Herman Trading (@RHerman)

@RHermanOct 21

London session. 62% probability for the short side. I took shorts only. +2R on the day. This is the full model. 1. Intelligent Balance Probability Map at 2 AM NY: 62% short, 38% long 2. Bias set: shorts and shorts only 3. Wait for a sweep reversal with confirmation 4. Short #1: too early, too aggressive. -1R 5. Short #2: small but valid confirmed sweep reversal. Entry 6. TP toward the more probable side. Level reached: +3R Basic ICT idea, done with numbers: trade toward the more probable side, and only with confirmation. Full walkthrough in the video. After the first short stopped out, would you have taken the second one?

Herman Trading educational graphic showing a 62% short directional bias on NQ during the London session. An original TradingView replay screenshot accompanies the message that probability selects the trading side while confirmation determines the entry.

ALT Herman Trading educational graphic showing a 62% short directional bias on NQ during the London session. An original TradingView replay screenshot accompanies the message that probability selects the trading side while confirmation determines the entry.

6,997

Herman Trading · Oct 30, 2026 · 9:00 PM UTC

@RHerman16h

𝗖𝗹𝗮𝘂𝗱𝗲, 𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗲𝗱 𝘁𝗼 𝗧𝗿𝗮𝗱𝗶𝗻𝗴𝗩𝗶𝗲𝘄. 𝟲 𝗽𝗿𝗼𝗺𝗽𝘁𝘀 𝗜'𝗱 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘂𝘀𝗲. 𝗡𝗼𝘁 𝗼𝗻𝗲 𝗼𝗳 𝘁𝗵𝗲𝗺 𝗮𝘀𝗸𝘀 "𝗜𝘀 𝗡𝗤 𝗯𝘂𝗹𝗹𝗶𝘀𝗵?" 1. Your stop was the liquidity Last 15 NQ sessions on the 5m: every sweep of the previous day's high or low by less than 15 points that closed back inside within 3 candles. What time, and how far did it run after? 2. NQ can lie. Its top 7 can't NQ just made a new high. How many of AAPL, MSFT, NVDA, AMZN, META, GOOGL and AVGO confirmed it on the 5m? Fewer than 4 and the high is suspicious. 3. Hold my leash NQ is ripping and I want to chase. Claude sets an alert that tells me to wait and look for my entry. 4. Who actually moved it? NQ dropped 150 points in 10 minutes at 14:00. News, calendar or one of the biggest Nasdaq-100 weights? Timestamps and sources. 5. The edge check in 30 seconds Last 15 sessions: how often did New York take out the London high or low before 10:30 ET, which side went first, and what happened after? 6. Good news is bad news, is it? Last 12 months of US CPI and unemployment data: did NQ go up or down on each release day? Ask a hypothesis. Pull the data. Test it. Measure it. TradingView stops being just a chart. What's the first question you'd ask Claude about your own chart?

922

Herman Trading · Oct 30, 2026 · 9:00 PM UTC

@RHerman16h

Observation → data → statistics → next hypothesis. All 6 prompts on one card.

Herman Trading graphic, dark, 16:9: "Claude + TradingView", "6 real trading prompts", "AI-assisted market research". On the left a glowing head outline with the lines Analyze, Find patterns, Challenge assumptions, Improve your edge, and a prompt bar "Ask anything about the markets..."; in the middle "AI + markets = insights"; on the right a candlestick chart with an "Alert, potential setup detected" badge. Six numbered tiles along the bottom: 1 Analyze markets, 2 Find setups, 3 Manage risk, 4 Backtest ideas, 5 Interpret data, 6 Improve decisions. "Herman Trading, @rherman" at the bottom. Using Claude with TradingView data to test NQ trading ideas.

ALT Herman Trading graphic, dark, 16:9: "Claude + TradingView", "6 real trading prompts", "AI-assisted market research". On the left a glowing head outline with the lines Analyze, Find patterns, Challenge assumptions, Improve your edge, and a prompt bar "Ask anything about the markets..."; in the middle "AI + markets = insights"; on the right a candlestick chart with an "Alert, potential setup detected" badge. Six numbered tiles along the bottom: 1 Analyze markets, 2 Find setups, 3 Manage risk, 4 Backtest ideas, 5 Interpret data, 6 Improve decisions. "Herman Trading, @rherman" at the bottom. Using Claude with TradingView data to test NQ trading ideas.

