RRektrospect

0x85ecf584f25db6f146718b86d493e33c5af72052

0x85ec...2052 wallet audit

0x85ec...2052 audit. -$81,716 realised trading PnL across 90 closed position cycles, using the latest 10,000 public fills from Apr 16, 2026 to Apr 21, 2026; older public fills may exist outside this audit.

loss-dominatedA quick bucket assigned from realised trading PnL, closed position-cycle count, and whether the public fill source was capped. Data covered: Apr 16, 2026 to Apr 21, 2026. Classification basis: closed net pnl after fees available window.latest 10,000 fillsHyperliquid's public fills source is capped for very active wallets. This audit used the latest 10,000 public fills it could retrieve, covering Apr 16, 2026 to Apr 21, 2026. Older trades may exist outside this page, so lifetime claims are avoided.
ModeProfessional keeps the tone factual. Roast uses the same numbers but writes the commentary more sharply.
ProfessionalRoast
Read trader analysisShare on X
Current account valueAccount value from Hyperliquid's public account state when this audit ran. This is a balance snapshot, not proof of lifetime deposits or withdrawals.$97,903latest 10,000 fills
Realised trading PnLClosed trading profit or loss after fees in the data covered: Apr 16, 2026 to Apr 21, 2026. Deposits and withdrawals are not counted as PnL. Older public fills may exist outside this audit.-$81,716-45.5%
Max drawdownLargest fall from a previous balance high to a later low inside the data covered: Apr 16, 2026 to Apr 21, 2026.-5.1%90 closed position cycles
Win rateShare of closed position cycles that ended positive. Profit factor compares total winning realised PnL with total losing realised PnL.+67.8%0.64 profit factor
Total volumeGross notional traded across 10,000 reconstructed public fills. A position cycle can contain many individual fills.$314,261,93190 position cycles
Trading PnL vs transfersRealised trading PnL comes from Hyperliquid closed-fill profit and loss. Deposits and withdrawals can change account value, but they are not counted as trading PnL here.

The dollar PnL is the realised result from closed trades in the data covered. The percentage uses an inferred starting value (current account value $97,903 minus closed trading PnL -$81,716 = starting estimate $179,618). This audit does not ingest a deposit or withdrawal ledger, so it can show that trades lost money, but it cannot prove whether the owner also moved funds in or out. Older fills may also exist outside the latest 10,000-fill window.

Data coveredHyperliquid's public fills source is capped for very active wallets. This audit used the latest 10,000 public fills it could retrieve, covering Apr 16, 2026 to Apr 21, 2026. Older trades may exist outside this page, so lifetime claims are avoided.Apr 16, 2026 to Apr 21, 2026

This is not a fixed last-week or last-month period. It is the actual span covered by the latest 10,000 public fills Hyperliquid exposed for this wallet. Because the public fill source hit its cap, older trades may exist but are not included here.

