--- title: "TJR Trading Strategy: Asia Sweep Backtest on NQ | PickMyTrade" slug: tjr-trading-strategy-asia-sweep-backtest date: 2026-09-17 modified: 2026-09-17 author: Bhavishya Goyal excerpt: "We backtested the TJR Asia session sweep strategy on NQ futures: 348 out-of-sample setups, profit factor 1.35. Rules, settings, and how to automate it." meta_description: "We backtested the TJR Asia session sweep strategy on NQ futures: 348 out-of-sample setups, profit factor 1.35. Rules, settings, and how to automate it." focus_keyword: tjr trading strategy canonical_url: "https://blog.pickmytrade.trade/tjr-trading-strategy-asia-sweep-backtest/" og_title: "TJR Trading Strategy: Asia Sweep Backtest on NQ | PickMyTrade" og_description: "We backtested the TJR Asia session sweep strategy on NQ futures: 348 out-of-sample setups, profit factor 1.35. Rules, settings, and how to automate it." og_image: "https://blog.pickmytrade.trade/wp-content/uploads/2026/09/tjr-trading-strategy-asia-sweep-backtest-featured-1024x576.avif" schema_type: FAQPage categories: - Automated Trading - Tradingview Strategy tags: - asia session sweep - fair value gap - nq futures - order block - Pine Script - tjr strategy - tjr trades reading_time: 12 word_count: 2923 robots: "index, follow" lang: en-US --- # TJR Trading Strategy: Asia Session Sweep Backtest on NQ _By PickMyTrade_ Table of Contents 1. [Who is TJR?](https://blog.pickmytrade.trade/#who-is-tjr) 2. [What is the TJR Asia session sweep model?](https://blog.pickmytrade.trade/#what-is-the-tjr-asia-session-sweep-model) 3. [How we tested it](https://blog.pickmytrade.trade/#how-we-tested-it) 4. [What were the backtest results?](https://blog.pickmytrade.trade/#what-were-the-backtest-results) 5. [Two settings changed everything](https://blog.pickmytrade.trade/#two-settings-changed-everything) 6. [Where did the profit actually come from?](https://blog.pickmytrade.trade/#where-did-the-profit-actually-come-from) 7. [Does it survive real trading costs?](https://blog.pickmytrade.trade/#does-it-survive-real-trading-costs) 8. [What the numbers don’t show](https://blog.pickmytrade.trade/#what-the-numbers-dont-show) 9. [TJR strategy settings](https://blog.pickmytrade.trade/#tjr-strategy-settings) 10. [How do you automate the TJR strategy with PickMyTrade?](https://blog.pickmytrade.trade/#how-do-you-automate-the-tjr-strategy-with-pickmytrade) 11. [Get the full script](https://blog.pickmytrade.trade/#get-the-full-script) 12. [FAQ](https://blog.pickmytrade.trade/#faq) 13. [What is the TJR trading strategy?](https://blog.pickmytrade.trade/#what-is-the-tjr-trading-strategy) 14. [Who is TJR?](https://blog.pickmytrade.trade/#who-is-tjr) 15. [Does the TJR strategy actually work?](https://blog.pickmytrade.trade/#does-the-tjr-strategy-actually-work) 16. [What broker does TJR use?](https://blog.pickmytrade.trade/#what-broker-does-tjr-use) 17. [What time frame does the TJR Asia sweep strategy use?](https://blog.pickmytrade.trade/#what-time-frame-does-the-tjr-asia-sweep-strategy-use) 18. [Can I use the TJR strategy in a prop firm challenge?](https://blog.pickmytrade.trade/#can-i-use-the-tjr-strategy-in-a-prop-firm-challenge) 19. [Sources](https://blog.pickmytrade.trade/#sources) TJR’s liquidity-sweep model is one of the most copied setups in futures trading right now. The idea fits in one sentence. Let the Asia session build a range. Wait for London or New York to run the stops on one side of it. Then trade back the other way once structure breaks. We coded that model in Pine Script and ran it on 3.7 years of 5-minute Nasdaq-100 data. We picked settings on 2023–2024 only, then tested them unchanged on 2025–2026. Below you’ll find the rules, every result (including the ugly parts), the exact settings, and how to automate it with PickMyTrade. _TJR Trades is not affiliated with PickMyTrade and hasn’t reviewed or endorsed this script. This guide summarizes publicly taught concepts and documents our own independent implementation and backtest._ > **Key Takeaways** > > > > > > > > - On 2025–2026 data the model never saw, the tuned version made **+$137,213 on 348 setups** with a **profit factor of 1.35** (2 NQ contracts, costs included). > - With default settings it was much weaker: **profit factor 0.99 in 2023** and 1.23 on 2025–2026. > - **Short trades made 91% of the out-of-sample profit.