kch123: $11.35 Million PnL and a Position List Full of Zeroes

kch123 (0x6a72f61820b26b1fe4d956e17b6dc2a1ea3033ee) ranked fifth on Polymarket's all-time PnL leaderboard on 8 November 2026 with $11,351,406.80 in reported profit on $293,444,002.75 of volume. That is a strong headline. The public position list is a useful antidote to it: a series of sports bets that resolved to zero, including one NBA position that cost nearly half a million dollars.

The contrast is the point. A cumulative PnL number nets winners and losers into a single figure. The position list shows the losers individually. If you are thinking about copying a trader, the losers are the part you need to understand.

The headline numbers

The Data API row showed $11.35 million of PnL on $293.4 million of volume, with PnL rank 5 and volume rank 39. Figures change; these are point-in-time readings. Reported PnL is about 3.87% of reported volume, a middle figure between the high-frequency sports accounts and the concentrated event bettors.

The profile page showed the account joined in June 2025, had 941,200 profile views, had traded 2,613 markets, and displayed a biggest win of $1.1 million. It showed $0.00 in positions value. The past-day PnL was $0.00. The newest public activity was dated 9 July 2026, so the account was not active on snapshot day.

The biggest win is about 9.7% of the all-time total. That is more diversified than a two-market or five-market record, but it is still concentrated: one result accounts for roughly a tenth of everything the account has earned.

The zeroes

The position endpoint returned a list of resolved, redeemable positions marked at zero. The largest visible losses were:

• Knicks versus Cavaliers, Cavaliers at 54.89 cents: 901,927.73 shares, entry cost $495,100.79 plus $3,427.13 in fees, current value $0.00.
• Thunder versus Spurs, Spurs at 53 cents: 385,239.8 shares, cost about $204,177.10, current value $0.00.
• Spread: Hurricanes -1.5, Senators at 66 cents: 150,000 shares, cost $99,000, current value $0.00.
• Will the Oklahoma City Thunder win the 2026 NBA Finals, Yes at 37.7 cents: 241,141.3 shares, a loss of about $90,890.74, current value $0.00.

Those four positions alone represent about $889,000 of cost that went to zero. Add the other visible losers, including a $22,630.67 loss on an Islanders position, a $21,875.37 loss on an Avalanche Stanley Cup position, and a $794.45 loss on an Islanders futures position, and the total is close to $941,000.

That is not the same as a drawdown. It is a collection of resolved losses, and some of them may have been offset by hedges or by profits elsewhere in the same event. But it is a floor on the amount of capital that was exposed to total-loss outcomes. A leaderboard row that shows +$11.35 million does not show this list.

Concentration and drawdown

The visible losses cluster in two sports: NBA playoff markets and NHL playoff and Stanley Cup markets. That is not a diversified book. A set of positions on the Knicks, the Cavaliers, the Thunder, the Spurs, the Hurricanes, the Islanders, and the Avalanche is a set of correlated playoff exposures. When the playoff bracket breaks one way, several positions can lose together.

The Knicks-Cavaliers position is the clearest example. More than 900,000 shares at an average of 54.89 cents is a $498,528 commitment to one outcome in one game. The position resolved at zero. A trader can be right about the season and wrong about one game, and the position size turns that into a half-million-dollar loss.

The account's all-time PnL is still strongly positive, which means the winning positions outweighed these losses. But the public record does not show the order in which the wins and losses occurred. An $11.35 million endpoint could have been reached after a much larger peak, or after a deep trough. Without a time series, the maximum drawdown is unknown.

This is where the leaderboard is at its weakest. It reports the sum, not the path. A trader who risked a large fraction of capital on each playoff series and survived can look identical to a trader who sized conservatively and compounded. The endpoint does not distinguish them.

Leaderboard PnL can also include unrealized gains when positions are open. On this date kch123 had none, so the reported total was settled. That is a snapshot property, not a guarantee.

PnL quality

The PnL-to-volume ratio of 3.87% is higher than swisstony's 1.00% or RN1's 0.90%, but far below the 51% to 68% ratios of the concentrated event bettors. That is consistent with a mid-frequency sports operation: more turnover than a pure event bettor, less than a market-making machine.

The 2,613-market count is also mid-range. It is enough markets that the result is not a one-off, but not enough to establish a stable statistical edge across thousands of independent trials. The biggest win at 9.7% of the total suggests that a handful of results matter more than the average.

The visible zeroes add another dimension: the account took total-loss risk at size. That can be part of a profitable strategy if the winning positions are priced to compensate. It can also be a sign that the position sizing is aggressive relative to the bankroll. The public data cannot tell us which, because the bankroll is not public.

Why high PnL is not copyable edge

The practical problem with copying kch123 is timing. The positions that produced the visible losses are already resolved. The account had no open positions on snapshot day and had not traded since June 2026. There is nothing to mirror.

Even if the positions were open, a copier would face three disadvantages. First, the copier sees the position after the original fill, so the price may have moved. Second, the copier may not have the same bankroll, so a $498,000 position cannot be scaled proportionally. Third, the copier may not know whether the position is a hedge against another exposure. A bet that looks like a standalone wager could be one leg of a portfolio.

The deeper issue is that the losses are the strategy. A trader who takes total-loss risk at size is not running a strategy that can be evaluated by the winning percentage alone. You need the payoff distribution: how often the position wins, how much it wins, and how much it loses when it loses. The public leaderboard gives you the net, not the distribution.

Public PnL is not proof of skill. It is the sum of a set of outcomes, and in this case the outcomes include several hundred thousand dollars of positions that went to zero.

What to check before following a sports trader

If you are evaluating a sports-focused account, ask:

1. What is the largest single position as a share of the account's visible capital?
2. How many positions are correlated through the same playoff series or the same slate of games?
3. What is the worst visible loss, and was it hedged?
4. Is the PnL concentrated in a few large wins or spread across many small ones?
5. Is the account active now, and are the positions still open?
6. Does the profile's all-time PnL match the leaderboard's? If not, which surface are you using?

For kch123, the answers are: a $498,528 position, a cluster of NBA and NHL playoff exposures, a visible loss of that full amount, a biggest win equal to about a tenth of the total, no current activity, and a small surface gap between the profile's +$11.4 million and the Data API's $11.35 million.

That does not make the account a bad trader. It makes the account a case study in why a high PnL and a copyable edge are different claims.