NQ, ES & Gold futures | Probability over opinion | ICT + backtests | TradingView indicators | https://xitter.cf/t.co/pruz2NQ6Tu | https://xitter.cf/t.co/7xIEUoPhSw

Portsmouth, United Kingdom
Joined December 2017
What does trading with PROBABILITIES actually look like? Today’s London Session $NQ was a perfect example. Intelligent Balance Probability Map indicator showed: 81% LONGS 📈 That doesn’t mean I blindly buy. It means I already know which side I want to focus on - and which side I can ignore. So I waited for an A+ setup using one of the most basic ICT concepts: Liquidity Sweep → iFVG → continuation towards the MORE PROBABLE side. No predicting every candle. No forcing shorts against an 81% long bias. No need for 20 different confirmations. Probability gives me the direction. Price action gives me the entry. That’s essentially how I trade. Would you rather trade every setup you see - or only the ones aligned with an 81% probability edge?
5
3
32
5,700
ChatGPT Pro weekly Codex limit: gone in 1 day. Claude Max can't run full speed for 7 days either. That's Herman AI Studio working at full power. Grok bot got higher limits. Still not enough. 2.32B tokens in the last 30 days. The result: 2 more online stores are live and running. Not trading related. Which limit do you hit first: Codex or Claude?
2
1
546
2 new online stores are open. Not trading related. The receipts: Codex weekly limit at 7%, 6 days before reset.
121
🚨 TradingView Section 22 UPDATE: November 1 is NOT a migration deadline. I have it in writing from @tradingview 's Creator Team. After weeks of uncertainty around commercial Invite-Only Scripts, I've received an official response. Here's what's CONFIRMED: ✅ November 1 is ONLY the deadline to contact TradingView, not to complete the transition. ✅ NOTHING changes for creators or customers on November 1. ✅ Existing Lifetime and fixed-term customers keep access without interruption or extra payment under the upcoming transition. ✅ Existing subscriptions remain valid until their end date. ✅ The 100-boost requirement to apply for the TradingView Marketplace has been DROPPED. TradingView says the new program will be more flexible for vendors. BUT the final commercial terms, transition requirements and effective date are still pending. A public announcement and FAQ are expected before the end of October. This is welcome progress, but the full picture isn't here yet. To all Pine Script developers selling premium indicators: Does this change your plans to stay with TradingView, or are you waiting for the final terms?
🚨 TradingView killed the $29.95/user fee drama - but replaced it with something MUCH bigger. From Nov 1: • Commercial Invite-Only scripts/indicators move to Marketplace • Whop → TradingView paid access effectively ends • 7-day refund window • No grandfathered pricing • Payout settlement can take up to 30 days • 0% platform fee today - but not guaranteed forever TradingView is now taking much more control over the paid Pine Script ecosystem. Better deal - or just a different kind of platform risk? Stay with @tradingview , move to @NinjaTrader / cTrader / MetaTrader / Sierra Chart, or build your own? #TradingView
3
5
1,034
TradingView just replied about the invite-only deadline. 1 November is a contact deadline, not a switch-off date. Quick context: Section 23 of the TradingView Terms bans selling access to invite-only scripts outside their Marketplace from 1 November. I applied to the Marketplace. Today the Creator team wrote back: 1. Vendors who already contacted them don't have to finish the move by 1 Nov. 2. Nothing changes for current users on 1 Nov. 3. The final terms will set the real date and the steps. 4. Paid before that date? Lifetime and fixed-term access keep working. No extra payment. 5. The 100-boost requirement to apply is gone. 6. More Paid Spaces payment options. Annual plans coming soon. Public announcement and FAQ expected before the end of October. Herman Macro Edge members: nothing changes for you on 1 November. Paying for an invite-only indicator? Has your vendor told you what happens after 1 November?
3
5
1,250
𝗣𝗹𝗮𝗶𝗻 𝗢𝗥𝗕 𝗼𝗻 𝗡𝗤 𝗶𝘀 𝗿𝗼𝘂𝗴𝗵𝗹𝘆 𝗯𝗿𝗲𝗮𝗸-𝗲𝘃𝗲𝗻. 𝗧𝗵𝗿𝗲𝗲 𝗳𝗶𝗹𝘁𝗲𝗿𝘀 𝘁𝘂𝗿𝗻 𝗶𝘁 𝗶𝗻𝘁𝗼 𝗣𝗙 𝟮.𝟮𝟰. The setup: OR30 (the 09:30 to 09:59 range), entry on the first 5-minute close beyond it, either side, stop at the other side, half off at 1R, rest to the close, no new entries after 12:00. The three filters: 1. London range (02:00 to 08:00) vs the last 24 days: normal 2. The 08:30 news window (08:30 to 08:34) vs the last 24 days: quiet, under 1x 3. The setup's own last 10 trades: a net loss PF 2.31 in 2015-21, 2.24 in 2022-26. $492 per trade on 1 NQ. Win rate 62.61%. 230 trades in 12 years, 1.6 a month. Positive in 12 of 12 years. The catch: 2022-26 was used to confirm it, so live data is the first clean test. Would you trade a setup that fires 1.6 times a month? Source: Herman Trading Backtesting Centre, NQ Opening Range Breakout (nq-orb-study), 2015-2026
2
4
21
2,605
Hey @grok Prop firm eval with a 30% consistency rule: no single day can be more than 30% of total profit. What is the minimum number of profitable days needed to pass?