220

Herman Trading · Oct 30, 2026 · 6:30 PM UTC

@RHerman18h

I tested 4,320 opening range setups on NQ. Retest entries lost in both periods.

Article · @RHerman

744

Herman Trading · Oct 30, 2026 · 4:00 PM UTC

@RHerman21h

I tested 246 trade versions of my NQ session plays. 30 made PF 1.2+ in both periods. How the filter works: 1. Find the plays on 2015-2021 only 2. Check them again on 2022-2026, data they never saw 3. Turn each play into simple trades: 1 contract, $4.50 round trip, 1 tick slippage The result: → 246 trade versions tested → 145 made money in both periods → 30 had PF 1.2+ in both periods and 8+ profitable years The luck check: the same search on shuffled filters found 1.2 plays at 08:00 NY on average. On real data: 69. How many of your backtested setups survive data they never saw? Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026

1,327

Herman Trading · Oct 30, 2026 · 4:00 PM UTC

@RHerman21h

Shuffled filters: 1.2 plays on average. Real data, same search and same gates: 69.

White editorial with orange accents, Herman Trading graphic in 4:5. Label: How the filter works. Headline: 30 of 246 survived. Two cards: 246 trade versions tested; 145 profitable in 2015-21 and 2022-26. A large 30: PF 1.2+ in both periods, and 8+ profitable years. Luck check: plays found at 08:00 NY: 69 on real data; 1.2 on shuffled filters, average. Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026. Herman Trading backtesting method on NQ futures: plays found on 2015-2021, checked out of sample on 2022-2026 and compared with shuffled filters.

ALT White editorial with orange accents, Herman Trading graphic in 4:5. Label: How the filter works. Headline: 30 of 246 survived. Two cards: 246 trade versions tested; 145 profitable in 2015-21 and 2022-26. A large 30: PF 1.2+ in both periods, and 8+ profitable years. Luck check: plays found at 08:00 NY: 69 on real data; 1.2 on shuffled filters, average. Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026. Herman Trading backtesting method on NQ futures: plays found on 2015-2021, checked out of sample on 2022-2026 and compared with shuffled filters.

161

Herman Trading · Oct 30, 2026 · 12:00 PM UTC

@RHermanOct 30

Hey @grok Why do 90% of traders fail prop firm challenges?

835

Herman Trading · Oct 29, 2026 · 9:00 PM UTC

𝟵𝟴 𝘄𝗶𝗻𝗻𝗲𝗿𝘀 𝗼𝘂𝘁 𝗼𝗳 𝟭𝟮𝟭 𝘁𝗿𝗮𝗱𝗲𝘀. 𝗣𝗿𝗼𝗳𝗶𝘁 𝗳𝗮𝗰𝘁𝗼𝗿 𝟮.𝟳𝟯. 𝗙𝗿𝗼𝗺 𝟮𝟬𝟮𝟬 𝘁𝗼 𝟮𝟬𝟮𝟲 𝗼𝗻 𝗠𝗡𝗤, 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲 𝗹𝗼𝗴𝗶𝗰 𝗻𝗲𝘃𝗲𝗿 𝗴𝗼𝘁 𝗯𝗮𝗰𝗸 𝗮𝗯𝗼𝘃𝗲 𝘇𝗲𝗿𝗼. Trend Rebalance Map [Herman] on TradingView, 16 Aug to 9 Sep 2026: 80.99% win rate Profit factor 2.73 +$4,541 No repainting, no lookahead, clean mechanical rules. It looked like a monster. Then I ran it in NinjaTrader's Strategy Analyzer from January 2020 to September 2026. The cumulative profit sits below zero the whole way. The strong weeks are a small slice at the very end. A win rate is not an edge. Robustness across years and market regimes is. The full Pine Script is public, so you can run the long test yourself. How many years of data do you need before you trust a win rate? github.com/HermanTrading/Tre…

GitHub - HermanTrading/Trend-Rebalance-Map-Herman-: Trend Rebalance Map [Herman]

Trend Rebalance Map [Herman]. Contribute to HermanTrading/Trend-Rebalance-Map-Herman- development by creating an account on GitHub.

github.com

199

17,974

Herman Trading · Oct 29, 2026 · 6:30 PM UTC

Venice. Paris. Now Normandy. The trading life I actually wanted. A family trip to France, and Mont Saint-Michel ticked off. For me it stands right next to Venice in Italy and the Eiffel Tower in Paris. One more place on my Europe map. Trading. Travel. Freedom. Traders, which place is still unticked on your map?

