QuantProp. Analyze a trade file ↗
Home / Research archive / Stockback research archive
stockback / Experiment result

EXP-009 — Connors RSI(2) 5-minute cost and slippage confirmation experiment

Full EXP-009 results: locked OOS NAVs and returns for 2 symbols × 7 variants under 0–5bp slippage, with charts.

Original record: 2026-09-04 · Migrated: 2026-10-27 · Translated from Korean: 2026-10-27
A historical record, not rerun in this migration.

The figures, hypotheses and verdicts of the time are preserved. A past positive result or candidate mention does not mean validated profitability today or a live-trading recommendation. Validation periods that were already used are not treated as new independent validation.

This is an English translation of a preserved Korean original. It describes conditions at the time of writing, not current provider specifications, costs or regulations.

EXP-009 — Connors RSI(2) 5-minute cost and slippage confirmation experiment

  • Run date: 2026-09-04
  • Review model: gpt-5.6-luna / high

| generated_at | 2026-09-04T00:00:00+09:00 |

  • Status: confirmation experiment complete · not a strategy promotion
  • Purpose: separate the effects of per-product costs and slippage on the 5-minute results of EXP-006/007.
  • Capital: initial 3,000,000 KRW per symbol and variant, reinvesting 95% of cash just before entry, integer quantities, no leverage

Data audit

SymbolInputBarsSessionsStartEndSHA-256
A005930[original local path withheld]96,1421,2502021-07-16T09:05:00+09:002026-08-26T15:30:00+09:00671f67506d52a6fd77da5b8a5ce7506bb81b83bcb08407c192eaf32966c62fbe
A069500[original local path withheld]96,1341,2502021-07-16T09:05:00+09:002026-08-26T15:30:00+09:001c910ee52d9e19a3113833ff33cc1e8d93de7ed650f732cfc74108cfe3e73032

The 5-minute input is the same CREON source as the existing EXP-006/007. Daily results were not included in this experiment.

Cost contract

ProfileOne-way commissionOne-way slippagePer-product taxesUse
zero0.0bp0.0bpNone0bp cost diagnostic line
actual_proxy1.5bp0.0bpA005930 KOSPI scheduleOne-way commission 1.5bp + per-product taxes
actual_proxy_slip11.5bp1.0bpA005930 KOSPI scheduleProvisional actual cost + 1bp one-way slippage
actual_proxy_slip21.5bp2.0bpA005930 KOSPI scheduleProvisional actual cost + 2bp one-way slippage
actual_proxy_slip51.5bp5.0bpA005930 KOSPI scheduleProvisional actual cost + 5bp one-way slippage

The 1.5bp/one-way commission in actual_proxy is a provisional value before checking actual CREON account fills. Slippage of 1/2/5bp is a sensitivity range, not a forecast, and the best value of any one range is not adopted.

Locked OOS results

Locked OOS is 2025-01-01~2026-08-26; each run ran continuously from 2021-07-16, and the period return was computed from the NAV just before the OOS start.

