"Buys Bitcoin at 13–32¢ and sells at $1": the profit came from 29 big bets, not thousands of small ones

The short version

The post went up on X on October 25, 2026 and has been viewed about 16,000 times. It describes Polymarket's 15-minute "Bitcoin Up or Down" markets and a trader who "waits until one side is priced as almost certain to lose, then buys it cheap", at 13–32¢ a share. It shows three winning trades, such as 6,322 "Down" shares bought for $864 that paid $6,321, and explains that at 15¢ you only need to be right about one time in six to break even. It credits "Cheap odds, strict position sizing, thousands of reps" and links the trader's profile, @berlinskies.

Every trade on Polymarket is public, so we downloaded the wallet's full history from Polymarket's data API: 31,243 records from July 22 to October 27, 2026. The three example trades are real; they are the wallet's three biggest wins, on October 5, 10 and 11 (the prices in the post include the trading fee). The month's figure is real too: Polymarket's own count of the wallet's profit before fees over the 30 days to October 25 is $62,371. The wallet paid $11,127 in fees over those 30 days, so its profit after fees was $51,244, plus $1,648 in fee rebates from Polymarket. Before October the account had barely made money: $3,434 after fees by September 30.

Most of the wallet's trading is not what the post describes. Of the $771,753 it spent buying shares, across 13,261 markets, 23.5% went into Bitcoin's 15-minute markets. About 70% went into esports, tennis, UFC and football, and about 7% into daily temperature markets, Bitcoin's 5-minute markets and other crypto; all of that together roughly broke even. Inside the Bitcoin 15-minute markets, most of its money bought at 32¢ or more ($146,615 of $181,669), and those buys made $19,622. The buying the post describes, at 13–32¢, was 139 positions costing $25,889: 3.4% of everything the wallet bought. They made $21,908.

That profit came from a short run of large bets. 29 of those 139 positions were $400 to $953 each, all placed between October 3 and October 23. 14 of the 29 won, where their prices implied about 6, and together they made $21,476. The other 110 cheap positions, of $1 to $400, made $432 between them, about break-even: they won 29 times where their prices implied 24.5. Stakes on cheap bets ran from $1 to $953 (median $36), so this is neither strict sizing nor thousands of reps. If those 29 prices were fair, winning 14 or more would happen about once in 700 tries, so luck alone is a stretch. But 29 bets in three weeks are too few to show a lasting edge, and the record cannot tell us why the trader chose them.

The post's arithmetic is right as far as it goes: at 15¢ a share, including the fee, one win in about 6.7 breaks even. But that is only the break-even point. Buying the cheap side pays only if it wins more often than its price says, and in this wallet that happened in 29 large bets, not across its cheap buying in general. Its buys below 10¢ lost 55% of what they cost. The post was published two days ago, which is too soon to judge what came after. Verdict: the trades and the month's profit are real, but the profit is before fees, most of it did not come from buying cheap Bitcoin shares, and the part that did came from 29 large bets.