checkpoint (@checkpoint_gg_) header
checkpoint (@checkpoint_gg_) profile picture

Trailers, release dates, and the stuff studios actually confirm - not copium, not rumors. I play the news so you don't have to doomscroll it.

New York, USA
Joined August 2026
Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)Photo by checkpoint (@checkpoint_gg_)
checkpoint retweeted
GTA 6 was leaked from a Travelodge room with an Amazon Fire Stick plugged into the TV. It cost Rockstar $5 million. Yesterday, 3 days before launch, the hacker pleaded guilty 🔥 His name is Arion Kurtaj, now 21. Per the BBC, who were in Woolwich Crown Court, he was 17 in 2022, out on bail and under police protection in a hotel because rival hackers were after his Bitcoin. His computer had been taken. So he used the hotel TV, a Fire Stick, a Bluetooth keyboard and mouse. What he did with it: → took 90 clips, nearly 60 minutes of unfinished GTA 6 → got into Rockstar's Slack and wrote: "If Rockstar does not contact me on Telegram within 24 hours I will start releasing the source code" → Rockstar didn't. He did → Rockstar says the cleanup cost $5M plus thousands of hours of staff time In the video below, the second chapter. In August 2026 a leaker called CyberLeak posted GTA 6 clips almost daily and cashed in through digital currency, until Take-Two went to court in New York. Rockstar's answer: GTA 6 launches with no disc, just a download code. A journalist told the BBC that also means no resale, so Rockstar keeps all the profit. To be fair, he was 16 and 17, he's autistic, and he was first sent to a secure hospital before being found fit for trial. Sentencing comes later. The scoreboard: one teenager, one TV stick, $5 million to clean up. The game he leaked: an estimated $1.45 billion in pre-orders. So who paid for the leak in the end: the hacker, Rockstar, or the players?
The GTA 6 leaker behind the massive 2022 leak has officially pleaded guilty, just 3 days before GTA 6 releases. - Hacked Rockstar Games using an Amazon Fire TV Stick from a hotel room. - Leaked 90 clips containing nearly 60 minutes of gameplay. - Cost Rockstar Games $5 million and thousands of hours of staff time. He will be sentenced at a later date. (Source: bbc.com/news/articles/cq99ex…)
5
1
1
9
2,735
GTA 6 was leaked from a Travelodge room with an Amazon Fire Stick plugged into the TV. It cost Rockstar $5 million. Yesterday, 3 days before launch, the hacker pleaded guilty 🔥 His name is Arion Kurtaj, now 21. Per the BBC, who were in Woolwich Crown Court, he was 17 in 2022, out on bail and under police protection in a hotel because rival hackers were after his Bitcoin. His computer had been taken. So he used the hotel TV, a Fire Stick, a Bluetooth keyboard and mouse. What he did with it: → took 90 clips, nearly 60 minutes of unfinished GTA 6 → got into Rockstar's Slack and wrote: "If Rockstar does not contact me on Telegram within 24 hours I will start releasing the source code" → Rockstar didn't. He did → Rockstar says the cleanup cost $5M plus thousands of hours of staff time In the video below, the second chapter. In August 2026 a leaker called CyberLeak posted GTA 6 clips almost daily and cashed in through digital currency, until Take-Two went to court in New York. Rockstar's answer: GTA 6 launches with no disc, just a download code. A journalist told the BBC that also means no resale, so Rockstar keeps all the profit. To be fair, he was 16 and 17, he's autistic, and he was first sent to a secure hospital before being found fit for trial. Sentencing comes later. The scoreboard: one teenager, one TV stick, $5 million to clean up. The game he leaked: an estimated $1.45 billion in pre-orders. So who paid for the leak in the end: the hacker, Rockstar, or the players?
