Notes

A read-only rerun of a saved script called job8 was completed, and it found the 50 most recent items. Out of these, 2 items matched the keywords we were looking for. The first match was a headline about Brent and European markets at 16:46, and the second match was a headline about jobs commentary at 11:02. Neither of these headlines was a signal that a company had better than expected earnings, and there was no claim or confirmation of this inside the headlines. This manual check does not prove that the scheduled job8 was completed. The news found was a secondary discovery and did not meet the rules for entering a trade. https://runway.adityaperswal.workers.dev/data/news?limit=50 ; job8 script manually executed September7 19:41UTC

The QTA9071940 system, which is a type of automated trading fund manager, has completed its monitoring and audit tasks. All 6 quant monitors, which are tools used to track and analyze data, were armed after the next run on September 7 at 21:00 UTC. There are 8 active jobs, which are tasks being performed by the system, and 15, 35, and 11, 57, and 8 remaining jobs. Job 35 was completed at 17:31 with an exit code of 0, which means it was successful, and each issuer had 3 matches, with the same first notice only, but the system cannot prove the absence of news. The next run is scheduled for 21:31, and the last planned run is on September 9 at 01:31 UTC, with 8 runs every 4 hours, but this is insufficient until the reviews on September 9 at 12:30, so the system needs to be renewed before it expires. Job 15 remains unchanged, with all 19 eligible tasks having a status of 0. Job 8 shows the next run at 19:15 in the past, and the last run was at 07:15, but the scheduled completion is unverified and not assumed to have failed. There is no action required for exposure between runs. list_jobs fund=quant; list_monitors fund=quant September7 19:41UTC; https://inhibrx.investorroom.com/news-releases https://tyrabio.investorroom.com/news-releases https://ir.opusgtx.com/press-releases

The primary internal ledger, which is a record of all transactions, only includes IDs 9/7/1, and all of these have been closed by the operator through a process called liquidation, meaning there are no open, pending, or closing transactions. A programming tool called Python numpy was used to analyze the data, and it found that the average win was 33.333333%, the average win over many trades was 7.939941%, and the average loss was 5.261432%. The mean return, which is the average return of all trades, was -0.860974%. The total amount of money made from selling assets, known as gross realized, was -1.046053 USD. The quarter Kelly, a measure of how much to bet, was 0. The same outcomes happened, but the intended time frames of 4-6 weeks were cut short, and there were no quotes available when trades were made, so the probabilities of slippage, which is when a trade is made at a different price than expected, and stop/target, which are types of orders, are unknown. The quant cash, which is the amount of money available for trading, was 245.69, the equity, which is the total value of the account, was 850.99, the position cap, which is the maximum amount that can be invested in a trade, was 212.7475, and the risk cap, which is the maximum amount that can be lost, was 8.5099. The N3, which refers to a type of calibration, was not used for forward calibration, meaning it was not used to adjust the model for future trades. query_trades fund=quant September7 19:41UTC; supplied account context September7 19:40:38UTC

The 6armed quant system is monitoring 46, 59, 47, 48, 60, and 42, with no changes expected before the next run at 19:40 UTC. There are currently 8 active jobs, with 15, 35, and 11, 57, and 9 remaining. Job 35 is scheduled to run next at 17:15, and the last run at 13:15 resulted in an exit code of 0. There are 9 four-hour runs scheduled to end on September 9 at 01:15 UTC, before the reviews at 12:30 UTC on the same day. A renewal is required before expiration, but not before the next manager review. The latest collector has not changed the first link notices for 3 issuers, each with 3 matches, but the partial output cannot confirm if there is no new news. There has been no new clinical output since the prior audit. Job 8 is scheduled to run next at 19:15 in discovery-only mode, and job 15 has 0 eligible items at 19:00. No duplicate or new monitor is needed while the system is flat, which means it is not experiencing any significant changes or fluctuations. list_jobs fund=quant; list_monitors fund=quant September7 16:31UTC; https://inhibrx.investorroom.com/news-releases https://tyrabio.investorroom.com/news-releases https://ir.opusgtx.com/press-releases

A news scan found the headline that on September 7, at 16:31 UTC, it was noted that the primary exchange's 2026 calendar lists Monday, September 7, as a Labor Day holiday, with the publication date unspecified. The next core open is scheduled for September 8 at 13:30 UTC, according to the supplied schedule. There is no holiday intraday signal. https://www.nyse.com/trade/hours-calendars

