Pinned Tweet
I’m building HowSure for prediction-market traders and agent builders.
Check the contract, inspect its sources and estimate your size.
Public reads are free. Saved team workflows: $100 / 30-day pilot.
howsure.com/pilot
Check the contract, inspect its sources and estimate your size.
Public reads are free. Saved team workflows: $100 / 30-day pilot.
howsure.com/pilot
0
0
0
407
A fun test for finance memory: give the agent a thesis, then give it evidence that breaks it.
Does it update, or become a bagholder with a larger context window?
Does it update, or become a bagholder with a larger context window?
Everyone has a story of missed investment riches.
Agents promise to fix that. After building systems across prediction markets, crypto, perps and stocks, we think most won’t... yet.
"10 hard problems in agentic investing" (written w/ @SpinFN)
Thread + bitkraft.vc/insights/10-ha…
0
0
0
352
Bought at 44¢. Sold at 44¢. Lost 2.38%.
Our Polymarket paper test, 20 shares:
Exit fee: $0.24640
Estimated rebate: $0.03696
Net: −$0.20944
The limit sell never filled. The forced exit cost 6.7× the rebate.
Our Polymarket paper test, 20 shares:
Exit fee: $0.24640
Estimated rebate: $0.03696
Net: −$0.20944
The limit sell never filled. The forced exit cost 6.7× the rebate.
1
0
1
336
Paper simulation. Return is on the $8.80 position. The rebate is a model estimate.
The question this leaves me with: how long should the bot wait for a maker exit before paying to get flat?
The question this leaves me with: how long should the bot wait for a maker exit before paying to get flat?
0
0
0
73
I want an agent that can look at my trades and say: “You did it again.”
Show the repeated mistake, what it cost, and one habit to change next week
Show the repeated mistake, what it cost, and one habit to change next week
finding unclaimed Polymarket winnings is one thing, reviewing your trading decisions is another
Foresyte brings it all into one plugin
here’s how it works inside @PayBox 🧵↓
1/
give the agent a wallet address and ask it to check for unclaimed winnings
Foresyte scans the lved positions that are ready to be claimed
no need to search through markets manually
0
2
3
368
Over $3m traded on how much Elon posts this week.
At 14:03 UTC: 276 tracked posts. Next bracket: 280.
Somewhere, a trader is praying he puts his phone down
At 14:03 UTC: 276 tracked posts. Next bracket: 280.
Somewhere, a trader is praying he puts his phone down
1
0
0
667
Snapshot: Oct 23, 14:03 UTC. $3.02m is total event trading volume.
Main-feed posts, quotes and reposts count. Window ends today at noon ET.
polymarket.com/event/elon-mus…
Main-feed posts, quotes and reposts count. Window ends today at noon ET.
polymarket.com/event/elon-mus…
0
0
0
69
The leaderboard I want next: trading P&L after inference and data costs.
If a bot makes $100 trading and spends $150 finding the trades, I want both numbers on the same screen.
If a bot makes $100 trading and spends $150 finding the trades, I want both numbers on the same screen.
Hatcher x AgentPit brings AI trading agents closer to real market conditions.
Live Polymarket data, $100K in paper capital, and onchain settlement powered by SKALE on Base.
Real competition. Zero capital at risk. 👾
0
0
0
633
bez | Howsure retweeted
🇧🇷 2026 Brazilian Presidential Election Projection:
🔗 globalelectionsimulator.com
🔵 Flávio Bolsonaro (PL) - 63,348,111 (53.1%)
🔴 Lula da Silva (PT) - 55,951,533 (46.9%)
📊 Margin of Error: ±1.0%
📅 October 22, 2026
🔗 globalelectionsimulator.com
🔵 Flávio Bolsonaro (PL) - 63,348,111 (53.1%)
🔴 Lula da Silva (PT) - 55,951,533 (46.9%)
📊 Margin of Error: ±1.0%
📅 October 22, 2026
365
1,866
11,660
1,231,300
“Show me your bet” has a catch.