The GTA 6 leaker behind the massive 2022 leak has officially pleaded guilty, just 3 days before GTA 6 releases. - Hacked Rockstar Games using an Amazon Fire TV Stick from a hotel room. - Leaked 90 clips containing nearly 60 minutes of gameplay. - Cost Rockstar Games $5 million and thousands of hours of staff time. He will be sentenced at a later date. (Source: bbc.com/news/articles/cq99ex…)
5
1
1
9
2,735
Nobody has played GTA 6 yet. It has already made an estimated $1,450,000,000. And on Monday, Take-Two stock fell 5.9% 📉 That's Sensor Tower's estimate as of Nov 16: 14.94M pre-orders, up 2.81M since Friday, 3 days before launch. In the video below, the money behind it: → the budget: Rockstar told Bloomberg it's "not quite as much as the Burj Khalifa," which cost an estimated $1.5 billion → the day-one math on CNBC: 25M copies at about $100 each is $2 billion. The whole budget back in 24 hours → an analyst: GTA V took six weeks to sell 29 million units. GTA 6 might do that on day one → Take-Two's CEO Strauss Zelnick, more careful: "I tend not to claim success before it occurs" → and the bill for everyone else: PTO requests for Nov 19 are up nearly 13%, with about 1 in 5 workers planning time off Now the part that matters if you own the stock. In September 2013, GTA V made $1 billion in 3 days, a record. Take-Two that week went from $17.66 to $16.99, down 3.8%. This week: $227.82 Friday, $214.46 Monday. The 2026 high was $242.40 back in August. The market had months to price in a hit. Launch week is often when people take the profit. To be fair, these are estimates, not Take-Two's numbers, and one week of trading says nothing about a game that could sell for a decade. GTA Online alone made about $500M a year for 13 years. The game will be huge. The question is whether huge was already in the price. Would you buy Take-Two before Thursday, or after? Not investment advice.
GTA 6 has now reached 15 MILLION pre-orders, generating $1.45 BILLION in revenue. (Via Sensor Tower)
6
7
1,091
GTA 6 comes out Thursday. This week the people who built it are voting on a union, and one of the things they want a say over is AI. The vote opened November 12 and closes November 26, across Rockstar's UK studios. It covers more than 1,600 people below director level, per Jason Schreier at Bloomberg. The union says a deal would let them negotiate pay, overtime, bonuses, leave and pensions, and get "consultation rights over HR policies and the use of AI." In the video below, how it got here. In October 2025 Rockstar fired 34 people, 31 in the UK and 3 in Canada, for "gross misconduct." They were called in and handed an envelope. The union says most, maybe all, of them were organizing, and calls it union busting. Why the AI line lands this month. On November 12 Xbox's CEO wrote that "a small team will be able to build worlds that today demand hundreds of developers." Meanwhile Rockstar, explaining the wait to the Guardian: "[We] could have made GTA 5.5 and done it faster. You have to take big swings." To be fair, Rockstar says the fired staff leaked confidential information in a roughly 300-person Discord, and a tribunal verdict is still pending. Its publishing head told Schreier: "If they want a union, they will have a union." And the fired workers asked fans not to boycott the game. The first new GTA in 13 years was made by thousands of people. Now they want a say in what happens to the next one. If your job held that vote, how would you vote?
Rockstar Games explain why they took so long to develop GTA 6. “If you were a developer with these resources, the only developer with these resources, wouldn't you go for the most ambitious thing you could? [We] could have made GTA 5.5 and done it faster. Ultimately, that's not what [we] wanted… You have to take big swings. It can't be just more of the same.” (Source: @Guardian)
6
6
420
The same company decides if you get the job and if you get paid. On Friday it couldn't do the second one. Oct 14, Workday CEO Aneel Bhusri on Fortune's podcast: payroll is "a pretty rote process." Can an agent replace a human at a lower cost? "Yes." Will companies want that? "Yes." Nov 13, around 5 PM ET, Workday's HR system went into maintenance mode. Workers at Walmart, Target, Chipotle and others couldn't clock in, check their pay, request time off or file for benefits in the middle of open enrollment. Workday said a problem found during testing left some data inconsistent. 11,500 organizations use it. Service came back the next day. Now the other door. Mobley v. Workday: more than 15,000 applicants over 40 joined an age claim against Workday's AI screening. Court filings put rejections through its tools at about 1.1 billion. One plaintiff says he applied at 12:55 and was rejected at 12:56. On Nov 10, Workday asked the judge to break the group up. Fair side: Workday denies any bias and says its AI "focuses on qualifications and is tested for potential bias." No court has ruled on the merits. The hearing is in March 2027. And the outage was a testing glitch, not an AI agent. But that's one vendor sitting between you and a job, and between you and your paycheck. When you got rejected in under a minute, do you think a person ever saw it?
