Mintblade: Five Markets, $9.1 Million, and the Small-Sample Trap

Mintblade (0x96cfcb0c30942cfcd1cdf76c7d408794d66b1acb) ranked sixth on Polymarket's all-time PnL leaderboard on 8 November 2026 with $9,110,689.71 in reported profit on $17,759,922.23 of volume. The ratio is striking: reported PnL is about 51% of reported volume, a high figure among the top ten.

Then comes the detail that should slow any reader down. The public profile says five markets traded. The account joined in June 2026. The biggest win was $7.2 million. There are no open positions. This is not a long track record with a high win rate. It is a very small number of very large event positions, most of them in the 2026 World Cup.

Five markets is not enough to establish a repeatable edge. It is enough to produce a spectacular leaderboard rank.

The numbers

The snapshot row showed $9.11 million of PnL on $17.76 million of volume, with PnL rank 6 and volume rank 1,644. Figures change; these are point-in-time readings. The volume rank is the lowest of the seven names in this series except fishalive, which is the point: this account did not get here by turning over a large book.

The profile page showed the account joined in June 2026, had 88,100 profile views, had traded five markets, and displayed a biggest win of $7.2 million. It showed $0.00 in positions value. The profile's total was +$9.2 million, a small rounding difference from the Data API's $9.11 million. The newest public activity was dated 8 July 2026.

The biggest win is about 79% of the all-time total. If that figure is a net result, the other four markets net roughly $1.9 million combined. That is still a large number, but it tells you how much of the record depends on one outcome.

What five markets look like

The public activity is dominated by 2026 World Cup markets settled in June 2026. Visible events include:

• A 200,000-share No position in "Will IR Iran win on 2026-06-15?" bought at 50 cents for $101,500.
• A redemption of $14,366,350.60 on that Iran-New Zealand No position.
• A redemption of $2,152,993.68 on a Uruguay No position in the Saudi Arabia versus Uruguay match.
• A redemption of $187,501 on a Saudi Arabia versus Uruguay draw position.
• Three rebate payments: two taker rebates of $18,721.49 and $9,156.41, and a maker rebate of $912.59.

The rebates total about $28,790, or roughly 0.3% of the reported PnL. They are not the reason the account ranks sixth. The settlement payouts are.

The activity feed also shows several fills at 50 cents in the Iran market, including a 200,000-share block. That is how a position gets built: not necessarily one order, but a series of fills. The profile's "markets traded" counter counts markets, not orders. Five markets can contain many fills.

The payout trap

The $14,366,350.60 redemption is the number that makes the account look extraordinary. It is a payout, not a profit. A redemption of 14.37 million shares at one dollar means 14.37 million shares were held to resolution. The cost of those shares is not in the redemption row.

The visible 200,000-share buy at 50 cents cost $101,500. If that were the entire position, the payout would be $200,000, not $14.37 million. The redemption is larger because the position was accumulated across many fills, or because the visible page is only a slice of the activity. Either way, the gross payout and the net profit are different numbers. The leaderboard's $9.11 million is the net figure; the activity feed's $14.37 million is gross.

This is the single most common mistake in reading Polymarket activity. A large redemption looks like a large win. It is actually the return of principal plus gain, and it says nothing about the cost basis unless you reconstruct the buys.

Small-sample arithmetic

Five markets is a tiny sample. Each market is 20% of the record. One loss would move an apparent win rate by 20 percentage points. A five-for-five record and a four-for-five record look very different in a table, but they are statistically indistinguishable at this sample size.

The rule of three makes the point concrete. If a record shows zero losses in five trials, the 95% upper bound on the underlying loss rate is still about 60%. In other words, five wins are consistent with a process that loses more often than it wins. The sample simply cannot rule that out.

That does not mean mintblade's process is bad. It means the public record cannot tell us whether it is good. The same five results could come from a sharp tournament model, a single correct read on one match, or a favorable draw. The data does not contain the counterfactual.

The 2026 World Cup is also a specific environment: a concentrated schedule and a small number of matches with global attention. A five-market record built around it is not evidence about behavior in ordinary months.

Why the 100% win-rate look is an artifact

Polymarket does not publish a win rate on the profile. It publishes a PnL total, a biggest win, and a market count. When an account has only a handful of markets, the visible record can look pristine for a simple reason: there are very few rows, and the biggest win dominates the display.

A five-market profile with no open positions is especially prone to this. The resolved winners are visible as large redemptions. The losers, if any, are smaller rows or are buried in the activity feed. A reader who counts only the visible winners can construct a 100% win rate that the underlying data never claimed.

The honest interpretation is not "five wins out of five." It is "five markets, one of which produced 79% of the PnL, in a single tournament, with no independent verification of the entry cost." That is a description of a sample, not a strategy.

What cannot be inferred

The public data does not show mintblade's bankroll, deposits, withdrawals, or position-sizing rules. It does not show whether the World Cup positions were hedged, whether the account had losing legs in the same matches, or whether the same wallet has other activity under a different name. It does not show the maximum drawdown, because there is no time series beyond a handful of settlements.

It also does not show whether the edge, if any, survives outside a World Cup. A tournament is a compressed calendar with a known end date. A trader who is good at pricing international football might be mediocre at club football, tennis, or politics. Five markets cannot answer that.

Leaderboard PnL can include unrealized gains when positions are open. Mintblade had none on this date, so the reported figure was settled.

Public PnL is not proof of skill. In this case, the PnL is proof that a small number of large event positions resolved favorably. The sample size is the story.

What to take away

Mintblade is the clearest illustration on the leaderboard of why sample size comes before rank. A $9.1 million PnL built on five markets is impressive as an outcome. It is not a track record you can generalize from, and it is not a strategy you can copy: the positions are closed, the account was not active at the snapshot, and the entry costs are not visible in the headline rows.

If you are evaluating a small-sample trader, ask three questions. How many independent markets are in the record? How much of the PnL comes from the single largest win? And can you reconstruct the cost basis of the biggest position from the public fills? If the answer to the last question is no, the PnL is a headline, not an analysis.

That standard would demote most of the leaderboard. It would also make the remaining names more interesting.