. As ICL mathematician Kevin Buzzard wrote, “it is still entirely unclear whether things will scale from machines being able to do mathematics which can be solved using high school techniques to machines being able to help with mathematics which can only be solved by having a deep understanding of modern research idea”. • Methodology, transparency and openness: How do we as a field evaluate and regard models that are not – and will not be anytime soon– available for outside review? • How much preprocessing was done and is necessary? The github files contain “Problem Submissions” that don’t seem to be in conventional English. I can’t follow what’s happening here. [For example, per the github, problem 3 is “\[P_f(a,b):\qquad  f(a)\mid b^a - f(b)^{\,f(a)}, \ \forall a,b\ge1.\]. Need universal linear majorant constant over all such functions.”] Who did this translation? Person? Machine? Were the problems be solved from conventional raw English descriptions? Or did a human translate into some sort of notation? • Economics: What was the cost per problem, both for inference time (clearly high) and any domain-specific expertise required for training/augmentation etc and what would the economics of using this model be when it is released? The result is obviously impressive; what it actually means remains to be seen. – With thanks to Ernest Davis for helpful discussion and initial thoughts, some incorporated above. Errors are mine.

Red Panda Mining@RedPandaMining· 22 days ago

RT @danbulmer70: OMG OMG OMG... Just did the Math. I don't have this much relatives! THANK YOU SO MUCH. My videos suck because I had a stroke and my body is saying to my brain.... Hold on buckshot.... Let's think this through and my new brain thinks it's invincible and worthy of an Oscar or some shit.... Again thanks. I"m on disability at the moment and.... If anyone was a first year apprentice anywhere.... Yeah it's that but not guaranteed so.... Trying to support myself with what I'm left with after a stroke. #Pron is definately out..... ;)

Mike Dudas@mdudas· 22 days ago

RT @mdudas: The Chads at @1kxnetwork invested in the @pudgypenguins CTO in 2022. The Chads at @1kxnetwork invested in the @moonbirds CTO in 2025. I’m a simple man who likes simple math.

Dan Bulmer@danbulmer70· 22 days ago

OMG OMG OMG... Just did the Math. I don't have this much relatives! THANK YOU SO MUCH. My videos suck because I had a stroke and my body is saying to my brain.... Hold on buckshot.... Let's think this through and my new brain thinks it's invincible and worthy of an Oscar or some shit.... Again thanks. I"m on disability at the moment and.... If anyone was a first year apprentice anywhere.... Yeah it's that but not guaranteed so.... Trying to support myself with what I'm left with after a stroke. #Pron is definately out..... ;)

Mike Dudas@mdudas· 23 days ago

The Chads at @1kxnetwork invested in the @pudgypenguins CTO in 2022. The Chads at @1kxnetwork invested in the @moonbirds CTO in 2025. I’m a simple man who likes simple math.

The New York Times@nytimes· 25 days ago

RT @anniekarni: .⁦@RoKhanna⁩ - “The math is obvious, we’re going to get a vote on the full release of these Epstein files. An overwhelming vote of Congress is not something the president can ignore, politically.” https://t.co/ht3CuOGFoG

sqd.ai@helloSQD· 25 days ago

SQD: Our 5-Year Evolution Everyone always asks when SQD will moon, but no one asks how did SQD even start? Like any big journey, it started with a niche obsession by one of the protagonists.  Act 1: The Meeting.  Despite much of his time being consumed by research for his PhD studies, @dizhel always found time to spend on crypto forums, quickly realizing the potential of the technology. He was convinced to the point of being willing to quit his math PhD.  In the end, he didn’t, but became a Messari community analyst on the side. After traveling to DAppCon, one encounter led to another, and after an evening on the roofs over Berlin, he met @i_1337_i, who ended up offering him a job.  Act 2: Trials and Tribulations  Unfortunately, this first collaboration did not result in a revolution but was torn apart by another crypto winter, leading our Co-Founders to part ways.  Undeterred by the cold, both remained in the crypto space. Dmitry joined a video streaming startup in the Polkadot ecosystem and, as part of the indexing team, quickly stumbled across a major problem: accessing blockchain data is hard.  His MVP wins the infra track at the Web3 Foundation hackathon. After reconnecting with MF, they both officially established Subsquid Labs, raising their seed round with 80% marketshare in the Polkadot ecosystem and an ambitious whitepaper powered by hard math.  Act 3: From Underdog to Established Since establishing the company, SQD (formerly Subsquid) has been on a steady growth trajectory, expanding the engineering team, increasing the network size, and gradually establishing ourselves as a serious contender in the blockchain data access space.  We went from just a little squid to having partnered with the Deutsche Telekom, integrations with Google Cloud, and powering the backend of PancakeSwap and other million-user apps.  But we’re far from done.