🚨 JUST IN: HUGE OUTAGES are now impacting WALMART, TARGET and other corporate systems used for pay info, time off, and benefits Workday says a problem surfaced while TESTING ITS BACKUP SYSTEMS, leaving some data scrambled. This is RIGHT after Microsoft's massive outage, and the FAA outages happening today Just an insane chain of events where things seem to just be breaking ‼️ A White House briefing is about to commence!
4
4
488
Over a third of game developers lost a job in the last two years. This week, Xbox's CEO wrote that soon "a small team will be able to build worlds that today demand hundreds of developers." Asha Sharma posted the memo on Nov 12 for Xbox's 25th birthday. Before Xbox, she ran AI products at Microsoft. The memo never says "AI" outright, but read the line before it: "A single person will soon be able to wield capabilities that once required entire nations." In the video below, her pitch at Microsoft Ignite last year, then FOX 35 from July: → Xbox: 1,600 jobs cut that day, another 1,600 over the fiscal year → id Software, the Doom studio: 96 of 136 people laid off, per Eurogamer → "a lot of folks… watching the clock… saying, am I next?" The same week, GDC put out its State of the Game Industry report, 2,600+ game workers: → over 1 in 3 lost a job in two years → 40% now use generative AI at work → 55% say AI is hurting games One respondent from California: "I got into this industry to collaborate with creative people, not to have an advanced chatbot riff on the ideas of someone who was laid off." The fair side. The July cuts were blamed on money, not AI: big bets on Game Pass and acquisitions that didn't pay off. And small teams with good tools have made some of the best games ever. But "hundreds of developers" is how she described the old way of making games. Those hundreds are the people in the GDC numbers. If you make games, is AI your tool or your replacement?
33% of game developers lost their jobs in the past two years, while AI continues to grow across the industry. According to the GDC State of the Game Industry 2026 report: - 40% now use generative AI at work. - 55% believe AI is having a negative impact on games. - 54% of laid-off developers still haven’t found another job. - 65% say their mental health has suffered because of work. - 81% are developing their next game for PC. - VR development has dropped from 8% to 3%.
4
6
590
Microsoft's AI couldn't fix Microsoft. Outlook went down in 74 countries, and a day later humans were still rebooting servers. It started around 16:00 GMT on Thursday. Outlook, Teams and Copilot. Downdetector logged 43,000+ Outlook reports from 72 countries by Friday morning. Police, NHS services, schools, government systems in Japan and Singapore. Microsoft's own update: "marginal success." Servers it brought back kept failing again. In the video below, NBC Los Angeles from the night shift. Microsoft posted that its changes were "not taking as expected." Workers across the country reported "simply going home." Then the part everyone online connected it to. → Satya Nadella, April 2025: "maybe 20, 30% of the code" in some Microsoft projects is "written by software" → Microsoft in 2026: buyouts offered to about 9,000 people, about 4,800 jobs cut in July → analyst Rupert Goodwins to Al Jazeera: "all the experienced people, the people who built the stuff up, have left" Now the fair part. Microsoft says this wasn't a cyberattack, and it blames servers that went unavailable, not staffing or AI code. It was rolling back a network change on Friday. No root cause is out yet. Outages like this happened long before anyone wrote code with AI. But if your job is "keep the system running," this week made the case for you better than any résumé. Coincidence, or the bill for cutting the people who knew how it worked?