SymbolVariantCostOOS tradesOOS starting NAVOOS ending NAVOOS returnFull MDDOOS win ratePF
A005930A0-baselineactual_proxy416751,652 KRW422,687 KRW-43.77%-85.95%43.3%0.194
A005930A0-baselineactual_proxy_slip1416663,228 KRW347,967 KRW-47.53%-88.44%38.9%0.163
A005930A0-baselineactual_proxy_slip2402586,451 KRW327,927 KRW-44.08%-89.11%36.1%0.138
A005930A0-baselineactual_proxy_slip5314408,328 KRW214,561 KRW-47.45%-92.88%23.6%0.077
A005930A1-stop7actual_proxy416751,652 KRW420,876 KRW-44.01%-86.01%43.3%0.194
A005930A1-stop7actual_proxy_slip1416663,228 KRW346,156 KRW-47.81%-88.50%38.9%0.163
A005930A1-stop7actual_proxy_slip2402586,451 KRW327,927 KRW-44.08%-89.11%36.1%0.138
A005930A1-stop7actual_proxy_slip5314408,328 KRW214,561 KRW-47.45%-92.88%23.6%0.077
A005930A2-time5barsactual_proxy455565,602 KRW288,416 KRW-49.01%-90.43%37.8%0.146
A005930A2-time5barsactual_proxy_slip1397488,712 KRW248,139 KRW-49.23%-91.73%29.0%0.113
A005930A2-time5barsactual_proxy_slip2373423,593 KRW223,124 KRW-47.33%-92.61%27.1%0.095
A005930A2-time5barsactual_proxy_slip5296282,369 KRW156,798 KRW-44.47%-94.79%17.9%0.062
A005930A3-stop7-time5barsactual_proxy455565,602 KRW288,416 KRW-49.01%-90.43%37.8%0.146
A005930A3-stop7-time5barsactual_proxy_slip1397488,712 KRW248,139 KRW-49.23%-91.73%29.0%0.113
A005930A3-stop7-time5barsactual_proxy_slip2373423,593 KRW223,124 KRW-47.33%-92.61%27.1%0.095
A005930A3-stop7-time5barsactual_proxy_slip5296282,369 KRW156,798 KRW-44.47%-94.79%17.9%0.062
A005930B0-daily-regimeactual_proxy631546,700 KRW291,625 KRW-46.66%-90.46%41.0%0.229
A005930B0-daily-regimeactual_proxy_slip1559470,009 KRW262,049 KRW-44.25%-91.48%36.5%0.187
A005930B0-daily-regimeactual_proxy_slip2479402,150 KRW236,611 KRW-41.16%-92.49%32.2%0.151
A005930B0-daily-regimeactual_proxy_slip5311270,934 KRW147,798 KRW-45.45%-95.08%21.2%0.073
A005930B1-daily-regime-day-onlyactual_proxy579527,387 KRW254,933 KRW-51.66%-91.61%37.5%0.187
A005930B1-daily-regime-day-onlyactual_proxy_slip1517453,757 KRW242,410 KRW-46.58%-92.14%32.7%0.152
A005930B1-daily-regime-day-onlyactual_proxy_slip2454390,791 KRW222,152 KRW-43.15%-93.03%31.1%0.126
A005930B1-daily-regime-day-onlyactual_proxy_slip5312250,257 KRW144,927 KRW-42.09%-95.19%19.2%0.062
A005930B2-daily-regime-five-sessionsactual_proxy631546,700 KRW291,625 KRW-46.66%-90.46%41.0%0.229
A005930B2-daily-regime-five-sessionsactual_proxy_slip1559470,009 KRW262,049 KRW-44.25%-91.48%36.5%0.187
A005930B2-daily-regime-five-sessionsactual_proxy_slip2479402,150 KRW236,611 KRW-41.16%-92.49%32.2%0.151
A005930B2-daily-regime-five-sessionsactual_proxy_slip5311270,934 KRW147,798 KRW-45.45%-95.08%21.2%0.073