5
3
2,782
𝗡𝗲𝘄 𝗬𝗼𝗿𝗸 𝗺𝗮𝗱𝗲 𝘁𝗵𝗲 𝗹𝗼𝘄 𝗼𝗳 𝘁𝗵𝗲 𝗱𝗮𝘆 𝗼𝗻 𝗮𝗹𝗹 𝟱 𝗱𝗮𝘆𝘀 𝘁𝗵𝗶𝘀 𝘄𝗲𝗲𝗸. 𝗜𝘁 𝗺𝗮𝗱𝗲 𝘁𝗵𝗲 𝗵𝗶𝗴𝗵 𝗼𝗻𝗰𝗲. NQ, 12 to 16 October, 1-minute bars, New York time: → New York printed the day's low in 5 of 5 sessions → Asia held the day's high in 4 of 5 → The London low was the first side taken, 5 days out of 5 → New York closed beyond a London level every day Across 2015 to 2026 the day's high usually belongs to New York: 69.1% of days, against 18.4% for Asia. The low is New York's on 62.6%. So the low side ran to script this week. The high side did the opposite. Monday morning, which level goes on your chart first: 1 the Asia high, 2 the London low, 3 the NY low? Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026
3
2
27
2,364
𝗧𝗵𝗲 𝗟𝗼𝗻𝗱𝗼𝗻 𝗟𝗼𝘄 𝗿𝘂𝗻𝘀 𝗳𝘂𝗿𝘁𝗵𝗲𝗿 𝘁𝗵𝗮𝗻 𝘁𝗵𝗲 𝗟𝗼𝗻𝗱𝗼𝗻 𝗛𝗶𝗴𝗵. 𝟬.𝟵𝟯𝘅 𝘃𝘀 𝟬.𝟴𝟬𝘅 𝘁𝗵𝗲 𝗟𝗼𝗻𝗱𝗼𝗻 𝗿𝗮𝗻𝗴𝗲, 𝗺𝗲𝗱𝗶𝗮𝗻, 𝗼𝗻 𝗡𝗤. Once New York takes a London level, the move past it is measured in London ranges, not in points. Past the London High: median 0.80x the London range (P30-P70 0.43-1.25x, n 1,285) Past the London Low: median 0.93x (P30-P70 0.52-1.57x, n 1,480) The whole New York range: median 2.35x the London range (n 2,877) Days where the 08:00 open is already at or beyond the level are left out. So the London range is the measuring stick for the New York session. One target, scaled to the day, instead of a round number of points. Where do you put your first target after New York takes a London level? Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026
3
17
1,852
𝗧𝗵𝗲 𝘁𝗲𝘅𝘁𝗯𝗼𝗼𝗸 𝗣𝗼𝟯 𝗱𝗮𝘆 𝗵𝗮𝗽𝗽𝗲𝗻𝘀 𝗼𝗻 𝟮.𝟭% 𝗼𝗳 𝗡𝗤 𝗱𝗮𝘆𝘀. 𝗔𝘀𝗶𝗮 𝗮𝗰𝗰𝘂𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻, 𝗟𝗼𝗻𝗱𝗼𝗻 𝗺𝗮𝗻𝗶𝗽𝘂𝗹𝗮𝘁𝗶𝗼𝗻, 𝗡𝗲𝘄 𝗬𝗼𝗿𝗸 𝗲𝘅𝗽𝗮𝗻𝘀𝗶𝗼𝗻. n 2,832 days, 2015-2026. Each piece on its own is common. New York alone is an Expansion session on 62.2% of days, Accumulation on 20.5%, Manipulation on 17.2%. The full three-session sequence almost never lines up. Do you trade the Po3 story, or just what New York does with the London range?