Microsoft services along with outlook is down for more than 12hrs now. When AI writes code, AI reviews codes, human engineers are mass laid off and your production code is totally dependent on Copilot for fixing but then copilot is also down, you get this Absolute Cinema.
3
5
1,516
Qantas is cutting 850 office jobs and says the reason is AI. The work goes to India first. The AI comes "in time." On Tuesday CEO Vanessa Hudson emailed staff: the "scale and speed of change in technology and AI is something none of us has seen in our lifetime." Finance, procurement, HR and marketing roles in Sydney move to Accenture, delivered "mostly by its teams in India." The first wave lands by mid-2027. In the video below, 7NEWS and ABC lay it out. Read Qantas's own statement closely: the outsourced work will "in time be transformed further with automation and AI." So the order is people in India now, software later. The union says it was told in August there was "no plan to offshore any Australian jobs." Qantas says it never gave that assurance. The stock rose 6% that day. One more receipt. In August 2025 a federal court fined Qantas A$90 million for illegally outsourcing 1,820 ground workers in 2020. Five years later the same airline is moving another 850 jobs out, and this time the word on the email is AI. It's not alone. The Guardian counts about 4,550 Australian corporate jobs targeted in a month: NAB 2,500, Canva 1,200, Qantas 850. Two ways to read it. Qantas says it runs 1,000+ old systems that don't talk to each other and it "cannot hire at the pace this work needs." The union says AI is the cover story for cheaper labour. If your job got the same email, would you believe the AI part?
3
1
7
561
Polymarket took Nvidia's crown in 30 minutes. The market is "Largest company in the world on December 31," $10.05M traded. On the morning of November 4, Nvidia was the favorite at 64.5%. At the US open the day after the midterms, it went from 59% to 43.5% in half an hour and never came back. In the video below, the market pages and the full price history: → Nvidia, December 31: 64.5% to 31% → Apple, December 31: 36% to 67% → the November 30 version: Nvidia 75% to 13% in six days → two more legs down: the November 5 AI rout and the Huang news on November 8-9 The real scoreboard agrees for now. Apple $5.10T, Nvidia $4.31T. A gap of about $780B with seven weeks left. Now the part I find interesting. In the same week, Polymarket gives Jensen Huang only 14.5% to be out as CEO by December 31. The crowd isn't betting on a board firing him. It's betting his company stays second. The one date that can change it is November 17, Nvidia's earnings. Closing $780B needs Nvidia to outrun Apple by about 18% from here. Market: polymarket.com/event/largest… Read the rules before you size anything: it resolves on "a consensus of credible reporting" of market cap at the close on December 31, not on any single site. Is 31% too cheap with earnings a week out, or did the crowd call the top of the Nvidia era?
5
6
521
Wall Street knocked about $230 billion off Nvidia before the open. Polymarket gives Jensen Huang 87.5% to keep his job. Last night he told staff "I was unfaithful to my wife," and Nvidia's board opened an investigation into whether he was the target of Chinese intelligence. By 4:17 PM ET Polymarket had a market on it: "Jensen Huang announced resignation or removal by December 31?" In the video below, the market pages and every trade from its first night: → YES opened near 1¢, jumped to 18% within two hours, and has sat at 12-15% since → $152K traded in 12 hours → at 05:35 UTC YES printed 99¢ on a thin book, and one wallet paid 48¢ for $2,971 of YES, triple the price a minute earlier → the biggest position: Parz1vaI, 46,537 NO shares at an average of 84¢. About $39K at risk to make about $7K. All-time +$895.6K on $74.4M volume → he's also on NO in the second market, "Jensen Huang divorce in 2026?", which trades at 9.5% on $104K Read the rules before you size anything. Retiring counts, and so does stepping down to take another role at Nvidia. Nvidia's board says it has "no reason to believe" anything was compromised, and Huang says he'll cooperate. My read: the stock is pricing risk to the company. The prediction market is pricing a decision by a board that just said it has confidence in him. Those can both be right for a while. Is 12.5% too cheap for a CEO under a spy probe eight days before earnings, or is the crowd right that boards don't fire founders over this?