A069500A0-baselineactual_proxy4842,307,764 KRW2,047,244 KRW-11.29%-33.40%57.4%0.744
A069500A0-baselineactual_proxy_slip14842,004,572 KRW1,628,099 KRW-18.78%-46.25%53.3%0.608
A069500A0-baselineactual_proxy_slip24841,740,692 KRW1,287,780 KRW-26.02%-57.42%47.7%0.488
A069500A0-baselineactual_proxy_slip54841,143,828 KRW659,757 KRW-42.32%-78.16%36.8%0.244
A069500A1-stop7actual_proxy4842,307,764 KRW2,047,244 KRW-11.29%-33.40%57.4%0.744
A069500A1-stop7actual_proxy_slip14842,004,572 KRW1,628,099 KRW-18.78%-46.25%53.3%0.608
A069500A1-stop7actual_proxy_slip24841,740,692 KRW1,287,780 KRW-26.02%-57.42%47.7%0.488
A069500A1-stop7actual_proxy_slip54841,143,828 KRW659,757 KRW-42.32%-78.16%36.8%0.244
A069500A2-time5barsactual_proxy5622,263,273 KRW1,912,272 KRW-15.51%-41.24%52.1%0.700
A069500A2-time5barsactual_proxy_slip15621,927,169 KRW1,476,296 KRW-23.40%-54.03%48.9%0.554
A069500A2-time5barsactual_proxy_slip25621,643,791 KRW1,131,125 KRW-31.19%-64.12%43.8%0.429
A069500A2-time5barsactual_proxy_slip55621,017,951 KRW511,125 KRW-49.79%-83.02%32.4%0.188
A069500A3-stop7-time5barsactual_proxy5622,263,273 KRW1,912,272 KRW-15.51%-41.24%52.1%0.700
A069500A3-stop7-time5barsactual_proxy_slip15621,927,169 KRW1,476,296 KRW-23.40%-54.03%48.9%0.554
A069500A3-stop7-time5barsactual_proxy_slip25621,643,791 KRW1,131,125 KRW-31.19%-64.12%43.8%0.429
A069500A3-stop7-time5barsactual_proxy_slip55621,017,951 KRW511,125 KRW-49.79%-83.02%32.4%0.188
A069500B0-daily-regimeactual_proxy7032,132,171 KRW1,936,660 KRW-9.17%-42.09%57.6%0.803
A069500B0-daily-regimeactual_proxy_slip17031,831,532 KRW1,454,134 KRW-20.61%-53.77%55.2%0.679
A069500B0-daily-regimeactual_proxy_slip27031,570,319 KRW1,109,519 KRW-29.34%-63.71%50.8%0.566
A069500B0-daily-regimeactual_proxy_slip5703994,326 KRW482,434 KRW-51.48%-84.17%41.8%0.299
A069500B1-daily-regime-day-onlyactual_proxy7262,080,151 KRW1,652,609 KRW-20.55%-46.84%56.3%0.720
A069500B1-daily-regime-day-onlyactual_proxy_slip17261,776,752 KRW1,237,960 KRW-30.32%-59.68%53.6%0.593
A069500B1-daily-regime-day-onlyactual_proxy_slip27261,516,250 KRW923,685 KRW-39.08%-69.78%49.3%0.476
A069500B1-daily-regime-day-onlyactual_proxy_slip5726942,131 KRW401,596 KRW-57.37%-86.80%39.5%0.226
A069500B2-daily-regime-five-sessionsactual_proxy7032,132,171 KRW1,936,660 KRW-9.17%-42.09%57.6%0.803
A069500B2-daily-regime-five-sessionsactual_proxy_slip17031,831,532 KRW1,454,134 KRW-20.61%-53.77%55.2%0.679
A069500B2-daily-regime-five-sessionsactual_proxy_slip27031,570,319 KRW1,109,519 KRW-29.34%-63.71%50.8%0.566
A069500B2-daily-regime-five-sessionsactual_proxy_slip5703994,326 KRW482,434 KRW-51.48%-84.17%41.8%0.299

