🔍 Read the full analysis: From Viral “100X Trade” Strategy To 13,000 Test Results on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Thorsten Meyer AI says it tested two versions of a viral Polymarket “100x trade” strategy using simulated money across four crypto markets and about 13,000 windows. Both versions lost money; the paired strategy’s required double-fill occurred three times in 9,486 attempts. The results come from a paper-trading test, not live trading with real funds.
Thorsten Meyer AI reported on May 21 that two simulated versions of a viral Polymarket trading strategy lost money across about 13,000 five-minute market windows on BTC, ETH, SOL and XRP. The test challenges the repeatability of a video’s striking trade examples: the paired strategy’s key double-fill happened just three times in 9,486 attempts, while the more frequent one-sided fills often left the simulated trader holding the losing position.
The strategy was based on trades shown by a YouTuber in Polymarket’s five-minute Bitcoin Up/Down market. According to the source article, examples included turning $1 into $50, as well as trades described as 48x and 100x. The source says the transactions could be checked on PolygonScan. Those examples establish that the trades occurred, but do not show how often the strategy could be repeated profitably. The YouTuber reportedly said the event was rare and that the video did not yet have results on whether it could be replicated.
Thorsten Meyer AI rebuilt the idea in Polybot as two separate simulated approaches, running them against live order books for two days across four coins. Its paper-fill model counted a simulated bid as filled only when a real taker sold into it. The article reports 24 parallel experiments and roughly 13,000 windows; it does not describe the tests as trades made with real funds.
In the paired version, the bot placed two-cent bids on both Up and Down. If both 50-share positions filled, one side would pay $50 at settlement, producing a reported $48 net gain on the $2 total cost, regardless of the outcome. But that double-fill occurred in 3 of 9,486 attempts. The source reports 1,297 loser-only fills, 22 winner-only fills and 8,162 attempts with neither side filled, with net simulated P&L of −$280.
Polymarket · Paper-trading report · May 21, 2026
From Viral “100X Trade” Strategy To 13,000 Test Results
A two-day simulation across BTC, ETH, SOL and XRP tested whether striking five-minute market trades could be repeated. Both reconstructed strategies lost money—and the paired trade’s crucial double-fill was exceptionally rare.
01 / How the test worked
From viral trade to replication test
The video showed large returns, including a reported $1-to-$50 trade and examples described as 48x and 100x. The follow-up asked how often a similar setup could capture fills across many attempts.
Five-minute market
Polymarket asks whether a crypto asset finishes above or below its opening price.
Buy both sides
Place two-cent bids on Up and Down. Each share costs $0.02.
Need both fills
If 50 shares on each side fill, the total cost is $2 and one side pays $50.
Measure every try
Paper fills counted only when a real taker sold into the simulated bid.
02 / Paired strategy
The edge depended on a rare event
The payout calculation works only if both low-priced sides fill. In the reported test, losing one-sided fills were much more common than the double-fill that could make the trade profitable.
3 double-fills among 9,486 attempts. Reported paired strategy net simulated P&L: −$280.
03 / Second approach
Post-close timing brought another risk
A separate “winner-snipe” version bid only on the side a price feed indicated had won, just after market close. Its results were also negative in the tested setup.
3,482 posts
The source reports eight fills from these posts.
4 of 8
Four fills settled; all four lost. The other four do not have a settled outcome reported in the supplied material.
Timing mismatch
The feed reading just after close did not always match Polymarket’s official resolution, according to the account.
04 / What the result can tell us
Execution matters as much as payout
A verifiable winning transaction shows that a trade happened. It does not show how often the same approach could be repeated profitably.
Count the misses
Rare double-fills and frequent loser-only fills shaped the paired strategy’s overall result.
Model the fills
Execution, time exposed, settlement rules and actual fill behavior all affect the outcome.
Keep the claim narrow
The findings apply to these reconstructions, markets and test period. They do not establish how every version or future market will perform.
Voices from the report
What was said
The source frames repeatability as an open question and describes its follow-up as part four of an ongoing series.
“This is going to take time because this is a rare event… probably I didn’t have anything for you in this video about the results.”
The YouTuber, as quoted by Thorsten Meyer AI
Test status: Every strategy described was run with simulated money. No real funds were involved at any point.
05 / Limits and next steps
A useful test, with boundaries
The report offers aggregate results from a two-day paper-trading test. The supplied source material does not include enough detail for an independent, full reproduction.
No full independent audit or detailed logs are provided in the material. The paper-fill model may not capture all live execution conditions.
No timestamp-by-timestamp comparison is supplied for the price-feed readings and official resolutions behind the four settled losses.
A longer run and more execution data could help assess repeatability, but no specific next experiment or publication date is announced.
Quick answers
Key questions
The reported results are specific to a simulated setup, not a record of live investment performance.