A news scan found the data line: {"sample":3,"win rate":33.33333333333333,"average win":7.939941307427634,"average loss":5.261431763773078,"mean return":-0.8609740733728405,"realized USD":-1.0460528968119984,"quarter Kelly":0,"position cap":212.7475,"risk cap":8.5099}. This data shows the results of a trading fund, including the number of samples, win rate, average win, average loss, mean return, and realized profit in US dollars. The quarter Kelly value, which is a measure of the optimal fraction of wealth to bet, is 0. A fresh ledger was started with only one ID, 9/7/1, which was closed by the operator due to liquidation. There are no open, pending, or closing trades. The same three trades were considered, but there were no new outcomes. The intended trading period of 4-6 weeks was shortened, and the arrival quotes are unknown, which means the actual prices at which trades were executed are not available, and slippage, which is the difference between expected and actual prices, is also unknown. The equity, which is the total value of the fund, is 850.99, and the quantcash, which is a type of cash reserve, is 245.69. The quarter Kelly value is zero, which is an empirical, or observed, value, and is only descriptive. N3, which refers to a type of probability calculation, is not calibrated, meaning it is not adjusted to reflect forward probability, which is the probability of future events. query_trades fund=quant September7 16:31UTC; user account context September7 16:31:05UTC

On September 7 at 13:51 UTC, an audit was performed on the QTA9071350 system, which uses a type of monitoring called quant monitors, a way to track and analyze data. The audit checked 6 quant monitors with the numbers 46, 59, 47, 48, 60, and 42, and found that none of them need to be checked again before the next run at 16:30 UTC. There are currently 8 active jobs, with job numbers 8, 15, and 35, and 11, 57, and 9 remaining to be completed. Job 35, which is a task that runs at a set time, last finished at 13:15 and will run again at 17:15 UTC, after the next manager run, which is a scheduled check. The system has been set to run 4 times every 4 hours, extending to September 9 at 01:15 UTC, before the clinical reviews on September 9, so the system needs to be renewed in a later run before it expires. The latest output from the system lists the same existing notices for INBX, TYRA, and IRD, with 3 matches for each, but only the first matching link is shown, and there is no proof that there are no other news items. Job 8 found some false positives, which are incorrect matches, and not actual signals, and job 15 found that none of the 19 names being watched are eligible, and the monthly watch and range reject, a way to filter out unwanted data, was applied. There is no action required between runs, and the existing monitors will be kept. list_monitors fund=quant; list_jobs fund=quant; query_research note733; https://inhibrx.investorroom.com/news-releases https://tyrabio.investorroom.com/news-releases https://ir.opusgtx.com/press-releases

A primary exchange page retrieved on September 7 at 13:51 UTC lists Monday, September 7 as a 2026 Labor Day holiday. The publication date is not specified, but it covers the 2026 calendar. The next open session is scheduled for September 8 at 13:30 UTC. There is no holiday intraday entry signal. https://www.nyse.com/trade/hours-calendars

A news scan found the following data line: a list of trades for QTA9071350, which is a type of automated trading fund. The trades are for the symbols SNOW, DELL, and GTLB. The return percentage for SNOW was -5.392364480519829, with a profit and loss of -2.426024762767998, and the exit was due to operator liquidation, meaning a person made the decision to sell. The return percentage for DELL was 7.939941307427634, with a profit and loss of 6.509957861609997, and the exit was also due to operator liquidation. The return percentage for GTLB was -5.130499047026326, with a profit and loss of -5.129985995653997, and the exit was due to operator liquidation. There were 3 trades in total, with a win rate of 33.33333333333333, meaning one third of the trades were profitable. The average win was 7.939941307427634, and the average loss was 5.261431763773078. The mean return was -0.8609740733728405, and the total profit and loss was -1.0460528968119984. The quarter Kelly, a measure of risk, was 0, and the position cap, the maximum amount that can be invested, was 212.7475. The risk cap, the maximum amount that can be lost, was 8.5099. There were no open, pending, or closing trades. All three trades that exited on September 4 were due to operator liquidation, not because the strategy had run its course. There were no new closed trades or outstanding lessons to be learned. The arrival quotes, the prices at which the trades were made, are not available, so slippage, the difference between the expected and actual prices, is unknown. The account equity, the total value of the account, was 850.99, and the quant cash, the amount of cash available for trading, was 245.69. These values are provided as context, but have not been independently verified. The empirical Kelly, a measure of risk based on past performance, is descriptive, but not reliable for predicting future performance. query_trades fund=quant retrieved 2026-09-07T13:51UTC; user account context 2026-09-07T13:50:36.129Z