7
8
733
Voters said no to AI data centers 40 times out of 41. The market barely moved. I pulled the Polymarket markets on a statewide data-center moratorium right after this. In the video below, the pages and the numbers: → Indiana by end of 2027: 18%, up half a point this week, on $5.1M traded → Texas: 21.5%, flat. Missouri 19.5% (+4), Louisiana 18.5% (+3), Oklahoma 19.5% (-1) → any state by Dec 31 this year: 28.5%, up 2.9 Why so calm? Read the rules. These markets only resolve YES on a statewide law or constitutional provision with legal force. Pausing tax incentives doesn't count. A county or town vote isn't a state law. Forty local "no" votes are a signal, not a resolution. Then the part I didn't expect. Indiana 2027 has $5.1M in volume, but the biggest YES holder owns 5,000 shares, about $900. The biggest NO holder has 2,963 shares. That volume is churn. Nobody is sizing up on the ballots turning into state law, in either direction. My read: the crowd is pricing a headline, the market is pricing a legislature. Legislatures move slower than ballots and rarely in one session. But 40 of 41 is the kind of number that ends up in a campaign speech, and 18% for a full year and a half is not nothing. Market: polymarket.com/event/indiana… Is 18% too cheap after 40 of 41, or is the crowd confusing a ballot with a law?
JUST IN: Americans vote to restrict or ban new AI data centers in 40 of 41 local ballot races nationwide.
5
2
352
Seven markets. Seven wins. Two of them right before a New York Times poll. The NYT reported Friday that Vanderbilt researchers flagged a Polymarket account that bet big shortly before two Times/Siena polls went public. Polymarket says it's investigating. I found the wallet. It's called dubdubdub2. In the video below, its profile and its trades from Polymarket's public data: → Nebraska Senate: $190.8K on Ricketts (plus "no independent") between Oct 30 evening and Oct 31 morning, most of it in one hour. Per the NYT, the poll came out about an hour later. Profit: $77.1K → Michigan Senate: $622.6K in two fills on the morning of Oct 31, about 40 hours before Times/Siena's Michigan numbers. Profit: $180.8K → all-time: +$405,346 across just 7 markets, all won, zero shares ever sold Wallet: polymarket.com/@dubdubdub2?v… Here's what bugs me in both directions. It doesn't trade like a classic leak. A leak usually buys the cheap side and sells into the spike after the news. This wallet bought the favorite at 71 to 79 cents and held to the end. Before October it did the same thing on Mamdani at 89 cents and 49ers NO at 95.7. But that still doesn't explain why these two races, right before these two polls. The lesson for everyone else: every fill on Polymarket has a public timestamp. Researchers lined them up against the minute each poll went live. If your timing is that good, someone will check. Leak, or a whale who likes favorites?
6
4
316
He made $3.05M on the midterms. He kept $610K. The wallet is called ndb1. Joined Polymarket in March, +$4.76M all-time, $154.4M traded across 3,028 markets. It's #2 on this week's leaderboard. Wallet: polymarket.com/@ndb1?via=che… The headline trade: since March 30 he bought $10.44M of YES on "Democrats Sweep" in the Balance of Power market at 30.2 cents. Democrats took both chambers. That line paid +$3,054,830. In the video below I searched his closed positions for that one market. Nine lines came back: → YES on all five outcomes, not just the one that won → Other: -$868K. R Senate, D House: -$807K. D Senate, R House: -$379K on YES and -$383K on NO → R sweep NO: -$102K. Two small winners: R sweep YES +$98K, and +$0.41 on one tiny line Add it up. $35.2M bought across that one market. Net: +$609,970. That's 1.7% on the money he put through it. My read: this isn't someone who called the election. It's someone running a book on every result, buying the cheap outcomes and the favorite together so no single night could wipe him out. It worked. But the price of that safety was about $2.5M of the $3.05M. The screenshots of a $3M win never show the other eight lines. Would you hedge a 30-cent favorite across every outcome, or ride the one call and accept the risk?