1
1
6
1,434
𝗡𝗤 𝘀𝘄𝗲𝗽𝘁 𝘁𝗵𝗲 𝗟𝗼𝗻𝗱𝗼𝗻 𝗹𝗼𝘄 𝗶𝗻 𝗮𝗹𝗹 𝟱 𝗡𝗲𝘄 𝗬𝗼𝗿𝗸 𝘀𝗲𝘀𝘀𝗶𝗼𝗻𝘀 𝘁𝗵𝗶𝘀 𝘄𝗲𝗲𝗸. 𝗔𝗹𝗹 𝟱 𝗰𝗮𝗺𝗲 𝗯𝗮𝗰𝗸 𝘁𝗼 𝘁𝗵𝗲 𝗹𝗲𝘃𝗲𝗹 𝗯𝗲𝗳𝗼𝗿𝗲 𝘁𝗵𝗲 𝗹𝗼𝘄 𝘄𝗮𝘀 𝗶𝗻. Thursday is the clean one. London range 02:00-08:00 ET: 29,748.50 to 29,941.00. 08:30: the low goes. Then price trades 111 points back above the level, 0.58 of the London range. 09:57: the real low at 29,622.50, that is 126 points below the level. Not a one off. In the session study, after New York takes the first London side, price trades back to that level before making its extreme on 92.8% of days (n 2,747). A clean break that never comes back is rare. And none of this week's five was a gift. At 08:00 price opened 0.15 to 0.37 of the London range above the level, so the sweep needed a real move. Which leg do you trade: the sweep, the retest or the low? This week measured on NQ1! 1-minute data from TradingView, Friday to 12:58 ET. Historical frequencies, not a promise: past results do not guarantee future results. Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026
5
4
39
3,965
One NQ opening range setup made money in 12 of 12 years. 230 trades, $492 a trade on 1 NQ. Plain, the same breakout is break-even: PF 1.03 on 2,629 trades. 3 filters took it to PF 2.26: 1. A normal London range 2. A quiet 08:30 candle 3. Only after its last 10 trades lost money The video: one real trade from 21 Sep 2026, +1.28R. Would you take a breakout only after a losing streak? Source: Herman Trading Backtesting Centre, NQ Opening Range Breakout (nq-orb-study), 2015-2026
6
25
2,490
The full numbers, 2015-2026: 62.6% win rate, max drawdown $9,943, longest losing run 4 trades. Historical backtest, net of costs. Past results do not guarantee future results. Research, not financial advice.
4
413
𝟴𝟳 𝗡𝗤 𝗙𝗿𝗶𝗱𝗮𝘆𝘀. 𝗧𝗵𝗲 𝗹𝗼𝘄 𝗼𝗳 𝘁𝗵𝗲 𝗱𝗮𝘆 𝗽𝗿𝗶𝗻𝘁𝗲𝗱 𝗶𝗻 𝗡𝗲𝘄 𝗬𝗼𝗿𝗸 𝗼𝗻 𝟳𝟱.𝟵% 𝗼𝗳 𝘁𝗵𝗲𝗺. The filter is known at 02:00 ET: it is a Friday and London opens near the 18:00 daily open, within 0.25 TR. Distance to the level alone predicts 65.5%, so the condition adds 10.4 pp. The edge was positive in 10 of 12 years. Over the full 2015-2026 sample it is 78.0% on 218 days. What London did on those Fridays: Took the Asia high only, 39.1% Took the Asia low only, 33.3% Took both Asia sides, 26.4% Stayed inside the Asia range, 1.1% It shows up about 1.5 days a month. Read it as session context, not as an entry signal. Do you give Friday its own session profile, or trade it like any other day? Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026
3
9
1,621
Hey @grok $50k prop eval: 6% target, 2.5% trailing drawdown, 0.5% risk per trade, 1:2 R:R. What win rate do I need for a 50/50 chance of passing?