Public fills
10,000
Position cycles
90 closed
Limit
latest 10,000 fills only
Equity curveA historical line showing how the wallet balance moved across the data covered: Apr 16, 2026 to Apr 21, 2026. It is not a prediction.$97,903
latest fills onlyHyperliquid's public fills source is capped for very active wallets. This audit used the latest 10,000 public fills it could retrieve, covering Apr 16, 2026 to Apr 21, 2026. Older trades may exist outside this page, so lifetime claims are avoided.
Equity curve by date and account valueX-axis shows date. Y-axis shows account value in US dollars. The line starts at Apr 16 with $180k and ends at Apr 21 with $98k.Account value (USD)Date$226k$149k$71kApr 16Apr 18Apr 21
Audit summaryA short extract from the full trader analysis below. It is built from the stored numbers and evidence pack.What matters immediately
  • Data used: latest 10,000 public fills from Apr 16, 2026 to Apr 21, 2026; older public fills may exist outside this audit because the source hit its cap.
  • This account is -45.5% in the data covered, having started with approximately $179.6k and fallen to $97.9k.
  • The headline loss masks a catastrophic behavioural collapse: the account began with edge (a 29.5k win on 17 April), then entered a revenge-trading spiral on 19–20 April that produced five oversized losses totalling $151.3k against a median loss of $556.
Analysis readoutA plain-language interpretation layer from the trader analysis. Use the cards and tables below for the raw evidence.Strengths & weaknesses
  • Directional accuracy exists but is overwhelmed by sizing discipline failure. The account won 67.78% of trades. On the single trade where it sized rationally (17 April long, $29.5k profit), it was right and made money. Every subsequent loss was paired with position notional that scaled into the deepest decline in this window rather than away from it.
  • Revenge trading is documented and severe. Five trades opened immediately after losses, four of them on 19–20 April. The largest loss ($55.9k) was followed within minutes by the second-largest loss ($40.6k). The account had no cooling-off mechanism and no position-size reset after adversity.
  • Averaging down into falling positions is a consistent pattern. The top win (17 April) used averaging down successfully; the second-largest loss (18–19 April) used it catastrophically. The difference was not skill—it was that one trade moved 1.35% against entry and the other
Trader analysisThis is the full written analysis for this wallet and mode. The metrics, flags, simulator, and tables below are the supporting evidence.Full trader analysis

Bottom line up front

Only the most recent public fills are visible, so this audit covers the data covered rather than full account history. This account is -45.5% in the data covered, having started with approximately $179.6k and fallen to $97.9k. The headline loss masks a catastrophic behavioural collapse: the account began with edge (a 29.5k win on 17 April), then entered a revenge-trading spiral on 19–20 April that produced five oversized losses totalling $151.3k against a median loss of $556. Fees consumed an additional $5.5k in net drag. The single dominant pattern is position sizing that scales into losses rather than away from them.

What the data shows

The data covered spans 4 days of trading (16–21 April 2026) across 90 closed episodes in ETH only. The account opened with roughly $179.6k and reached a highest balance in this window of $2.91m on 18 April before collapsing to a lowest balance in this window of $2.76m on 20 April—a deepest decline in this window of 5.06%. The account closed the window at $97.9k, a net loss of $81.7k realised.

Long positions lost $43.9k (66.67% win rate); short positions lost $37.8k (68.89% win rate). Neither side showed edge. The account made money on exactly one trade of substance: an ETH long opened 17 April at $2,329.55, closed at $2,342.38 after 8.5 hours, with averaging-down behaviour and a notional highest balance in this window of $1.96m, yielding $29.5k. This trade worked because the account sized rationally and the market cooperated.

Everything after 18 April was a loss sequence. The account then entered a documented revenge-trading cycle: after a $556 loss on 19 April, it opened a $1.54m notional position. After a $20.4k loss later that day, it opened a $2.78m notional position. After a $859 loss, it opened a $2.93m notional position. After a $4.6k loss, it opened a $1.57m notional position. After a $55.9k loss on 20 April, it opened a $4.08m notional position on 20 April at 13:09, held for 1.83 hours, and lost $40.6k. The five largest losses in the data covered—totalling $151.3k—all occurred within this 24-hour window and all carried the "revenge_trade" flag. The account's structural stops (ATR 14 1h) were set 0.87–1.37% away; the account was running 50–60x leverage into a $2.9m balance, making any adverse tick lethal.

Fees paid totalled $12.8k gross; net fee drag was $5.5k after rebates. Realised PnL before fees was -$76.2k. The account was profitable on 67.78% of closed trades but lost 3.2x as much per loss as it made per win ($7,910 average loss vs $2,421 average win), producing a profit factor of 0.64 and an expectancy of -$908 per trade.

Trade quality

Win rate of 67.78% is superficially strong but is a trap: the account won two-thirds of its trades and still lost 45% of capital. Profit factor of 0.64 means every dollar of gross profit was paired with $1.56 of gross loss. Win/loss ratio of 0.31 is the operative metric: the account's winners were one-third the size of its losers. Expectancy of -$908 per trade is the mathematical summary: on average, each closed episode destroyed $908 of equity.