5
1
6
308
11 green months in a row. Same platform as yesterday's -$3.2M. A trader called bernardbulletin posted his year last night. Started with about $16.6K. Every month of 2026 so far is green, the worst one +$6.31K. November alone: +$373K. Across all the platforms he trades, his chart shows $1.1M. Wallet: polymarket.com/@bernardbulle… I checked the Polymarket side. In the video below, his profile: +$407,817 all-time on Polymarket $7.5M volume across 2,461 markets +$162,370 in the last 30 days, mostly the midterms biggest win: Nithya Raman for LA mayor, bought at 32.8 cents, +$97,987 Then I pulled his 2,571 closed positions. 1,803 wins, 762 losses. A 70% hit rate, with a median entry of 63 cents. That's the whole story. At 63 cents you need to win about 63% of the time to break even. He wins 70%. Seven points over the line. Yesterday's daily #1 bought at 51 cents and won 50% of the time. One point under the line, on $511M of volume, and he's down $3.2M. He loses too. -$9,112 on Trump saying "Job" 20+ times at the State of the Union. -$8,140 on a "Two Percent" mention market. His worst ticket is about a tenth of his best one. Volume doesn't make you a good trader. The gap between your price and your hit rate does. What's your hit rate versus your average entry? Most people have never checked.
6
7
335
Polymarket's #1 trader today is down $3.2M. Yesterday one wallet topped the daily leaderboard with +$456,094. Overnight it collected $617,507 on a Grizzlies spread, $170,603 on the Flames and $119,247 on the Senators. Wallet: polymarket.com/profile/0x492… Then you switch the profile from 1D to ALL. Joined December 2025. $511.5M traded across 1,741 markets. All-time: -$3,165,527. In the video below, the profile and his record. I pulled all 1,749 of his closed positions: - 875 wins, 874 losses - won $124.97M on the winners, lost $128.25M on the losers - 94% of his bets were bought between 40 and 60 cents - 826 over/unders, 684 spreads, 239 moneylines Biggest single win: +$473K on Chelsea vs Bournemouth over 3.5 goals. Biggest single loss: -$467K on a college basketball over, Canisius vs Iona. Here's the math that matters. His median entry is 50.9 cents. At that price you need to win about 50.9% of the time just to stay even. He wins 50.0%. Under one point short, on half a billion dollars of volume, and it's enough. My read: this isn't a sharp having a bad year. It's someone flipping coins worth up to $400K a toss on NBA, NHL and college games, with no visible edge. The daily leaderboard put him at #1 because yesterday the coins landed. Tomorrow he can be at the bottom with the same strategy. So when you see a "top trader" screenshot, switch it to ALL before you copy anything. Would you call this a skilled bettor on a bad run, or just variance with a big bankroll?
5
4
340
$920K on red states. $3,885 left. Kansas hadn't sent a Democrat to the Senate since 1932. One Polymarket wallet, lomneivan, bet a lot of money that it still wouldn't. In the video below, his profile and the races that took most of it: - South Carolina: -$284.9K, in since July at ~86 cents - Kansas: -$229.5K, in since September at ~70 cents - Ohio: -$195.1K, in since August at ~44 cents - Iowa: -$88.2K - plus Maine and the balance-of-power market 21 election-night positions. $920,237 in, $3,885 left. His profile shows -$816,020 for the day. Before Tuesday he was up $224.7K on closed trades. These weren't longshots. At 70 or 86 cents the market agreed with him. That's what makes it worth studying. My read: he didn't make five bets. He made one bet five times. Every race hung on the same thing, the national mood. When it turned, it turned everywhere at once. Spreading money across states felt like diversification. It was concentration with extra steps. Compare the other end of the same night. Wallet 033033033 bought at 99.9 cents after the result was in, and has lost once in 868 positions. lomneivan bought the favorite months early and lost almost all of it in one evening. Was he wrong about the races, or wrong about how connected they were?