Annie Karni@anniekarni· 25 days ago

.⁦@RoKhanna⁩ - “The math is obvious, we’re going to get a vote on the full release of these Epstein files. An overwhelming vote of Congress is not something the president can ignore, politically.” https://t.co/ht3CuOGFoG

TinyTap EDU@TinyTapEDU· 25 days ago

Start your engines... All roads with TinyTap lead to curiosity! From early reading and math skills to fun facts about the world, give them TinyTap, and watch them grow while they play. It’s screen time everyone can feel good about. https://t.co/gmA3US6bXz

OneKey@OneKeyHQ· 25 days ago

APR vs APY — Don’t Let These Numbers Fool You "10% APR" vs "10% APY" — same number, totally different story. Some protocols use shiny APY figures to sell you a dream that never pays out. To most beginners, yield is yield. But APR and APY are built on entirely different assumptions — and mixing them up can cost you. What APR and APY Actually Mean APR (Annual Percentage Rate) shows your annual yield without compounding — it assumes you never reinvest your rewards. Let’s say you deposit $1,000 into a USDT/USDC pool showing 20% APR. If you do nothing — no reinvesting, no compounding — you’ll earn $200 after one year. That’s simple interest. But what if you keep reinvesting the rewards? Re-depositing fees, token incentives, all of it? That’s where APY (Annual Percentage Yield) comes in. It calculates your return assuming you compound consistently — earning interest on your interest. Let’s walk through the math with an example 👇： Without compounding: Profit = Principal × APR $1,000 at 20% APR → $200 after one year With compounding: Profit = Principal × APY APY = (1 + APR / n)^n - 1, where n is how often you compound per year. If you auto-compound daily (n = 365), then 20% APR → ~22.13% APY, or $221.30 in a year. That’s an extra $21 — just from reinvesting regularly. What They Don’t Tell You About Yield Understanding APR and APY is just the beginning. Knowing the terms doesn’t mean you’re safe. In DeFi protocols, clever design choices and hidden mechanics can still leave you with less than what you thought you’d earn. Let’s go over a few things most people overlook 👇 1/ High numbers don’t always mean high returns: Whether it’s APR or APY, these are often estimates based on past performance. Actual returns can drop fast due to things like declining pool activity, token price depreciation, or diluted rewards. 2/ You need to understand where the yield comes from: Some protocols help by breaking down the components of APR/APY directly in the UI. But in most cases, your first step should be clicking the "Docs" button — and digging. Look for the fine print. That’s where the real story usually hides. Here’s a real example from Kamino Finance: two SOL lending vaults — "MEV Capital SOL" and "Allez SOL". At first glance, "Allez SOL" looks better with an APY of 8.54%, compared to 7.37% from "MEV Capital". But when you check the breakdown, the actual "Lending APY" — the yield from real borrowing activity — tells a different story: > MEV Capital SOL: 7.35% > Allez SOL: only 5.1% The higher total APY on "Allez SOL" comes from extra token incentives added on top. That might look great now — but it also comes with risk: token price drops, reward dilution, and less sustainable yield in the long run. Objectively, neither pool is strictly better than the other. It’s not about chasing the highest number — it’s about understanding where the yield comes from, and choosing the one that aligns with your risk profile. 3/ When Fees Eat Your Yield: When manually compounding, don’t forget to factor in gas fees, swap fees, and other costs — especially if you’re working with a small bag. Let’s say you are farming a 9% APR vault. You might yield about $2.25 on a $100 deposit over 3 months. But if you swapped or bridged tokens before entering, and gas was high on Ethereum, you could lose $1+ in fees — nearly half your profits gone. With smaller deposits, fees eat APY fast. 4/ When APR Pretends to Be APY Some protocols run rewards using APR, but display them as APY — making yields look better than they are. That’s not a rounding error. It’s a red flag, especially in new ecosystems where low-quality protocols mix with legit ones. That's when you need to judge a protocol not just by numbers, but by product quality, security details, and whether the team actually seems trustworthy. Throwback to the airdrop farming era: On ZKsync, before $ZK was launched, a lending protocol called Era Lend became a hotspot thanks to sky-high APYs and endless farming tutorials. The result? Millions were lost to an exploit. The team went silent. No compensation. Many still believe it was a pure rug. In DeFi, while you’re chasing yields TradFi can’t offer, someone might be chasing your principal. End By now, you should have a clear understanding of what APR and APY really mean — and how yield in DeFi isn’t always what it seems. But remember: knowing these concepts is just a start. Making good decisions in DeFi takes more than that — it takes context, curiosity, and caution. Disclaimer: This content is for educational purposes only and does not constitute financial advice. DeFi protocols carry significant market and technical risks. Token prices and yields are highly volatile, and participating in DeFi may result in the loss of all invested capital. Always do your own research, understand the legal requirements in your jurisdiction, and evaluate risks carefully before getting involved.