A fresh operational audit was done on September 7 at 12:33 UTC. This audit found that quant IDs 46, 59, 47, 48, 60, and 42 were armed, but none are due now or before 13:50 UTC. There are currently 8 active jobs out of a total of 15, with 35 jobs in the system and 11 remaining. Job 31 was cancelled. Job 35 last ran at 09:15 UTC and will run again at 13:15 UTC. The first matching issuer link for INBX, TYRA, and IRD still uses the existing notice, with 3 matching links for each. Just because there is partial coverage does not mean there is no news. The primary listings for these companies can be found at https://inhibrx.investorroom.com/news-releases, https://tyrabio.investorroom.com/news-releases, and https://ir.opusgtx.com/press-releases. Job 8 had false positives for earnings due to broad words, but this was just for discovery and no duplicate job is required. list_monitors fund=quant; list_jobs fund=quant; query_research notes699

A news scan found the headline 'QTA9071230 September 7 exchange holiday verified'. Retrieved on September 7 at 12:32 UTC from the exchange's primary calendar, it was verified that Monday, September 7, 2026, is Labor Day and there will be no core session. According to the user's clock, the next core session will open on September 8 at 13:30 UTC. There is no current-session intraday signal. The publication date is unspecified, but it covers the 2026 calendar. https://www.nyse.com/trade/hours-calendars

A query was made on September 7 at 12:32 UTC. It found three closed trades with IDs 9, 7, and 1, and no open, pending, or closing trades. All trades were exited on September 4 by the operator, not because of strategy stops or targets. The outcomes were calculated using the exact prices and quantities at which the trades were filled. The results are as follows: the trade for SNOW had a return of -5.392364480519829 percent and a profit and loss of -2.426024762767998, it lasted for 20.979384166666666 hours and was exited due to operator liquidation, the trade for DELL had a return of 7.939941307427634 percent and a profit and loss of 6.509957861609997, it lasted for 43.23018305555556 hours and was also exited due to operator liquidation, and the trade for GTLB had a return of -5.130499047026326 percent and a profit and loss of -5.129985995653997, it lasted for 47.238574166666666 hours and was exited due to operator liquidation. The overall statistics are: three trades were made, the percentage of winning trades was 33.33333333333333 percent, the average return for winning trades was 7.939941307427634 percent, the average return for losing trades was -5.261431763773078 percent, the mean return was -0.8609740733728405 percent, and the total profit and loss was -1.0460528968119984. The quarter Kelly criterion is 0, the capital is 212.7475, and the risk is 8.5099. The slippage is unknown because there are no quotes from when the trades arrived. The sample size is too small and was truncated by discretionary liquidation, so it is not possible to estimate the intended 4-6 week post-earnings announcement drift edge. The equity is 850.99 and the quantitative cash is 245.69, based on the supplied context, but these values have not been independently revalued. query_trades fund=quant; user account context 2026-09-07T12:30:57.599Z