BREAKING: Polymarket projects Adam Hamilton has won Kansas' U.S. Senate race, becoming the first Democrat elected to Senate from the state since 1932 — 99% chance.
6
3
143
$1.4 million to win $1,530. While everyone watched the map last night, one Polymarket wallet called 033033033 waited until the result looked certain. Then it put $1,423,070 on "Democrats control the House" at 99.9 cents. Best case, it makes $1,530. That's not a one-off. That's the whole strategy. In the video below, the profile: joined November 2024 $183.5M traded across 880 markets total profit: +$318.2K 868 closed positions, exactly one at a loss: -$9 Almost every buy is at 99 cents or higher. His book right now includes "No" on Jesus returning before 2027 ($20,478 at 98.5 cents) and "No" on Trump saying "Burger" in October. Same night, the other end of the market: colombiaman, the fresh wallet that put $176K on a Republican sweep at 6.1 cents, sits at -$216.3K. My read: 033033033 isn't predicting anything. He's running a bank. He parks money in markets that are basically decided and collects about 0.17 cents per dollar traded. Across $183M, that's $318K. The risk fits in one line. At 99.9 cents, a single wrong resolution erases about a thousand wins. And this morning the House market's losing side still says "In Review." Who has the better business: the 6-cent dreamer or the 99.9-cent landlord? Trader: polymarket.com/@033033033?vi…
BREAKING: Polymarket projects Democrats have officially swept the 2026 midterm elections, seizing control of both the House & Senate — 99.9% chance.
6
1
7
370
$3.59M bet to win $191K. That's what 614 Polymarket wallets did in the 19 hours after one brand-new account put $176,201 on a Republican sweep at 6.1 cents. The account is called colombiaman. First trade ever, Nov 2 at 18:12 UTC. 2.9M shares in 4 minutes. If Republicans hold both the House and the Senate tonight, it gets paid $2.9M. Right now it's down $23.8K. I pulled every trade on the House, Senate and Balance of Power markets after that moment: 90c or higher (the favorites): $3,587,881 from 614 wallets. Best case, they make $191K combined 10c or less (the longshots): $299,640 from 378 wallets. Best case, $5.4M colombiaman alone is 59% of all the longshot money In the video below: the market pages, both sides of the trade, and the math. Everyone calls the 95c crowd smart money and the 6c guy a degen. I don't buy it. At 95c you have to be right 19 times out of 20 just to break even, and one miss wipes out 19 wins. At 6c you need to be right 1 time in 16. Both sides are pricing the same thing: how likely a surprise is. 614 wallets are selling insurance against one. One wallet is buying it. Polls close tonight. One side gets paid. Who has the better bet: 614 wallets collecting 5 cents, or one wallet paying 6 for a shot at $2.9M?
5
1
1
5
298
Someone bought the Broncos at 11 cents. Two touchdowns later: $432,554. Sunday, Chiefs vs Broncos. Fourth quarter, Denver down 13-24. On Polymarket, a Broncos win was trading at 11 cents. One wallet put $47,105.13 in at that price. Denver scored twice, won 28-24, and the ticket paid $432,554. The video below shows the market, the swing on the chart and the ticket itself. That's 9.2x in one quarter, roughly $385,000 in profit, on a moneyline that traded $4.8M. Here's the part that matters if you trade this stuff. An 11-cent price means the market thinks this happens about 1 time in 9. Buying it is only a good trade if you're right more often than that, consistently, at size. One win doesn't tell you that. And you only saw this ticket because it won. The account that posted it doesn't post the comeback bets that expired at zero. Survivorship bias is the best marketing a betting product has. What I'd want to know before calling it smart: does this wallet do this every week, and what's its record on 10-15 cent live bets? If it's profitable over hundreds of them, that's a real edge in live-game pricing. If it's one shot, it's a lottery ticket that hit. Edge or lottery?
3
5
255