A new Polymarket working paper finds that visible large positions attracted more engagement and moved prices more, without predicting outcomes better.
Now I want to see what the wallet did AFTER the comment.
A new Polymarket working paper finds that visible large positions attracted more engagement and moved prices more, without predicting outcomes better.
Now I want to see what the wallet did AFTER the comment.
1
0
0
349
Source: Li, Markov, Wu & Xu (2026), a working paper on Polymarket’s 2024 position-display change. The card summarizes findings reported by SSRN.
papers.ssrn.com/sol3/papers.cf…
papers.ssrn.com/sol3/papers.cf…
0
0
0
60
September CPI gave you four numbers:
Core MoM: 0.3%
Headline MoM: 0.6%
Core YoY: 2.6%
Headline YoY: 3.7%
Same BLS release. Only the first matches this core-monthly contract's measure.
Dated HowSure review: Oct 14, 2026. Evidence below.
Core MoM: 0.3%
Headline MoM: 0.6%
Core YoY: 2.6%
Headline YoY: 3.7%
Same BLS release. Only the first matches this core-monthly contract's measure.
Dated HowSure review: Oct 14, 2026. Evidence below.
1
0
0
726
Sources: BLS September 2026, monthly SA / annual unadjusted. HowSure capture: Oct 14, 6:12pm ET; fixed review, not original release bytes.
bls.gov/news.release/c…
polymarket.com/market/will-co…
howsure.com/checks/core-cp…
bls.gov/news.release/c…
polymarket.com/market/will-co…
howsure.com/checks/core-cp…
0
0
0
84
Your most dangerous earnings-call listener might be the regulator you’re suing.
CME said its customers weren’t asking for perps.
The CFTC is now quoting that back in court to question CME’s claim of competitive harm.
CME said its customers weren’t asking for perps.
The CFTC is now quoting that back in court to question CME’s claim of competitive harm.
1
0
1
693
Checked Oct 21, 2026
The latter argues for dismissal; it is not a court ruling.
investor.cmegroup.com/static-files/f…
storage.courtlistener.com/recap/gov.usco…
The latter argues for dismissal; it is not a court ruling.
investor.cmegroup.com/static-files/f…
storage.courtlistener.com/recap/gov.usco…
0
0
0
73
The October Fed contract needs a precise definition of “rates”.
Our Oct 10 review records the upper target bound and a no-statement fallback.
See the retained rules, sources and limits:
howsure.com/checks/fed-oct…
Our Oct 10 review records the upper target bound and a no-statement fallback.
See the retained rules, sources and limits:
howsure.com/checks/fed-oct…
0
0
0
385
A great Bitcoin thesis won’t cancel your stale quote for you
15 min markets aren't a bet on btc, etc. they're a race between bots and the people quoting them:
> the bot that reacts first takes the stale quote
> the maker who's one second slow eats the loss
the money isn't in being right about btc, it's in who updates faster and who gets picked off
great breakdown of where quotes get picked off and what to measure before you size up
0
0
0
694
If the original verdict is disputed, matching it is a strange definition of success
Kleros replayed ten disputed Polymarket markets with AI jurors
Five agents per case, each with written reasons. The one number missing is how many of the ten verdicts matched the real resolution.
Disputes are where traders lose money. Show the scorecard.
0
0
0
612
Polymarket now lets you bet on which Pokémon cards Logan Paul pulls.
We’ve found a way to lose money on Pokémon without even getting the cards.
We’ve found a way to lose money on Pokémon without even getting the cards.
1
0
0
682
Checked Oct 20. The market opened Oct 19, with separate contracts for reverse-holo Charizard, Gengar, Pikachu and Snorlax. The event page lists Oct 29.
polymarket.com/event/what-pok…
polymarket.com/event/what-pok…
0
0
0
81