933

Herman Trading · Oct 29, 2026 · 6:30 PM UTC

Inside: Saint Michael, the archangel the Mont is named after.

ALT Travel postcard style graphic in 4:5: cream paper with a red, white and blue airmail border, the Herman Trading wordmark and a PAR AVION stamp at the top. The red label "Inside Mont Saint-Michel" sits above the headline "Saint Michael. The Mont's namesake.", the second line in red italic script. Below, a tilted postcard photo of the silver statue of the archangel Saint Michael with wings and a crown, sword raised, standing on a dragon, in front of a carved wooden altarpiece under a white stone arch, with candles and an altar below. A handwritten paper note in the corner reads "Trading. Travel. Freedom." Footer: www.hermantrading.pro.

235

Herman Trading · Oct 29, 2026 · 4:00 PM UTC

Gold, 02:00 New York, one Asia-session filter. London took the Asia high first on 92.2% of those days. The setup, known at 02:00 New York time: → One Asia-session filter, known before London opens 2022-2026, days London took an Asia level: the Asia high went first on 92.2% of 103 days. Distance alone predicts 75.1%. 2015-2026: 90.0% of 230 days. London took only the Asia high on 91.3% of them. Most common window: 02:00-02:30. Median time 02:55. About 1.6 days a month. Do you trade gold in London, or wait for New York? Source: Herman Trading Backtesting Centre, GC Session Study v2 (gc-session-study-v2), 2015-2026

1,472

Herman Trading · Oct 29, 2026 · 4:00 PM UTC

Above what distance alone predicts in every year since 2015. 2023 was 100%.

White editorial with orange accents, Herman Trading graphic in 4:5. Label: Gold, every year since 2015. Headline: Above expected 12 of 12 years. A large 92.2%: Asia high taken first, 103 days, 2022-2026. Chips: Expected from distance 75.1%, 2015-2026: 90.0% of 230 days, About 1.6 days a month. A bar chart per year with a line for the expected rate: 2015 83, 2016 97, 2017 87, 2018 92, 2019 88, 2020 79, 2021 93, 2022 75, 2023 100, 2024 93, 2025 94, 2026 90 percent. Source: Herman Trading Backtesting Centre, GC Session Study v2 (gc-session-study-v2), 2015-2026, days London took an Asia level, with an Asia-session filter at 02:00 NY. A gold futures (GC) session study from Herman Trading backtesting: which side of the Asia range London takes first.

ALT White editorial with orange accents, Herman Trading graphic in 4:5. Label: Gold, every year since 2015. Headline: Above expected 12 of 12 years. A large 92.2%: Asia high taken first, 103 days, 2022-2026. Chips: Expected from distance 75.1%, 2015-2026: 90.0% of 230 days, About 1.6 days a month. A bar chart per year with a line for the expected rate: 2015 83, 2016 97, 2017 87, 2018 92, 2019 88, 2020 79, 2021 93, 2022 75, 2023 100, 2024 93, 2025 94, 2026 90 percent. Source: Herman Trading Backtesting Centre, GC Session Study v2 (gc-session-study-v2), 2015-2026, days London took an Asia level, with an Asia-session filter at 02:00 NY. A gold futures (GC) session study from Herman Trading backtesting: which side of the Asia range London takes first.

275

Herman Trading · Oct 29, 2026 · 12:00 PM UTC

Your plan needs 3 things to go right. Each one is 80% likely. All three happen together 51% of the time. Every "and" in a trade plan costs you probability. How many conditions does your entry need?

974

Herman Trading · Oct 28, 2026 · 9:00 PM UTC

𝗧𝘄𝗼 𝗰𝗮𝗻𝗱𝗹𝗲𝘀 𝗯𝗲𝗳𝗼𝗿𝗲 𝘁𝗵𝗲 𝗡𝗤 𝗼𝗽𝗲𝗻. 𝗣𝗗𝗛 𝗴𝗼𝘁 𝘀𝘄𝗲𝗽𝘁 𝗶𝗻 𝟴𝟭% 𝗼𝗳 𝗼𝗻𝗲 𝗿𝗲𝗴𝗶𝗺𝗲 𝗮𝗻𝗱 𝟳.𝟬% 𝗼𝗳 𝗮𝗻𝗼𝘁𝗵𝗲𝗿. Compare yesterday's candle with the day before, on each side separately: CONT: it broke the prior extreme and closed beyond it REV: it broke the prior extreme but failed to hold it into the close None: it never reached the prior extreme High side and low side together give the regime. Chance of a Previous Day High sweep: CONT / REV → 81% CONT / None → 79.4% None / REV → 68.8% None / None → 63.4% REV / None → 43.8% None / CONT → 25.2% REV / CONT → 7.0% The raw two-candle pattern was 70% accurate. Candle filters took it to 77%, with each day predicted by a model built only from the other sessions. NQ futures study, 2015-2025. Classify the regime first, then look for an entry. Which regime did NQ print yesterday?