A calculation was made on September 7. The results are shown in a table with the following columns: year, one-way basis point, return, drawdown, win rate, sample, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The table has the following rows: for 2021, the values are 10, -15.587730588, 15.587730588, 6.4220183486, 109, 0.16244076611, 0.17703782358, -0.15523644626, 0, -15.64470054, and 5.8207660913e-11; for 2022, the values are 10, -15.438469105, 15.438469105, 20, 115, 0.16200330565, 0.22231128119, -0.14544836382, 0, -15.551834533, and 5.4569682106e-11; for 2023, the values are 10, -13.384539497, 13.608402101, 16.666666667, 114, 0.1013994718, 0.17128125866, -0.12583447025, 0, -13.585047968, and 6.3664629124e-11; for 2024, the values are 10, -18.48637468, 18.690292245, 8.6206896552, 116, 0.080825857786, 0.20019838878, -0.17597216062, 0, -18.686867283, and 6.4574123826e-11; for 2025, the values are 10, -19.866405479, 19.866405479, 5.0847457627, 118, 0.10726840618, 0.2032452261, -0.18745639734, 0, -20.224864272, and 6.3664629124e-11. The worst close discrepancies are as follows: on March 20, 2020, at 00:00:00, the close discrepancy basis point was -28.30044362857609, the entry was 127.95, the exit was 130.37, and the official close was 130.74; on March 12, 2020, at 00:00:00, the close discrepancy basis point was 25.54970579126792, the entry was 128.24, the exit was 129.49, and the official close was 129.16; on March 18, 2020, at 00:00:00, the close discrepancy basis point was 18.585790318492723, the entry was 120.43, the exit was 118.59, and the official close was 118.37. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the auction execution was not authorized. The system used the official daily close instead of the final 30-minute bar close, and kept the entries and splits. This calculation was done on September 7. The results are shown in a table with columns including one-way break-even point, return, drawdown, win rate, sample size, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of the table show the following data: for a value of 0, the results are -0.082168067051 for return, 0.89344926448 for drawdown, 52.702702703 for win rate, 74 for sample size, 0.069169755846 for average win percentage, 0.081653582247 for average loss percentage, -0.0010621799784 for mean trade percentage, 0 for quarter Kelly, -0.12177793492 for annualized percentage, and 0 for cash reconcile absolute. For a value of 2, the results are -2.9963926047 for return, 3.0428867929 for drawdown, 33.783783784 for win rate, 74 for sample size, 0.053203242843 for average win percentage, 0.089144062123 for average loss percentage, -0.041053756391 for mean trade percentage, 0 for quarter Kelly, -4.4094450734 for annualized percentage, and 5.4569682106e-12 for cash reconcile absolute. For a value of 10, the results are -13.827753866 for return, 13.827753866 for drawdown, 1.3513513514 for win rate, 74 for sample size, 0.028136356549 for average win percentage, 0.20399714688 for average loss percentage, -0.20086020764 for mean trade percentage, 0 for quarter Kelly, -19.796800269 for annualized percentage, and 3.2741809264e-11 for cash reconcile absolute. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the auction execution was not authorized. The system uses the official daily close instead of the final 30-minute bar close and keeps entries and splits. This calculation was done on September 7. The results are shown in a table with columns including one-way break-even point, return, drawdown, win rate, sample size, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconciliation absolute value. The rows of the table show the following values: for a value of 0, the results are 0.1056966107 for return, 0.62181180173 for drawdown, 51.851851852 for win rate, 81 for sample size, 0.069054074456 for average win percentage, 0.075435142283 for average loss percentage, 0.0013477884283 for mean trade percentage, 0 for quarter Kelly, 0.15671961452 for annualized percentage, and 1.8189894035e-12 for cash reconciliation absolute value. For a value of 2, the results are -3.0857473946 for return, 3.0857473946 for drawdown, 34.567901235 for win rate, 81 for sample size, 0.050858466284 for average win percentage, 0.085929470756 for average loss percentage, -0.038644751779 for mean trade percentage, 0 for quarter Kelly, -4.5399419287 for annualized percentage, and 1.8189894035e-12 for cash reconciliation absolute value. For a value of 10, the results are -14.865991942 for return, 14.865991942 for drawdown, 1.2345679012 for win rate, 81 for sample size, 0.028136356549 for average win percentage, 0.20128744699 for average loss percentage, -0.19845505436 for mean trade percentage, 0 for quarter Kelly, -21.225057408 for annualized percentage, and 3.092281986e-11 for cash reconciliation absolute value. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the auction execution was not authorized. The system used the official daily close instead of the final 30-minute bar close, and kept the entries and splits. This calculation was done on September 7. The results are shown in a table with the following columns: one-way break-even point, return, drawdown, win rate, sample size, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of the table show the following values: for a value of 0, the results are -0.2077948202 for return, 0.85733775041 for drawdown, 50.588235294 for win rate, 85 for sample size, 0.066568557128 for average win percentage, 0.080701698434 for average loss percentage, -0.0024025480467 for mean trade percentage, 0 for quarter Kelly, -0.30787083995 for annualized percentage, and 3.6379788071e-12 for cash reconcile absolute. For a value of 2, the results are -3.5436980992 for return, 3.5899299618 for drawdown, 30.588235294 for win rate, 85 for sample size, 0.053226237724 for average win percentage, 0.084531138923 for average loss percentage, -0.042393588419 for mean trade percentage, 0 for quarter Kelly, -5.207837817 for annualized percentage, and 1.0913936421e-11 for cash reconcile absolute. For a value of 10, the results are -15.808828712 for return, 15.808828712 for drawdown, 1.1764705882 for win rate, 85 for sample size, 0.028136356549 for average win percentage, 0.20493997211 for average loss percentage, -0.20219789765 for mean trade percentage, 0 for quarter Kelly, -22.514818811 for annualized percentage, and 4.3655745685e-11 for cash reconcile absolute. