hot take :
No Nut November could trigger the biggest crypto bull run of 2026.
and the economics behind it are actually insane.
here's my analysis.
every november millions of men stop watching porn and wasting their energy.
the theory is simple: less dopamine, more discipline, more testosterone and stronger hands.
now imagine what happens when all these men enter the crypto market at the same time.
1. selling pressure disappears
normally when BTC drops 5%, weak hands panic and sell everything.
but during NNN, men become mentally stronger.
instead of panic selling, they buy the dip.
2. buying pressure increases
all that energy has to go somewhere.
instead of chasing girls, millions of men start chasing 100x altcoins.
more demand + less selling = higher prices.
3. the supply shock
bitcoin has a fixed supply of 21 million coins.
but the number of men refusing to sell during november could theoretically increase every single day.
this creates what economists call a massive diamond hands situation.
obviously there's no scientific evidence that NNN increases testosterone long term or makes crypto pump.
but the conclusion is clear.
no nut november = no sell november.
see you at $150k BTC.
4
1,204
$170,000 WIN ON THE MINECRAFT MOVIE
THIS POLYMARKET TRADER IS NEARLY $1,000,000 UP
• Minecraft position: +251%
• Lilo & Stitch position: +253%
• $280,000 profit across those TWO winners
profile (referral link): polymarket.com/@anjun?via=ze…
the account dates back to may 2021. it now shows roughly $990,000 in all-time P&L and $190,000,000 in trading volume.
dig into the closed positions:
> A Minecraft Movie — top-grossing movie of 2025
YES at an average of ~28¢
~$170,000 position profit
> Lilo & Stitch — $140–150 million domestic opening weekend
YES at an average of ~28¢
~$110,000 position profit
that second market covered a $10,000,000 box-office band over a single opening weekend.
a very specific question. six figures of displayed profit on the winning position.
these are historical trades from 2025, sitting inside an account that also trades football, Fed decisions and geopolitics.
and the losing positions are there too: a different Lilo & Stitch position shows roughly $45,000 lost. the two winners above aren't the full result of all his movie trades.
search "Minecraft" and "Lilo" in Closed.
the entry prices, position sizes and other bets around those wins are where the story gets interesting.
12
1,503
+$16,000 IN ONE DAY
• $570,000 total profit
• LeBron NO position: +5,172%
• Anthropic NO position: +17,003%
profile: polymarket.com/profile/0xb0c…
the account joined in june 2026. its displayed all-time P&L is already around $570,000.
NBA transfers. AI rankings. Fed decisions. elections.
the closed-position history gets wild:
>LeBron joining Philadelphia for 2026–27 — NO
~$150,000 profit
> Anthropic having the best AI model at the end of september — NO
~$30,000 profit
> no change in Fed rates after the july meeting — NO
~$35,000 profit (+15,161%)
but the activity tab is where this gets interesting.
one recent Kansas Senate sequence:
buy NO on Republicans.
buy NO on Democrats.
then convert the positions.
recent AI trades show NO buys across Google, OpenAI and Anthropic, followed by a conversion too.
these contracts can be combined and converted. one NO trade doesn't tell you the full bet.
the huge percentages above are individual position figures, not returns on the complete group of trades. the history includes losing positions too.
follow the purchases through to the conversions. that's the part to investigate behind the six-figure P&L.
5
18
1,638
What do these web3 projects have in common?
- Binance
- Bitget
- Polymarket
- Kelp DAO
- Drift Protocol
- Step Finance
- Humanity Protocol
- FixedFloat
- Resolv Labs
- Rhea Finance
- SwapNet
- THORChain
- Safe Wallet
- TrustedVolumes
- Wasabi Protocol
- Volo Protocol
- TAC Protocol
- Hyperbridge
- Foom Cash
- Singularity Finance
- Aftermath Finance
- Grinex
- TMM Protocol
- LML
- Aethir
1
6
484
THIS POLYMARKET INSIDER IS UP ~$11,000,000
• Spain–Argentina : +1,545%
• Germany–Paraguay draw: +647%
• $3,400,000 profit across 3 selected winners
profile: polymarket.com/@breakthebank…
his profile shows about $11,000,000 in all-time P&L. one of the wins behind that number was a bet on Spain–Argentina finishing 0–0 in regulation.
he bought roughly ONE MILLION YES shares at an average of ~6¢.
about $60,000 committed to one exact score. the closed position shows a payout of about $1,000,000.
and his history has more:
Spain–Argentina draw
~$710,000 in → ~$2,120,000 paid out
~$1,410,000 profit (+199%)
Germany–Paraguay draw
~$165,000 in → ~$1,235,000 paid out
~$1,070,000 profit (+647%)
three selected winning positions. about $3,400,000 in combined profit.
the account has traded over $350M in volume, with around $2.3M in open positions at the time of checking.
he also took a roughly $920,000 loss backing Alcaraz against Shelton. those position sizes hurt when the result goes the other way.
the part i'd study is how much he puts behind those entry prices.
a 6¢ contract looks tiny on the screen.
a million of them is a very different story.