This is a classic case of high win rate masking negative expectancy through position sizing asymmetry. The account was right more often than it was wrong, but wrong in the direction that mattered: size.

Post-mortems

ETH long, 20 April 13:09–15:52, $2,323.30 entry to $2,301.24 exit: -$40,591.64

This trade carried three flags: FOMO re-entry, oversized loser, and revenge trade. The account had just closed a $55.9k loss (an ETH short opened 19 April at $2,260.20, exited at $2,289.31 after 13.97 hours). Within minutes, it re-entered long at $2,323.30 with a notional position of $4.08m—the largest single position in the data covered. The structural stop was 1.28% away. The trade lasted 1.83 hours and hit the stop, closing at $2,301.24. The account had no margin for error and no time to be right. This was pure revenge sizing.

ETH long, 18 April 15:26–19 April 00:19, $2,351.65 entry to $2,320.12 exit: -$32,508.99

This trade carried two flags: averaging down and oversized loser. The account opened at $2,351.65 and added to the position five times as price fell, reaching a notional highest balance in this window of $2.79m. It held for 8.53 hours and exited at $2,320.12, a 1.35% move against the entry. The structural stop was 0.87% away—tighter than the actual loss. The account was averaging into a falling knife and sized the position large enough that a normal intraday move became a $32.5k loss. This is the second-largest loss in the data covered and the clearest example of averaging-down discipline failure.

What the risk simulator reveals

Under a 1% hard stop rule applied historically, the account would have realised -$40.1k (vs actual -$81.7k), with a deepest decline in this window of -41.34% and a win rate of 58.43%. Six episodes would have been stopped early.

Under a 2% rule, the account would have realised -$80.1k (nearly identical to actual), with a deepest decline in this window of -70.43%.

Under a 4% rule, the account would have realised -$160.3k, with a deepest decline in this window of -108.67%.

The simulator is gross of fees. The 1% rule would have cut losses in half by preventing the largest revenge trades from running to full catastrophe. The 2% rule would have made almost no difference because the account's structural stops were already tight; the damage was done at entry, not at exit.

Open positions

No open positions at window close.

Honest summary

  • Directional accuracy exists but is overwhelmed by sizing discipline failure. The account won 67.78% of trades. On the single trade where it sized rationally (17 April long, $29.5k profit), it was right and made money. Every subsequent loss was paired with position notional that scaled into the deepest decline in this window rather than away from it.
  • Revenge trading is documented and severe. Five trades opened immediately after losses, four of them on 19–20 April. The largest loss ($55.9k) was followed within minutes by the second-largest loss ($40.6k). The account had no cooling-off mechanism and no position-size reset after adversity.
  • Averaging down into falling positions is a consistent pattern. The top win (17 April) used averaging down successfully; the second-largest loss (18–19 April) used it catastrophically. The difference was not skill—it was that one trade moved 1.35% against entry and the other

Behaviour checksRule-based warnings found in the trading history. They are not moral judgements; they mark patterns worth reviewing.

Rule-based position-cycle checks
FOMO re-entryReopened the same market and direction soon after a winning close, but at a worse entry.
29
Examples
  • ETH on Apr 16, 2026: re-entered at 2,343.4 after closing at 2,349.85 (Apr 16, 2026 prior close); outcome $27.
  • ETH on Apr 16, 2026: re-entered at 2,342 after closing at 2,342.4 (Apr 16, 2026 prior close); outcome $2,496.
+27 more matching cycles
Averaging downAdded size while the position was already moving against the entry.
43
Examples
  • ETH on Apr 16, 2026: added to the position; while it was already moving against entry; outcome $1,166.
  • ETH on Apr 16, 2026: added to the position; while it was already moving against entry; outcome $393.
+41 more matching cycles
Oversized loserA losing position cycle more than 3x the wallet's median closed loss.
11
Examples
  • ETH: -$4,913 realised loss; 8.8x median closed loss.
  • ETH: -$17,122 realised loss; 30.8x median closed loss.
+9 more matching cycles
Revenge tradeOpened a larger-than-normal position within one hour after a closed loss.
9
Examples
  • ETH on Apr 19, 2026: followed a -$557 loss; larger-than-normal size.
  • ETH on Apr 19, 2026: followed a -$20,396 loss; larger-than-normal size.
+7 more matching cycles
ExpectancyAverage result per closed position cycle after wins and losses are blended. Positive means each completed cycle added money on average.-$907.95
Fees / realised PnLFees as a share of realised trading PnL. High values mean execution cost is eating a meaningful part of the edge.n/a
Maker fill rateShare of fills that added liquidity rather than crossed the spread. Higher maker share usually means more patient execution.+51.9%