Saito Network 🟥@SaitoOfficial· 25 days ago

Core Node ⚙️ Update Summary ♦️ Browser Theme fallback logic adjusted to default to ‘lite’ when no preference is set. Removed unused React container rendering method. Clipboard and file input logic cleaned and simplified. Improved theme icon handling, now safer with try/catch around class assignment. Console debug lines trimmed, old commented debug removed. ♦️ Wallet & Crypto Type annotations for returnInstalledCryptos and returnActivatedCryptos cleaned up. Fixed type to use typeof CryptoModule[] rather than tuple. ♦️ QRCode / Helpers Large refactor for QR code table drawing: merged duplicated paths, better margin handling, removed legacy comments. Simplified drawImage polyfill for older Android support. ♦️ Game Template / Game Queue / Game UI Console debug, trace, and info logs removed or commented out for cleaner output. Removed legacy sanity check for card decryption in RESOLVEDEAL. Simplified conditional checks and math (e.g., resetDeck keeps old xor if present). Better queue initialization and pending TX handling cleanup. Updated move notification sound logic to catch errors if autoplay is blocked. Small CSS and UI adjustments in game help and overlays. ♦️ Stream Capturer Refactored captureStream and mozCaptureStream checks to be shorter and clearer. ♦️ Profile Refactored description handling: empty state now shows “Add description”, and edit button shows consistently. Banner and profile image logic simplified. Removed direct console logs. ♦️ UI / Overlays Simplified show/hide logic in overlays and loaders. Removed or merged redundant DOM update lines. Cleaned up innerHTML assignments to be clearer and shorter. ♦️ Arcade Cleaned and shortened many event handlers, removed old commented debug. Refactored peer service up logic for offline games. Improved console debug style, replaced info logs with warn or debug. Updated theme options, removed unused theme icon. General tidy-up across invite handling and game status transitions. ♦️ ModTemplate Theme option ‘arcade’ line commented out as not used.

teller.eth@useteller· 26 days ago

RT @defi_mago: Farming USDC in a short period on @useteller is currently giving me the best experiences and yield! I'm getting 65.71% APY just for supplying USDC, no leverage. → The best part is its time-based loans, just 1 month, and my earnings are locked in right when the loan gets taken. → No APY change halfway through. →UI even shows me the numbers super clearly. ─────── For example: If you drop 1000 USDC into the USDC-MASA pool, you’re getting: → 56.96% Base APY → 8.75% Reward APY = 65.71% total, fixed So over a 1-month term, that’s $54.66 in yield. Easy math. ─────── And if you wanna borrow? It works the same way. Say you borrow $1K worth of MASA, it straight up tells you you'll pay: → $60.82 interest → $20.08 in fees No surprise APR changes. ─────── Why this setup is so nice: 👉 I don’t have to touch anything. After 1 month, it auto-settles. No need to time exits or micro-manage positions. When to use this strat: > Got some spare stables and want to park them short-term (<1 month) with solid, fixed yield > Wanna borrow ETH/BTC to farm points or airdrops short-term and know exactly your cost upfront > Or even short tokens that are borrowable on Teller ─────── Bonus alpha: Teller raised like $8M from legit funds. They’ve got a points system + rankings, so there’s a chance you get extra rewards down the line, not just only APY.