6
13
10,403
𝗢𝗻 𝗙𝗿𝗶𝗱𝗮𝘆𝘀, 𝗡𝗤'𝘀 𝗹𝗮𝘀𝘁 𝗵𝗼𝘂𝗿 𝗹𝗲𝗮𝘃𝗲𝘀 𝘁𝗵𝗲 𝗵𝗶𝗴𝗵 𝗮𝗹𝗼𝗻𝗲. 𝟴𝟰.𝟵% 𝗻𝗲𝘃𝗲𝗿 𝘁𝗮𝗸𝗲 𝘁𝗵𝗲 𝗽𝗿𝗲𝘃𝗶𝗼𝘂𝘀 𝗵𝗼𝘂𝗿'𝘀 𝗵𝗶𝗴𝗵. Backtest of the week, parameters to copy: Market: NQ futures, 1H candles, New York time Candle: the 16:00 hour, the last hourly candle of the day Filter: Friday only Level: the high of the 15:00 hour Outcome: that high is never traded through 84.9% of 159 candles in 2022-2026. 79.3% across 363 candles since 2015. Distance to the level alone predicts 68.3%, so Friday adds 16.6 pp. 11 of 12 years sat above their own baseline. When the high does get taken, it goes early: most first sweeps land in the first 10 minutes of the hour. Free sweeps are excluded, meaning candles that open within 0.10 TR of the level or already beyond it, where TR is the median range of the last 24 hourly candles. Same play on the other indices: ES 85.6%, YM 87.3%, RTY 78.7%. On gold it breaks at 53.6%, and that is the check that matters. These are event rates, not win rates. No entries, stops or costs are modelled. Do you trade the Friday close on NQ, or are you flat by 16:00 New York? Source: Herman Trading Backtesting Centre, NQ Candle Path Study v2 (nq-path-study-v2), 2015-2026
1
4
26
2,439
Herman AI Studio. The central brain of Herman Trading. Claude, ChatGPT and Grok now work in one place. In the Council they debate, then agree on the best call for my business. → Backtesting Centre: 12 verified studies on NQ, ES, GC, YM and RTY, 2015-2026 → 197/197 checks passed across the study library → Claude, ChatGPT and Grok Bot connected inside the Studio → Council: the three AIs consult on a question and land on the best decision for the business → X, YouTube, Instagram, TikTok, LinkedIn, Facebook and Discord managed from one screen I can't imagine running a business without tools like this anymore. Would you let AI help make your business decisions, or keep it to research?
5
1
1
19
3,750
Claude, ChatGPT and Grok share one screen now. I built it to run my trading business. This is Herman AI Studio. Private, built for one business: mine. What runs through it: → Backtesting: 12 verified studies, the newest an ES candle path study (2015-2026) → X posts: ideas, drafts and the schedule, straight into Typefully → LinkedIn, Instagram and TikTok through Buffer → Followers across 7 networks, tracked day by day → A Council where the AIs consult each other before I decide Claude builds it. I approve every step. The numbers in my posts come from verified backtests, not from a chatbot guessing. What's the first job in your trading you'd hand to an AI?
4
4
1,417
Recorded last night, 14 Oct, 23:07. Straight from my screen. 12 verified studies. 33 post ideas. 18 posts in the pipeline. One place.
1
296
@𝗴𝗿𝗼𝗸 𝗜𝘀 𝘁𝗵𝗲 𝗰𝗹𝗮𝘀𝘀𝗶𝗰 𝗼𝗽𝗲𝗻𝗶𝗻𝗴 𝗿𝗮𝗻𝗴𝗲 𝗯𝗿𝗲𝗮𝗸𝗼𝘂𝘁 𝘀𝘁𝗶𝗹𝗹 𝗽𝗿𝗼𝗳𝗶𝘁𝗮𝗯𝗹𝗲 𝗼𝗻 𝗘𝗦 𝗳𝘂𝘁𝘂𝗿𝗲𝘀? Define it as the 5-minute opening range, stop entry on either side, stop at the opposite side, 2R target, entries until 12:00 ET. Yes or no, and a profit factor. Then I'll post mine from 2015 to 2026.
7
2
12
9,983
𝗧𝗵𝘂𝗿𝘀𝗱𝗮𝘆 𝗶𝘀 𝗻𝗼𝘁 𝗮 𝘀𝗲𝘁𝘂𝗽. 𝗢𝗻 𝗡𝗤, 𝗡𝗲𝘄 𝗬𝗼𝗿𝗸 𝗲𝘅𝗽𝗮𝗻𝗱𝘀 𝗼𝗻 𝟱𝟵.𝟰% 𝘁𝗼 𝟲𝟱.𝟵% 𝗼𝗳 𝗱𝗮𝘆𝘀, 𝗠𝗼𝗻𝗱𝗮𝘆 𝘁𝗼 𝗙𝗿𝗶𝗱𝗮𝘆. The day on the calendar barely moves the number. On NQ, 2015-2026: New York is an Expansion session (takes a London side and closes beyond it) on 62.2% of all days. Across Monday to Friday that rate sits between 59.4% and 65.9%. The weekly high lands on Friday in 36.8% of weeks, about what shuffling the same days gives. What does move it is the session before: After a London Manipulation, New York expands on 67.4% of days. After a London Accumulation, New York stays in Accumulation on 34.6%, against 20.5% on all days. So the context is London, not the date. Historical frequencies, not a promise. Before the New York open, do you log what London did, or just the day of the week? Source: Herman Trading Backtesting Centre, NQ Session Study v2 (nq-session-study-v2), 2015-2026
1
16
1,798