Expectancy is not a forecast. It is the historical average result per closed position cycle in this reconstructed sample.

Risk simulatorA counterfactual replay of the same historical trades using fixed risk limits. It is for comparing risk shape, not predicting future returns.

Replays the same closed position cycles with 1%, 2%, and 4% account-risk sizing. It shows what the wallet would have made or lost if each eligible cycle was sized from account value at entry and a structural stop.

1% account-risk ruleThis scenario limits each eligible position cycle to about 1% of account value at the simulated stop.-$40,067
Max drawdownLargest high-to-low account-value drop inside this simulated replay.
-41.3%
Stopped earlyHow many historical position cycles would have exited before the real close because the simulated stop was hit.
6
2% account-risk ruleThis scenario limits each eligible position cycle to about 2% of account value at the simulated stop.-$80,135
Max drawdownLargest high-to-low account-value drop inside this simulated replay.
-70.4%
Stopped earlyHow many historical position cycles would have exited before the real close because the simulated stop was hit.
6
4% account-risk ruleThis scenario limits each eligible position cycle to about 4% of account value at the simulated stop.-$160,269
Max drawdownLargest high-to-low account-value drop inside this simulated replay.
-108.7%
Stopped earlyHow many historical position cycles would have exited before the real close because the simulated stop was hit.
6

The 1%, 2%, and 4% rules are account-risk limits per position cycle, not leverage settings. If the simulated stop is breached, the cycle is stopped early. Outputs are gross of fees and funding, so use them as risk-shape comparisons rather than exact alternate realised trading PnL.

Equity curve by date and account valueX-axis shows date. Y-axis shows account value in US dollars. The line starts at Apr 16 with $186k and ends at Apr 21 with $100k.Account value (USD)Date$255k$165k$76kApr 16Apr 18Apr 21

Top lossesThe largest realised losing position cycles in the data covered by this audit.

Click a row for the trade breakdown
MarketThe traded Hyperliquid market or coin.SideLong means the wallet benefited if price rose. Short means it benefited if price fell.SizeLargest notional exposure reached during the reconstructed position cycle.PnLRealised profit or loss when the position cycle closed.DateClosed date when available; otherwise the cycle open date.

Top winsThe largest realised winning position cycles in the data covered by this audit.

Realised position-cycle outcomes
MarketThe traded Hyperliquid market or coin.SideLong means the wallet benefited if price rose. Short means it benefited if price fell.SizeLargest notional exposure reached during the reconstructed position cycle.PnLRealised profit or loss when the position cycle closed.DateClosed date when available; otherwise the cycle open date.
ETHlong$1,960,965$29,4832026-04-17
ETHshort$1,908,688$15,4442026-04-21
ETHshort$1,538,816$8,9522026-04-19
ETHlong$2,444,389$8,5122026-04-20
ETHshort$2,081,268$8,1572026-04-21

By marketBreaks the audit down by traded market or coin so you can see which markets helped or hurt the account.

Realised results by coin
CoinThe traded Hyperliquid market.CyclesClosed reconstructed position cycles for this market. One cycle can contain many fills.WinShare of that market's closed position cycles that ended positive.PnLRealised PnL attributed to this market's closed position cycles in the data covered by this audit.
ETH90+67.8%-$81,716
Share this audit on X