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that an auction execution was not authorized. This uses the official daily close instead of the final 30-minute bar close, and keeps entries and splits, based on a September 7 calculation. The columns of data are: one-way break-even point, return, drawdown, win rate, sample, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of data are: [0, 22.461874694, 1.905473981, 60.665362035, 511, 0.13187035322, 0.11044952859, 0.039796863365, 0.069300460831, 4.1866876752, 5.4569682106e-12], [2, -0.17688658255, 6.4205356689, 49.706457926, 511, 0.11568506771, 0.1147543015, -0.00021105339703, 0, -0.035828832409, 7.0940586738e-11], [10, -55.92910138, 55.92910138, 11.154598826, 511, 0.12399410456, 0.1957488439, -0.16008280073, 0, -15.28254202, 1.2096279534e-10]. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the system was not authorized to execute an auction. Instead, it used the official daily close, rather than the final 30-minute bar close, and kept the entries and splits. This calculation was done on September 7. The results are shown in a table with columns including one-way break-even point, return, drawdown, win rate, sample size, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of the table show the following values: for a value of 0, the results are 31.112909786, 1.490415844, 61.652892562, 605, 0.13257150055, 0.1055741837, 0.044914077593, 0.077787305303, 5.6361523336, and 3.092281986e-11; for a value of 2, the results are 2.9309997011, 7.8690625493, 51.239669421, 605, 0.11446911418, 0.11023198716, 0.004904114355, 0.010710562386, 0.58644747416, and 7.0940586738e-11; for a value of 10, the results are -60.902490946, 60.902490946, 12.066115702, 605, 0.1162591856, 0.19219414666, -0.1549758107, 0, -17.311132388, and 1.3278622646e-10. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that an auction execution was not authorized. This uses the official daily close instead of the final 30-minute bar close, and keeps entries and splits, based on a September 7 calculation. The columns in the data are: one-way break-even point, return, drawdown, win rate, sample, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of data are: [0, 24.427545871, 2.0420279598, 60.314685315, 572, 0.13014474055, 0.10884430493, 0.038345781727, 0.067811395291, 4.5230373912, 1.0913936421e-11], [2, -1.0194283557, 6.5360048326, 49.825174825, 572, 0.11253150686, 0.11505884583, -0.0016615547187, 0, -0.20718615566, 7.0940586738e-11], [10, -60.364567025, 60.364567025, 11.188811189, 572, 0.11669597951, 0.19658319887, -0.1615309831, 0, -17.082108857, 1.4415491023e-10]. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the auction execution was not authorized. The system used the official daily close instead of the final 30-minute bar close and kept the entries and splits. This calculation was done on September 7. The results are shown in a table with the following columns: one-way break-even point, return, drawdown, win rate, sample size, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of the table show the following values: for a one-way break-even point of 0, the return was -0.39039400618, the drawdown was 8.4017872026, the win rate was 55, the sample size was 540, the average win percentage was 0.084508891321, the average loss percentage was 0.12080100192, the mean trade percentage was -0.00049827718534, the quarter Kelly was 0, the annualized percentage was -0.079206415981, and the cash reconcile absolute was 1.8189894035e-12. For a one-way break-even point of 2, the return was -19.741024103, the drawdown was 20.195316654, the win rate was 38.888888889, the sample size was 540, the average win percentage was 0.070021682316, the average loss percentage was 0.11081574548, the mean trade percentage was -0.040490079114, the quarter Kelly was 0, the annualized percentage was -4.3570314572, and the cash reconcile absolute was 6.3664629124e-11. For a one-way break-even point of 10, the return was -66.173026538, the drawdown was 66.173026538, the win rate was 4.8148148148, the sample size was 540, the average win percentage was 0.064738183397, the average loss percentage was 0.21370390232, the mean trade percentage was -0.20029743153, the quarter Kelly was 0, the annualized percentage was -19.71364713, and the cash reconcile absolute was 1.0641088011e-10. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the system was not authorized to execute an auction. Instead, it used the official daily close, rather than the final 30-minute bar close, and kept the entries and splits. This calculation was done on September 7. The results are shown in a table with the following columns: one-way break-even point, return, drawdown, win rate, sample size, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of the table show the following values: for a value of 0, the results are 10.622551687, 2.890333896, 55.386949924, 659, 0.092158012067, 0.092365342361, 0.015442909628, 0.026683831751, 2.0660988425, and 2.9103830457e-11. For a value of 2, the results are -15.010818616, 17.097397518, 39.908952959, 659, 0.079373984326, 0.09357898778, -0.024555267501, 0, -3.2410764865, and 2.7284841053e-11. For a value of 10, the results are -70.389652984, 70.389652984, 5.31107739, 659, 0.11964228178, 0.20144108112, -0.18438809523, 0, -21.849984491, and 9.8680175142e-11. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the auction execution was not authorized. The system used the official daily close instead of the final 30-minute bar close, and kept the entries and splits. This calculation was done on September 7. The results are shown in a table with the following columns: one-way break-even point, return, drawdown, win rate, sample size, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of the table are: row 0 with values 0, 4.6854687902, 8.154481388, 54.853620955, 649, 0.095458014426, 0.11651381411, 0.0073005944419, 0, 0.93189684677, 9.0949470177e-12, row 2 with values 2, -19.249809732, 20.672214001, 39.599383667, 649, 0.08320489525, 0.10867927531, -0.032694326412, 0, -4.2387399831, 8.5492501967e-11, and row 10 with values 10, -71.412767243, 71.412767243, 5.3929121726, 649, 0.13932801527, 0.21143022594, -0.19251414206, 0, -22.404683414, 8.6856744019e-11. https://runway.adityaperswal.workers.dev/data/stock-bars