Buy both sides of a five-minute Up/Down market at very low prices. If both positions filled, the winning side’s payout could exceed the combined cost.
Did the test use real money?
No. Thorsten Meyer AI says it used simulated money only; no real funds were involved.
How often did both sides fill?
Three double-fills in 9,486 attempts, or about 0.032%. The paired version reported −$280 net simulated P&L.
Why did the post-close version lose?
The source attributes its four settled losses to a mismatch between the bot’s post-close feed reading and Polymarket’s official resolution.
Rare Fills Undercut the Trade’s Edge
The results matter because the viral examples highlight the size of a payout, while the test measures the frequency and cost of trying to capture it. In the paired strategy, profit depended on both sides filling at a very low price. The reported data show that this happened rarely, while loser-only fills were much more common. According to the source, changing how long the bids remained live did not produce a profitable result: longer exposure brought more loser-only fills, and the occasional double-fill gains did not offset those losses.
The second version illustrates a separate risk. It placed a bid after market close only on the side a price feed indicated had won. The source reports eight fills from 3,482 posts; four of those fills settled, and all four lost. Meyer’s account attributes the losses to timing: the price-feed reading made shortly after close did not always match Polymarket’s official resolution. That finding is specific to the tested setup and does not establish how every post-close strategy performs.
For readers evaluating viral trading claims, the report is a reminder that a verifiable winning transaction is not the same as evidence of a repeatable edge. A strategy’s outcome depends on execution, fill rates, settlement rules and losing attempts, not just its largest payout. The test supports caution about this particular reconstruction; it is not a guarantee about future market behavior.
crypto trading simulation software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Polymarket’s five-minute crypto markets ask whether an asset’s price will finish above or below its opening level. As described by Thorsten Meyer AI, the winning side pays $1 per share at settlement and the losing side pays nothing. The proposed paired trade attempts to buy both sides cheaply, so that a fill on each would cover either possible outcome. Its central challenge is not the payout calculation but obtaining both fills at the desired prices.
The source article says the original video used on-chain transactions to show unusually large returns, then used an AI coding agent to examine how the trades may have worked. The video’s creator reportedly acknowledged that the events were rare and that repeatability had not yet been established. The May 21 report is presented as a follow-up test of that question, as the fourth installment in an ongoing series.
Meyer’s article explicitly says the strategy tests used simulated money only, with no real funds involved. Its reported figures are outcomes from the described paper-trading model and testing period, rather than a verified record of live investment returns.
“This is going to take time because this is a rare event… probably I didn’t have anything for you in this video about the results.”
— The YouTuber, as quoted by Thorsten Meyer AI
As an affiliate, we earn on qualifying purchases.
Limits of the Paper-Trading Results
The report describes a two-day test and gives aggregate counts, but the supplied source material does not provide a full independent audit, detailed test logs or enough information to reproduce every result. It also does not establish whether the paper-fill model captures all execution conditions that would apply to a live trader. No real-money performance is reported.
The source says four winner-snipe fills settled and all four lost; it does not give a settled outcome for the other four fills. It attributes the losses to a mismatch between the bot’s price-feed reading after close and Polymarket’s official resolution, but the material supplied does not include a detailed timestamp-by-timestamp comparison. The report’s conclusion applies to the particular strategies, markets and period tested. Future results, other implementations and market conditions remain unknown.
As an affiliate, we earn on qualifying purchases.
Further Tests Could Check Repeatability
Thorsten Meyer AI presents this report as part four of an ongoing series. The supplied material does not name a publication date for another installment or set out a specific next experiment. A follow-up could add a longer test period, publish more execution data and examine how paper fills compare with actual live fills, but those steps are not announced in the source.
For now, the published result is a simulated test with negative reported returns for both versions. The original trades may be verifiable, but whether any revised approach could capture similar opportunities repeatedly has not been established.
paper trading platform for cryptocurrencies
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What was the viral “100x trade” strategy?
It was a proposed way to buy both sides of a five-minute Polymarket Up/Down market at very low prices. If both positions filled, the winning side’s payout could exceed the combined cost. The test found that the necessary double-fill was rare.
Did the test use real money?
No. Thorsten Meyer AI says the tests used simulated money only and that no real funds were involved.
How often did both sides fill in the paired test?
The report counted three double-fills in 9,486 attempts, or about 0.032% of those attempts. It reported a net simulated P&L of −$280 for the paired version.
Why did the post-close version lose when it selected the apparent winner?
The source attributes the four settled losses to timing: the bot’s price-feed reading just after close did not always match Polymarket’s official resolution. The report does not provide detailed timestamps for each case.
Do these results prove that no version of the strategy can work?
No. They report negative outcomes for two tested versions under the described simulated conditions. The source does not establish results for every implementation, longer time period or future market environment.
Source: ThorstenMeyerAI.com
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