DeFi Mago@defi_mago· 26 days ago

Farming USDC in a short period on @useteller is currently giving me the best experiences and yield! I'm getting 65.71% APY just for supplying USDC, no leverage. → The best part is its time-based loans, just 1 month, and my earnings are locked in right when the loan gets taken. → No APY change halfway through. →UI even shows me the numbers super clearly. ─────── For example: If you drop 1000 USDC into the USDC-MASA pool, you’re getting: → 56.96% Base APY → 8.75% Reward APY = 65.71% total, fixed So over a 1-month term, that’s $54.66 in yield. Easy math. ─────── And if you wanna borrow? It works the same way. Say you borrow $1K worth of MASA, it straight up tells you you'll pay: → $60.82 interest → $20.08 in fees No surprise APR changes. ─────── Why this setup is so nice: 👉 I don’t have to touch anything. After 1 month, it auto-settles. No need to time exits or micro-manage positions. When to use this strat: > Got some spare stables and want to park them short-term (<1 month) with solid, fixed yield > Wanna borrow ETH/BTC to farm points or airdrops short-term and know exactly your cost upfront > Or even short tokens that are borrowable on Teller ─────── Bonus alpha: Teller raised like $8M from legit funds. They’ve got a points system + rankings, so there’s a chance you get extra rewards down the line, not just only APY.

Alex Gladstein 🌋 ⚡@gladstein· 26 days ago

RT @TryMapleAI: Maple has two powerful new models: Qwen 2.5 72B and Mistral Small 3.1 24B. 💻 Qwen brings multilingual support, and advanced math & programming skills 🌄 Mistral delivers vision capabilities for image analysis and fast chat performance Try them✌️ https://t.co/8FOocmZRst

Maple AI@TryMapleAI· 26 days ago

Maple has two powerful new models: Qwen 2.5 72B and Mistral Small 3.1 24B. 💻 Qwen brings multilingual support, and advanced math & programming skills 🌄 Mistral delivers vision capabilities for image analysis and fast chat performance Try them✌️ https://t.co/8FOocmZRst

Swing | Unified Cross-chain UX 🤖@swing_xyz· 26 days ago

The pursuit of perfect swaps goes beyond price. Today's builders and users demand something fundamental: privacy as a right, not luxury. Our partnership with @HoudiniSwap delivers exactly that. We've fused elite routing with untraceable transactions. Houdini Swap solves the privacy problem completely. Their non-custodial system breaks visible wallet links, letting users execute cross-chain swaps without exposing source or destination addresses. Here's how it works: Houdini queries our API for the most efficient path. That optimal route then vanishes into their dual-exchange architecture. Two technologies, one seamless experience. Our routing, their shield. For developers, this eliminates a critical trade-off. Your users no longer choose between great rates and privacy. Access Houdini through Swing and deliver both to your users. For users, the math is simple: Swing execution + Houdini privacy = trading without compromise. They get optimal rates while maintaining complete transaction confidentiality. The efficiency-privacy trade-off is dead. The infrastructure is here. Optimal execution with complete privacy. The no-compromise stack developers have been waiting for, powered by Swing and Houdini.

MathWallet@MathWallet· 26 days ago

MathWallet has listed the $PUMP token! 🚀Explore new possibilities in the crypto world with @pumpdotfun ! $MATH https://t.co/g2K5g0LLEb

Immutable@Immutable· 26 days ago

RT @InsikWeb3: Starting today, I’m staking every asset I’ve got on @Immutable. Kicking things off with 35 $IMX. And I’m not stopping there. I’ll keep stacking until I hit the full 1,000 $IMX stake. Why? Because the math is loud. That’s a 30.97% Annualized Rewards Rate. One of the juiciest plays in the game right now. Stakers win on @Immutable.

Kyle Chassé / DD🐸@kyle_chasse· 26 days ago

You think $269K BTC is wild? Let's do the math: USD down 10% DXY has been at its worst since 1973 Recession odds collapsing 0 sell signals from CoinGlass Fed pivot coming Now layer in Trump’s ETF. Still think $125K is the top? https://t.co/DZjYdfLtu4

AGI.Eth@ceobillionaire· 27 days ago

RT @PhysInHistory: Vector fields help us see invisible forces like gravity, magnetism, and wind currents. They use arrows to show direction and strength at every point in space, like a weather map showing wind flow. From planets orbiting to air moving over wings, vector fields help us understand how forces shape our world! 📷 math.animations

BowTiedBull.eth - Read Pinned or NGMI@BowTiedBull· 27 days ago

It's pretty hilarious that you need a promotion to offset inflation over a 3-5 year span. Big hoop-la about getting promoted to the next order on the chain only to do the math and realize it's to keep you flat with inflation

ARK Invest@ARKInvest· 27 days ago

RT @CathieDWood: The price of any good is determined at the margin, demand relative to supply. The math is mathin’ !