1,493

Herman Trading · Oct 28, 2026 · 9:00 PM UTC

The gap between the strongest and the weakest regime: 74%.

Herman Trading study card, light background. Big red "77%", "Accuracy after candle filters", and the title "Two-candle classification separated NQ into sharply different PDH sweep regimes." Key figures: 8 pattern states, 74% regime spread, 70% raw pattern accuracy. Horizontal bars for the chance of a Previous Day High sweep: CONT / REV 81%, CONT / NONE 79.4%, NONE / REV 68.8%, NONE / NONE 63.4%, REV / NONE 43.8%, NONE / CONT 25.2%, REV / CONT 7.0%. Note: the same two-candle structure can imply a high-probability expansion regime or a weak sweep regime. Legend: CONT = continuation, REV = reversal, NONE = no prior-extreme test. NQ futures study 2015-2025.

ALT Herman Trading study card, light background. Big red "77%", "Accuracy after candle filters", and the title "Two-candle classification separated NQ into sharply different PDH sweep regimes." Key figures: 8 pattern states, 74% regime spread, 70% raw pattern accuracy. Horizontal bars for the chance of a Previous Day High sweep: CONT / REV 81%, CONT / NONE 79.4%, NONE / REV 68.8%, NONE / NONE 63.4%, REV / NONE 43.8%, NONE / CONT 25.2%, REV / CONT 7.0%. Note: the same two-candle structure can imply a high-probability expansion regime or a weak sweep regime. Legend: CONT = continuation, REV = reversal, NONE = no prior-extreme test. NQ futures study 2015-2025.

274

Herman Trading · Oct 28, 2026 · 6:30 PM UTC

Plain ORB loses money on ES. Longs only plus 3 filters: PF 2.35, 12 of 12 years.

Article · @RHerman

1,145

Herman Trading · Oct 28, 2026 · 4:00 PM UTC

426 narrow 8:00 ET candles on NQ. The 9:00 hour missed both the high and the low only once. After a narrow 8:00 ET candle, the 9:00 candle took both its high and its low 59.4% of the time. Split by how wide the 8:00 candle was: → Narrow: 59.4% took both sides → Normal: 47.0% → Wide: 20.1% Both halves of the test agree: 57.9% in 2016-2021, 63.2% in 2022-2026. Narrow means under 0.7x the typical range for that hour. The 9:00 candle holds the 9:30 cash open, and a tight 8:00 range leaves both its levels close to price. Before 9:30, what do you mark first: 1 the 8:00 high, 2 the 8:00 low, or 3 both? Source: Herman Trading backtest, NQ 1H candles, 2016-2026

1,898

Herman Trading · Oct 28, 2026 · 4:00 PM UTC

The same setup as a still: 1 the narrow 8:00 candle, 2 its high taken, 3 its low taken.

Dark EVIDENCE-style Herman Trading graphic in 4:5. Header: NQ1! 1H, concept sketch, not real prices, and in amber: 1 in 426 missed both. A drawn candlestick sketch rises from left to right. Near the right edge a small narrow candle sets dashed amber lines labelled 8:00 high and 8:00 low. The next candle is wide and trades through both lines inside an amber box. Three numbered amber markers point at the chart: 1 narrow 8:00, 2 high taken, 3 low taken. The caption bar reads: 9:00 missed both sides, 1 in 426 in amber. What do you mark first? Three options: 1 8:00 high, 2 8:00 low, 3 Both, with Reply 1, 2 or 3. Herman Trading and hermantrading.pro in the footer. It illustrates an NQ futures backtest: after a narrow 8:00 ET hourly candle, the 9:00 hour almost always takes its high, its low or both.