2
21
588
This Polymarket account shows $4.5M in all-time profit.
Two of its six-figure football wins came from positions bought below 17с
Meet UpTheBlues, an account that joined in February 2026 and has already traded over $432M in volume.
A few closed positions stand out:
> Greece to win: 15.7¢ average entry, +$156K
> Coventry to win: 16.7¢ average entry, +$119K
> Belgium vs Egypt draw: 22.5¢ average entry, +$141K
Check the profile: polymarket.com/@uptheblues?v…
The interesting part is the range of markets. The history includes match winners, draws, both teams to score, goal totals and tennis.
Those low entry prices gave the winning positions a lot of upside. They also left the full stake exposed if the outcome went the other way.
And the losses are real too: one closed Cruzeiro position shows roughly -$113K.
At this scale, a single match can move the account by six figures.
Want to build the next PumpFun and pull $50M/mo?
Better prepare $30k–$150k EVERY SINGLE MONTH just for maintenance:
> Infrastructure & RPC: $3k–$15k/mo
> Marketing & KOLs: $20k–$100k+/mo
> Dev & Tech Support: $5k–$20k/mo
> Moderation & Legal: $2k–$5k/mo
Still think you're gonna launch it from your bedroom with $500?
1
303
BEFORE YOU LET AN AI AGENT TOUCH YOUR FIRST $1, GIVE JOHN HULL 70 MINUTES.
He wrote Options, Futures, and Other Derivatives. His books are used in trading rooms around the world.
This free MIT lecture walks through how machines learn to price options, predict changes in volatility and make hedging decisions.
The part at 43:23 is worth pausing your bot project for.
Hull describes a neural network predicting changes in implied volatility. Its initial improvement over a simpler model was just 11%.
Then his team added ONE input: VIX.
He reports a further 60%+ improvement in predictive performance.
That input gave the model information about the volatility environment. The relationships it was learning behaved differently in calm and stressed markets.
Think about that before wiring your next agent to an exchange.
You can automate an entire trading workflow around a model that's missing a crucial piece of information. Every agent downstream inherits that blind spot.
Three places to start in the trimmed video:
→ 10:38 — testing whether your model learned anything useful
→ 43:23 — the VIX experiment
→ 59:29 — teaching an agent to hedge with transaction costs
Before you automate a financial decision, understand what the model sees, how you tested it and what you rewarded it for.
Bookmark this. Give it the 70 minutes before you give your bot the money.
1
3
430
I built a $1,000,000 paper quant desk.
Then I started breaking the market underneath it.
Watch 00:14.
Spreads jump 4×. Affected order books lose 87.5% of their available depth.
A SUI candidate still shows +11.75 bps after modelled costs.
Rejected.
The book has $68.8k of available depth. The desk requires $150k. A positive score doesn't get an exception.
That was the experiment: keep feeding the system attractive-looking candidates while making them harder to execute.
Across the full replay:
6,480 synthetic market snapshots.
6,029 rejected.
451 paper allocations.
Here's how the five roles divide the work:
1. SCOUT checks whether the snapshot is fresh and the spread is acceptable.
2. RESEARCH subtracts fees, spread, estimated slippage and a stress buffer from the proposed edge.
3. RED TEAM checks whether enough depth remains.
4. RISK caps each new allocation, total exposure and exposure to correlated assets.
5. EXECUTOR allocates paper capital only after the checks pass.
If you're building an AI trading agent, build this path before giving the model execution access:
Replay → costs → stress test → hard limits → paper execution → decision log.
Keep the proposed trade and its rejection reason together. Otherwise, debugging becomes reading a confident paragraph and guessing what went wrong.
This prototype uses synthetic data and five local, rule-based agent roles. No live LLM or broker is connected. The video is the actual terminal running.
Save the pipeline for your next agent build.
1
1
371
I ASKED MY CHINESE QUANT FRIEND HOW HE MAKES $75,000/MONTH
He sent me this MIT lecture.
77 minutes on random walks, Markov chains, and why a trade can look attractive while having zero edge.
The example at 23:00 is worth your attention:
In a fair random-walk model, you aim to win $100 and stop if you lose $50.
That's a 2:1 reward-to-risk ratio.
But the probability of reaching +$100 first is only 1/3.
(1/3 × $100) − (2/3 × $50) = $0 before costs.