MathWallet@MathWallet· 27 days ago

🚀BTC has smashed through $120K! History in the making – where do you think we're headed next? $MATH #Bitcoin https://t.co/pS7nJzT19e

hexens@hexensio· 27 days ago

Early FHE was a math marvel but a compute nightmare. Modern schemes like BGV/BFV/CKKS flip the model, packing vectors with CRT and unlocking SIMD via Galois automorphisms. Here’s how the math makes FHE practical: https://t.co/MQlyxER8ez

Cryptor ⚡️@cryptorinweb3· 27 days ago

Aped a bag of $YNE @yesnoerror at $7M MC. It just had a revive, here's why 👇 Been following the project for a while now, and I’m excited about its potential. YesNoError is building a decentralized platform to audit scientific research papers (think @arxiv) for errors. They’re using AI (like OpenAI’s o1 model) combined with blockchain for transparency. It’s not just about AI papers but also catching math, data, or methodology flaws. This help researchers cut through the noise by verifying sources, saving time for legit findings. Some interesting social interactions happening around it right now: - @jackclarkSF (co-founder of @AnthropicAI) is giving feedback directly to the founder. Anthropic runs a $60B AI conglomerate. - @jessepollak (creator of Base) is supporting their Base integration. They’ve got an active Telegram chat with the Base team. According to the team, they’re also supported by other notable names like: - @reidhoffman (LinkedIn founder) - @BoostVC (early Coinbase investor) - @davemorin (founder of @OfflineVentures and chairman at @Esalen) $YNE holders get priority paper reviews, with more features teased. Nothing concrete yet but interesting if you’re deep in AI or research. Price just pumped, probably because of the news and @Soloxbt aping $27K. Decided to follow in, and I’ve got more DCA buys ready if there’s a retrace. Personally, I do a lot of research myself. It’s annoying to constantly triple-check sources or wonder if a paper is actually valid. A tool that helps with that would save a lot of time. Worth keeping @yesnoerror on the radar.

Insik@InsikWeb3· 27 days ago

Starting today, I’m staking every asset I’ve got on @Immutable. Kicking things off with 35 $IMX. And I’m not stopping there. I’ll keep stacking until I hit the full 1,000 $IMX stake. Why? Because the math is loud. That’s a 30.97% Annualized Rewards Rate. One of the juiciest plays in the game right now. Stakers win on @Immutable.

Physics In History@PhysInHistory· 27 days ago

Vector fields help us see invisible forces like gravity, magnetism, and wind currents. They use arrows to show direction and strength at every point in space, like a weather map showing wind flow. From planets orbiting to air moving over wings, vector fields help us understand how forces shape our world! 📷 math.animations

Mike Alfred@mikealfred· 28 days ago

Look, at some point, if you are good enough at investing, becoming a billionaire becomes somewhat inevitable. It’s a simple compound growth math problem. Nothing more. And the significance of the title itself loses meaning. It’s just something that happens organically. NBD.

Mario Nawfal@MarioNawfal· 28 days ago

GROK 4 CRUSHES CHATGPT - WIPING THE FLOOR WITH RIVALS Grok 4 shreds ChatGPT and Gemini, claiming the AI throne without mercy! • ARC-AGI jumps 15.9% — brains with bite, not bloated fluff • 200,000 GPUs = pure muscle, wrecking math & logic pretenders • Elon’s relentless grind fuels Grok 4’s feral rise past the weaklings The AI arena just got steamrolled by Grok 4! Source: @NotebookcheckNet

Cathie Wood@CathieDWood· 28 days ago

The price of any good is determined at the margin, demand relative to supply. The math is mathin’ !

Crypto King@CryptoKing4Ever· 28 days ago

It's better to get yours while you can. You are super early if you are reading this post. Simple math! https://t.co/2yPJS6ViqM

Peter Brandt@PeterLBrandt· 28 days ago

Some absolute gems from from Alex @thiccyth0t at thiccythot@substack.com, one of the deans of trading math -- a MUST FOLLOW " 'Just win more' isn't scalable advice" For those seeking moonshots, you must come to terms with the "Jackpot Paradox," what economists call the "ergodicity problem" "That taste for outsized risk has seeped into everyday culture" "We're becoming a culture that worships the jackpot and increasingly prices survival at zero" and this leads to "degeneracy and desperation that has arisen from the culture" Desperate for economic and social parity, the Zs and Ms "risk too much and drawdown too deep." "Build more edge rather than risk more size. Don't kill yourself chasing the jackpot." And here is my favorite: "'Just win more' isn't scalable...." Long term success in market speculation is not so much about looking for a set up as it is about understanding the math behind trading

