How did "smart money" with a total profit of USD 820,000 end up with a net loss of USD 230,000? I. Kkunn's Two Turnarounds At 4:43 AM on November 6th, a deposit of USD 19,952 was made into a Polymarket wallet named Kkunn. After removing this newly arrived money, this wallet, which had been a top performer in multiple leagues over the past three months, generating profits exceeding USD 800,000, was left with only USD 94.39. Kkunn used to be my main trading partner. It made a lot of money in women's tennis, and my trading followed its significant drawdowns. Today, re-analyzing the wallet's transaction history, I want to understand: how could someone who accumulated over USD 500,000 in net profit on the WTA end up with a cumulative loss of USD 230,000 in the same account? What most easily traps people on this trajectory is that it actually managed to recover before. On September 27th, the official cumulative return curve was still +80,800 USD. By October 3rd, it had plummeted to -128,100 USD. Then, it rebounded sharply. On October 22nd, the curve surged to its highest verifiable observation: +382,800 USD. This recovery from the deep trough far exceeded previous profits. Anyone who followed this phase could easily interpret the next drop as another temporary pullback. However, four days later, on October 26th, the curve had fallen to -74,100 USD. On October 30th, it briefly returned to +21,400 USD. This small rebound was enough to encourage further waiting. Then, the cumulative return continued to decline, reaching approximately -230,000 USD on November 6th. From the highest observation on October 22nd to the latest point, the cumulative return has fallen by $613,000 USD. This refers to the difference in profit/loss, not the drawdown rate. The previous successful comebacks made another one seem very likely. However, every loss incurred in the account has already occurred. Breaking down the leagues, the answer is even more glaring. WTA Women's Tennis: Net profit of $504,200. ATP Men's Tennis: Loss of $564,100. CS2: Loss of $216,600. Other markets combined: Loss of approximately $2,900. A total of $780,700 was lost in ATP and CS2 alone. All the net profit accumulated in the WTA couldn't even cover these two losses. Total cash profit/loss for all trades: -$279,500. Even adding back the $49,100 in platform rebates and rewards, the cumulative cash profit/loss remains -$230,400. By retrospectively analyzing complete deposit and withdrawal records and transaction flows, the on-chain pUSD balance at the time of verification matches. Therefore, the $500,000 profit on the WTA is real, and the $230,000 loss on the overall ledger is also real. A wallet that makes money in one league can still lose it back in other leagues. II. Even the most profitable league is unreliable. At this point, it's easy to draw a seemingly clever conclusion: just follow its WTA. I initially valued its performance on the WTA more. But breaking it down further, this answer is also untenable. As of the time of verification, the 62 WTA markets entered for the first time in the last 15 days have collectively lost $223,900, with an ROI of -13.63%. This figure is grouped by the market's initial purchase date, calculating the total transaction cash flow from the initial purchase date to the deadline. It shows that the areas where Kkunn excels most have recently run into problems. The two purchase directions on the Ningbo site are particularly glaring. Against Rybakina, it placed four initial bets on Shnaider, totaling approximately $105,300 including fees, all of which were lost. Against Fernandez, it placed seven initial bets on Bouzkova, totaling approximately $104,200 including fees, also all of which were lost. The total loss across both directions was $209,500. The saying "lose a few games, win one more and you'll make it all back" has an easily overlooked premise: the losing positions must be small enough. Once $100,000 is bet in each direction, this premise disappears. Kkunn has a history of placing many WTA bets in the low to mid-price range. These positions, if successful, offer high returns, and a few big wins can create a very attractive profit curve. Based on the current complete sample, the five most profitable markets in the WTA contributed a total of $354,200, equivalent to about 70% of the WTA's total net profit. Even excluding these 5 markets, the remaining markets still netted approximately $149,900. It did indeed make money in the past. However, such a profitable record couldn't prevent later heavy-position mistakes. More noteworthy is its behavior after price drops. In the sample verified before October 24th, this situation occurred in 32 directions: the subsequent average purchase price was at least 10 cents lower than the initial average purchase price, yet it continued to add to its position. Ultimately, only 9 of these 32 directions were profitable. This doesn't prove that not adding to the position would necessarily be better, nor can it be used to judge the wallet owner's psychology. But for copy traders, the risk is very concrete: when the price moves against them, their investment may continue to increase. What we follow is not only the profit when it correctly predicts the direction, but also the process of it continuing to invest after it's wrong. III. The Wear and Tear of Copy Trading Copy trading adds another layer of trouble: the profitable portfolio of the source wallet often changes by the time it reaches me. (1000 words of operational details omitted here) Just because we're both in the WTA and have even bet on the same direction doesn't mean we're getting the same returns. After this review, I'm more concerned with: which markets did the profits come from, and what position sizes were used to generate them; and where did the funds go when losses occurred? A beautiful total return figure can mask these differences. Kkunn is currently in my observation pool. The deposit from this morning has arrived. Whether it can recover, I don't know. What I do know is that the new $20,000 has given this address cash again; the previously lost $230,000 remains in the main account. Data as of 11:49:19 Beijing time, November 6, 2026. Trading cash profits and losses include fees; rebates and rewards are additional. The official curve ended at 11:00 AM that day, approximately -$230,200; the cash reconciliation showed -$230,400, a difference of $117.91, with each figure retained separately. The $94.39 refers only to the pUSD balance at that address before the deposit and does not represent the holder's total assets.