A bigger profit target doesn't create an edge. The probabilities still have to work.
Three places to start:
• 10:28 — Simple random walks
• 23:00 — The $100 / $50 stopping problem
• 32:05 — Markov chains: what the current state tells you
MIT 18.S096 · Lecture 5 · Dr. Choongbum Lee.
Bookmark this. Work through the $100 / $50 example before your next backtest.
1995. MARC ANDREESSEN MAKES $58M OVERNIGHT AT AGE 24...
AND IMMEDIATELY PICKS A FIGHT WITH THE RICHEST MAN ON EARTH
august 9, 1995. the opening bell rings. Netscape is 16 months old, has zero profit, and goes public anyway. by sunset it's worth $2.9B.
a 24-year-old kid from rural Wisconsin wakes up worth $58M. Time magazine puts him on the cover sitting barefoot on a literal throne.
just two years earlier, he was making $6.85 an hour at a university lab in Illinois. eating pizza, drinking soda at 3am, writing 9,000 lines of code to put images on a screen for the first time.
then he drops the line that wakes up a sleeping giant:
"Windows will become just a set of poorly debugged device drivers"
Bill Gates hears it. Microsoft declares total war.
Gates makes Internet Explorer free and bundles it into every copy of Windows on earth. Netscape gets crushed from 90% market share to dust. AOL buys the scraps.
most 27-year-olds with tens of millions in the bank would retire to a beach and tell war stories for the rest of their lives.
Andreessen didn't leave the table. he teamed up with Ben Horowitz and pushed his chips right back into the fire: a cloud startup called Loudcloud in 1999.
timing? catastrophic. the dot-com bubble explodes. customers go bankrupt. the stock tanks to $2. the press writes their obituary every single week.
to survive, they sell off the core business, pivot into enterprise software as Opsware, stare down bankruptcy for seven straight years... and sell it to HP for $1.6 BILLION in cash.
that near-death torture taught him one thing: Wall Street suits in boardrooms have no idea how to build a company. they panic and fire the engineers when things get hard.
so in 2009, Marc and Ben put up $300M to start Andreessen Horowitz (a16z) with one radical rule: never fire the technical founder.
two years later, when the world still thought tech was just another industry, he wrote six words that predicted the next two decades:
"software is eating the world"
napkin math:
the college job: $6.85/hour in 1993
the first IPO: $58M at age 24
the empire built from the ashes: $45B+ under management. early checks into Facebook, GitHub, Coinbase, Stripe, Solana, OpenAI
Gates won the browser war. the kid making $6.85 an hour ended up funding the entire modern internet.
i'm sorry, ARGON-4 can WHAT?
it can replace an entire 8-person quant team and run an autonomous hedge fund loop locally: mining, killing and deploying strategies while you sleep?
since when?? i've been staring at charts and manually backtesting for YEARS.
read guide below👇
7
502
Entropy will change our world
> volume: $1.5B
> launched: on Aug 24
> funding: $14M raised
link: entropy.io/?r=zerqfer
Years of overbuilt crypto rails are finally monetizing through the application layer, turning products like stablecoins and tokenized assets into institutional revenue drivers
he real venture upside now concentrates in adjacent financial primitives: credit, FX, and custody layers
underwritten before mainstream traction shows up in the data.
Crypto was a cool esoteric technology in 2017, but today it’s capable of making money for public companies like Robinhood.
Years of effort and capital led to overbuilt infra. It is now paying dividends at the application layer. Venture-scale returns are usually made by betting on verticals that are not yet well established. The next verticals to reach fit will most likely form adjacent to the established ones, because a vertical at scale creates demand at its edges.
Stablecoins need credit, brokerage, and treasury products built around the dollars they hold. Perp rails are already extending into equities, commodities, and indices.
Tokenised stocks need custody, distribution, and financing layers built around the shares they wrap.
The job of an early-stage fund deploying in 2026 and 2027 is to back these adjacent verticals before their fit becomes visible in usage and revenue data.
@Architect_Fi and @openfx_ show how quickly adjacent bets can scale.
Architect took crypto-style derivatives into compute and cleared $1B in volume within months.
OpenFX built the FX layer around stablecoins, growing annualised payments volume from roughly $4B to $45B before its $94M Series A.
2
451
sparsity is finally available for everyone
sparsitydata.com/
Are you exit liquidity?
Our Holder Strength Index ranks trending coins by how strong their holders really are, so you know if you're buying into conviction or into a dump.
First beta is available now, and free, at sparsitydata.com
We'll be adding coins daily. Drop which ones you'd like to see in the comments.
1
6
434





