MathWallet@MathWallet· 28 days ago

Introducing the Bonk Ecosystem Map! @theuselesscoin @Hosico_on_sol @pomkori @ikuncoinx @Crybaby_on_sol @BluechipDotSol @bonk_fun @AgentNyla @bapOnBonk @gibthefrog @bestcoinbonk @kakaforsol @bonk_fun @RaydiumProtocol @JupiterExchange @orca_so @phantom $MATH #BONK https://t.co/o49kX2qT9R

Yelay@YieldLayer· 28 days ago

DeFi keeps moving. Whether you’re in or not. It’s not hype. It’s just how crypto works now. And unless the whole space disappears (which feels unlikely), DeFi’s not slowing down. From Maker and Aave to Pendle loops, undercollateralized loans, intent-based lending, and social trading dashboards — the tools keep getting better. Smarter. More efficient. Now we’re entering a new phase: AI-native DeFi. No more babysitting positions or doing LTV math by hand. Agents will handle loops, manage risks, and personalize strategies in real time. It’s DeFi with a brain. Yelay is already there. Our Trusted Yield Environment is made for this shift — ready for agents, automation, and smarter yield flows. Don’t wait around for “the future.” You can start building it here: https://t.co/hGlQhHEWP8

Stephen | DeFi Dojo@phtevenstrong· 28 days ago

Anywho, please feel free to audit my math! I love when people add to the points speculation conversation. One thing to note is that just because S1 is six months doesn't mean TGE will be exactly after S1. I hope it is, but I know how DeFi works, delays should be expected.

Stephen | DeFi Dojo@phtevenstrong· 28 days ago

In just a few weeks, @infinifilabs broke through $50M in TVL. This reinforces my belief the YTs are being undervalued on @pendle_fi. I did some math 🧵👇 https://t.co/9uQihCM8sM

Choice Coin@ChoiceCoinDAN· 29 days ago

RT @BitcoinMagazine: 21 million #Bitcoin, 8.1 billion people. Do the math ✨ https://t.co/TbdUJCer96

Bitcoin Magazine@BitcoinMagazine· 29 days ago

21 million #Bitcoin, 8.1 billion people. Do the math ✨ https://t.co/TbdUJCer96

Kaspa@kaspaunchained· 29 days ago

Math, code, and people in motion. https://t.co/0M9gTJxDPA

Massa@massachain· 29 days ago

RT @DusaLabs: Yesterday was a significant milestone for Dusa The token release testifies 3 years of hard work by the whole team since June 2022, when we decided to commit full-time to the development of DeFi on Massa. This release is also the fruit of the investment of 40 angels who chose to commit to a project that puts decentralization back at the heart of thinking, with real vestings respecting the community. But despite the energy and good intentions demonstrated by the tokenomics, backer commitments and fair launch, we've taken a beating. But that doesn't change our energy and motivation. Especially with the next releases (revenue sharing / ve3,3 / limit orders) due before the end of the year to align the motivations of all Dusa stakeholders and push $DUSA higher. So don't forget to stake your tokens because the staking APR is still ultra juicy. From Sept 11th, the token will be aligned with traders' fees and redistributed to active stakers. FYI Dusa Protocol generated +$70k of fees since its deployment and the current market cap is less than that number. NFA but math isn't mathing here...

Dusa Labs@DusaLabs· 29 days ago

Yesterday was a significant milestone for Dusa The token release testifies 3 years of hard work by the whole team since June 2022, when we decided to commit full-time to the development of DeFi on Massa. This release is also the fruit of the investment of 40 angels who chose to commit to a project that puts decentralization back at the heart of thinking, with real vestings respecting the community. But despite the energy and good intentions demonstrated by the tokenomics, backer commitments and fair launch, we've taken a beating. But that doesn't change our energy and motivation. Especially with the next releases (revenue sharing / ve3,3 / limit orders) due before the end of the year to align the motivations of all Dusa stakeholders and push $DUSA higher. So don't forget to stake your tokens because the staking APR is still ultra juicy. From Sept 11th, the token will be aligned with traders' fees and redistributed to active stakers. FYI Dusa Protocol generated +$70k of fees since its deployment and the current market cap is less than that number. NFA but math isn't mathing here...