ALT Dark EVIDENCE-style Herman Trading graphic in 4:5. Header: NQ1! 1H, concept sketch, not real prices, and in amber: 1 in 426 missed both. A drawn candlestick sketch rises from left to right. Near the right edge a small narrow candle sets dashed amber lines labelled 8:00 high and 8:00 low. The next candle is wide and trades through both lines inside an amber box. Three numbered amber markers point at the chart: 1 narrow 8:00, 2 high taken, 3 low taken. The caption bar reads: 9:00 missed both sides, 1 in 426 in amber. What do you mark first? Three options: 1 8:00 high, 2 8:00 low, 3 Both, with Reply 1, 2 or 3. Herman Trading and hermantrading.pro in the footer. It illustrates an NQ futures backtest: after a narrow 8:00 ET hourly candle, the 9:00 hour almost always takes its high, its low or both.

222

Herman Trading · Oct 28, 2026 · 12:00 PM UTC

Hey @grok How long does it REALLY take to become a consistently profitable trader?

28,534

Herman Trading · Oct 27, 2026 · 6:30 PM UTC

@RHermanOct 27

40% win rate at 1:2. You lose 6 trades out of 10. And still make money. Breakeven win rate at 1:2: 33.3%. Expectancy: +0.20R per trade. After 0.1R of costs per trade: +0.10R. The catch: over 100 trades, a run of 5+ losses in a row is about 98% likely. The label is fine. The question is whether you can stomach it. How many losses in a row before you'd start doubting this strategy?

Herman Trading Trade Facts card styled like a food nutrition label. Headline: Read the label before you trade it. Serving size 1 trade, 100 per sample. Win rate 40%, average win 2.0R, average loss 1.0R, breakeven win rate 33.3%, profit factor 1.33, expectancy +0.20R, after costs of 0.1R per trade +0.10R. Chance of 5 or more losses in a row: 98%. Note: illustrative math, not a backtest.

ALT Herman Trading Trade Facts card styled like a food nutrition label. Headline: Read the label before you trade it. Serving size 1 trade, 100 per sample. Win rate 40%, average win 2.0R, average loss 1.0R, breakeven win rate 33.3%, profit factor 1.33, expectancy +0.20R, after costs of 0.1R per trade +0.10R. Chance of 5 or more losses in a row: 98%. Note: illustrative math, not a backtest.

1,185

Herman Trading · Oct 27, 2026 · 5:48 PM UTC

@RHermanOct 27

UPDATE: TradingView just changed course on invite-only indicators. November 1 changes nothing. A month ago the plan was clear: from Nov 1, no more selling access to invite-only scripts outside the TradingView Marketplace. Today the Creator Team sent vendors an update: → New terms are coming, shaped by vendor feedback → Selling your indicators on your own website or inside your trading course will be allowed under the new terms → On Nov 1 nothing changes for vendors or their users → Updated terms and an FAQ land the week after Nov 1 More than 1,500 script authors reached out to them. I was one of them. If you use my invite-only indicators: keep using them as normal. If you sell invite-only scripts and haven't emailed [email protected] yet, the window closes Nov 1. Would you rather buy an indicator through the TradingView Marketplace or straight from the creator?

Phone screenshot of the TradingView Blog. A black SOCIAL label sits above the headline: Marketplace update: More flexibility and features are coming, dated Oct 27. Below it is a dark banner with a glowing purple edge: on the left a button reading Indicator marketplace with a shopping bag icon, on the right a card titled Terms of Use, Policies, and Disclaimers with small legal text. Under the banner the article begins: A month ago, we shared our plans to make selling indicators and strategies on TradingView clearer and more transparent. Since then, we've received a wealth of thoughtful suggestions and positive feedback. Context: TradingView's 27 October 2026 update for vendors of invite-only indicators and Pine Script strategies, explaining that updated Marketplace terms are coming and nothing changes on November 1.

ALT Phone screenshot of the TradingView Blog. A black SOCIAL label sits above the headline: Marketplace update: More flexibility and features are coming, dated Oct 27. Below it is a dark banner with a glowing purple edge: on the left a button reading Indicator marketplace with a shopping bag icon, on the right a card titled Terms of Use, Policies, and Disclaimers with small legal text. Under the banner the article begins: A month ago, we shared our plans to make selling indicators and strategies on TradingView clearer and more transparent. Since then, we've received a wealth of thoughtful suggestions and positive feedback. Context: TradingView's 27 October 2026 update for vendors of invite-only indicators and Pine Script strategies, explaining that updated Marketplace terms are coming and nothing changes on November 1.

1,273

Herman Trading · Oct 27, 2026 · 5:48 PM UTC