Review insights

  1. Under actual_proxy, 0 of 14 cells are positive, and the mean OOS return is -30.23%.
  2. At 2bp one-way slippage, 0 of 14 cells are positive, and the mean OOS return is -37.18%.
  3. 5-minute bars are not simply a scaled-down daily RSI(2) strategy. On 5-minute bars, RSI(2), SMA(200) and the 5-bar exit carry the economic meaning of about 10 minutes, 2.5 trading days and 25 minutes respectively.
  4. Because tracks A/B and costs/slippage were separated, it can be checked whether a result survives only under one particular execution assumption.
  5. Two symbols and about 5 years of intraday bars alone do not promote a live-trading candidate. Positive OOS cells are only subjects for the next stage's review of year, time-of-day and trade concentration.

Charts

Locked OOS cost and slippage comparison for the main variants

-43.8%A005930 A0-baseline actual_proxy-47.5%A005930 A0-baseline actual_proxy_slip1-44.1%A005930 A0-baseline actual_proxy_slip2-47.5%A005930 A0-baseline actual_proxy_slip5-46.7%A005930 B0-daily-regime actual_proxy-44.2%A005930 B0-daily-regime actual_proxy_slip1-41.2%A005930 B0-daily-regime actual_proxy_slip2-45.4%A005930 B0-daily-regime actual_proxy_slip5-11.3%A069500 A0-baseline actual_proxy-18.8%A069500 A0-baseline actual_proxy_slip1-26.0%A069500 A0-baseline actual_proxy_slip2-42.3%A069500 A0-baseline actual_proxy_slip5-9.2%A069500 B0-daily-regime actual_proxy-20.6%A069500 B0-daily-regime actual_proxy_slip1-29.3%A069500 B0-daily-regime actual_proxy_slip2-51.5%A069500 B0-daily-regime actual_proxy_slip5

Baseline-cost equity curves by product

A005930 A0-baselineA005930 B0-daily-regimeA069500 A0-baselineA069500 B0-daily-regime

Reproducibility

  • Run script: scripts/backtest/run_connors_rsi2_m5_confirmation.py
  • Summary ledger: reports/ledgers/EXP-009-summary.csv
  • Trade ledgers: reports/ledgers/EXP-009-m5-*-trades.csv
  • Existing EXP-005/006/007 results were not overwritten.
  • Original 5-minute CSVs and private material are not included in Git.

Source and migration evidence

Source: reports/experiments/EXP-009-connors-rsi2-m5-confirmation.md

Original SHA-256: a2713ca6718228c6203f21eb01853fb59c0229ffc5125f345cca17b9b15b425b

Raw data and account information are not part of the public package. The original results do not replace statistical or scientific revalidation. The English text is a translation; numbers, tables and charts are carried over unchanged.

Linked records

Evidence and follow-up records

archive / hub

Stockback research archive

64 historical Korean equity research records, translated into English. Read the experiment design, actual report and verdict together.

READ THE EVIDENCE
archive / hub

The record behind the research.

Stockback and Coinback research records in English, translated from the Korean originals and preserved with source hashes and explicit historical context.

READ THE EVIDENCE
methodology / hub

Evidence has a chain of custody.

Every result should tell you what was known, what was assumed, and what remains untested.

READ THE EVIDENCE
policy / policy

The assumptions belong beside the result.

Research, education and tools are not personalized investment advice.

READ THE EVIDENCE
archive / article

EXP-008 — Turtle + Bollinger squeeze daily validation

Daily backtest of a Turtle 55-day breakout with a Bollinger squeeze filter on 7 Korean stocks: 6,720 result rows, costs, slippage and a hold verdict.

READ THE EVIDENCE
archive / article

EXP-010 — Connors RSI(2) daily, generalization across symbols

EXP-010 results: data audit, cost contract, locked OOS baseline, all four exit variants and equity charts.

READ THE EVIDENCE
archive / article

EXP-009 verdict — Connors RSI(2) 5-minute cost and slippage confirmation

EXP-009 verdict: 0 of 14 out-of-sample cells were positive at provisional costs or 2bp slippage, so the 5-minute RSI(2) contract was held.

READ THE EVIDENCE
archive / article

2026-09-04 — 5-minute Connors RSI(2) confirmation experiment EXP-009

70 combinations of 2 symbols, 7 variants and 5 cost profiles: no positive OOS cell even before slippage on 5-minute bars.

READ THE EVIDENCE
archive / article

EXP-009 design — Connors RSI(2) 5-minute cost and slippage confirmation

Preregistered EXP-009 contract: 2021–2026 5-minute data, 7 variants, a cost sensitivity grid and rules against picking a best cell.

READ THE EVIDENCE