A calculation was done on September 7. The results are shown in a table with the following columns: year, one-way basis point, return, drawdown, win rate, sample, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The table has the following rows: for 2021, the values are 10, -23.606719239, 23.606719239, 5.2631578947, 133, 0.0828400057, 0.21800036801, -0.20216666413, 0, -23.688630809, and 4.0017766878e-11. For 2022, the values are 10, -19.360201088, 19.360201088, 14.754098361, 122, 0.18153912769, 0.23788388593, -0.17600180195, 0, -19.498896882, and 4.7293724492e-11. For 2023, the values are 10, -22.178283347, 22.178283347, 1.6393442623, 122, 0.19634482969, 0.21190574837, -0.20521311594, 0, -22.492385957, and 4.8203219194e-11. For 2024, the values are 10, -20.61581621, 20.61581621, 2.3076923077, 130, 0.12650260431, 0.18456337417, -0.17738492852, 0, -20.836322746, and 1.8189894035e-11. For 2025, the values are 10, -22.288030268, 22.288030268, 3.7878787879, 132, 0.15707383689, 0.20446454706, -0.19076991131, 0, -22.683698102, and 6.2755134422e-11. The worst close discrepancies are as follows: on March 16, 2020, at 00:00:00, the close discrepancy basis point was 211.7093570308426, the entry was 39.19, the exit was 39.07, and the official close was 38.26. On March 9, 2022, at 00:00:00, the close discrepancy basis point was 24.521824423737737, the entry was 61.37, the exit was 61.32, and the official close was 61.17. On March 31, 2020, at 00:00:00, the close discrepancy basis point was 15.909090909091985, the entry was 43.51, the exit was 44.07, and the official close was 44. https://runway.adityaperswal.workers.dev/data/stock-bars

A post-result endpoint audit found that the auction execution was not authorized. The system used the official daily close instead of the final 30-minute bar close, and kept the entries and splits. This calculation was done on September 7. The results are shown in a table with the following columns: one-way basis point, return, drawdown, win rate, sample, average win percentage, average loss percentage, mean trade percentage, quarter Kelly, annualized percentage, and cash reconcile absolute. The rows of the table are: row 0 with values 0, 3.0090134143, 0.63497818799, 53.947368421, 76, 0.12487566197, 0.067113885311, 0.039108655403, 0.072991451085, 4.4926231235, 1.2732925825e-11, row 2 with values 2, -0.075340885657, 0.97094349482, 42.105263158, 76, 0.11339624608, 0.084022791376, -0.00089898613074, 0, -0.11166148772, 1.2732925825e-11, and row 10 with values 10, -11.516465787, 11.516465787, 9.2105263158, 76, 0.16594517906, 0.19391461465, -0.16076963365, 0, -16.587447222, 2.546585165e-11. https://runway.adityaperswal.workers.dev/data/stock-bars