** Longs were close to flat. > - It still held a profit factor of **1.29 with 4 ticks of slippage and $5 commission** per side. > - The catch: a **$29,413 max drawdown** at 2 NQ contracts. Scaled to 2 MNQ it’s about $2,900, more than many $50K prop evaluations allow. ## Who is TJR? {#who-is-tjr} TJR is the trading name of Tyler Riches, a futures and forex educator who runs TJR LLC. His [official About page](https://www.jointjrtrades.com/about) lists **1.8 million+ YouTube subscribers** and a Discord community of more than 50,000 members. He says he started trading crypto at 16 and now lives in Puerto Rico. He built his audience by trading live on stream, wins and losses both. That’s also why his name gets searched so often. People see a trade on a clip, then go looking for the rules behind it. TJR’s own site lists the concepts he teaches: break of structure, fair value gaps, order blocks, session highs and lows, and liquidity. None of them is new. Most traders will recognize the smart-money (ICT) vocabulary. A [breakdown by SnappChart](https://www.snappchart.app/blog/beginner-playbook/tjr-ict-trading-strategy) describes his contribution as a sequencing layer: a fixed order for applying those ideas, not a new theory. ## What is the TJR Asia session sweep model? {#what-is-the-tjr-asia-session-sweep-model} The Asia session sweep is the version of TJR’s teaching that traders code most often. Several community scripts on TradingView already use it, including [this Asia sweep + FVG strategy](https://www.tradingview.com/script/2XfMBwdR/). TJR hasn’t published one official coded ruleset. So treat any script, ours included, as an interpretation of the concepts, not his exact trading. Here’s the sequence our script follows: 1. **Mark the Asia range.** Track the high and low from 8:00 PM to midnight New York time. 2. **Wait for a sweep.** During London (3:00–8:00 AM ET) or the New York morning (9:30 AM–12:00 PM ET), price wicks beyond the Asia high or low and closes back inside. The wick has to be at least 0.3× ATR, so tiny pokes don’t count. 3. **Confirm a break of structure.** A strong candle (body larger than half an ATR) closes past the last 8-bar swing in the opposite direction. It has to happen within 40 bars of the sweep. 4. **Mark the entry zone.** That’s the last opposing candle before the break (the order block), or a fair value gap that forms after it. 5. **Enter on the retest.** Price trades back into the zone and closes in the new direction. 6. **Stop beyond the sweep.** The stop goes past the sweep wick plus half an ATR of buffer. 7. **Scale out.** Half the position exits at 1.5R. The stop on the rest moves to breakeven, and it runs to 4R. 8. **Stay flat by the afternoon.** Anything still open closes when the New York morning session ends. The script takes at most 2 trades and 2 losses per day. Here’s what a typical winning trade looked like. It isn’t the best trade in the test. It’s the one closest to the median winner. ![Example TJR Asia sweep short on NQ: Asia range, liquidity sweep of the high, break of structure, order block retest entry, stop, TP1 and TP2](https://blog.pickmytrade.trade/wp-content/uploads/2026/09/tjr-asia-sweep-example-trade-nq.avif) Price swept the Asia high at 3:30 AM, broke structure lower at 4:15, and came back into the order block. The short triggered at 4:20 AM. TP1 filled within the hour, and the runner closed at breakeven. > **In short:** The TJR Asia sweep model waits for London or New York to take out the stops sitting beyond the overnight Asia range. It only enters after a break of structure confirms the move has actually turned. The stop sits beyond the sweep wick, and profits are taken at fixed multiples of that risk. ## How we tested it {#how-we-tested-it} Most strategy backtests you’ll find online use a few weeks of chart data. Our first TradingView run had the same problem. A Basic plan loads only about a month of 5-minute bars, which gave us 8 setups. That’s nowhere near enough to judge anything. So we rebuilt the strategy bar by bar in Python and copied TradingView’s order-fill rules. The emulator assumes the path within each bar runs open → high → low → close (or open → low → high → close), [as TradingView documents](https://www.tradingview.com/support/solutions/43000786181-broker-emulator/). Market orders fill on the next bar’s open. To check the rebuild, we ran both on the same month. The results lined up almost exactly: | Aug 17 – mid-Sep 2026 | TradingView (NQ1!) | Python rebuild | | --- | --- | --- | | Net P&L | −$13,465 | −$14,439 | | Profit factor | 0.13 | 0.12 | | Sep 3 short | Entry 6:35 AM ET, stopped 9:30 AM | Entry 6:35 AM ET, stopped 9:30 AM | | Aug 31 short | Entry 11:05 AM, out 11:25 AM | Entry 11:05 AM, out 11:25 AM | Then the full test, set up like this: | Test setting | Value | | --- | --- | | Data | Nasdaq-100 5-minute prices from Dukascopy, Jan 2023 – Sep 2026 | | Contract math | $20 per point, 0.25-point ticks (CME E-mini Nasdaq-100 specs) | | Size | 2 contracts per trade (the 50% scale-out is 1 contract) | | Costs | $2.50 per contract per side, 1 tick slippage on market and stop fills | | Settings search | 384 combinations, scored on 2023–2024 only | | Out-of-sample test | Jan 2025 – Sep 2026, settings unchanged | The Dukascopy feed is an index CFD, not NQ futures. Price levels differ slightly, but the sessions and intraday structure line up closely, and the one month we could check against real NQ1! data matched. > **In short:** We validated a bar-by-bar Python rebuild against TradingView’s own Strategy Tester, trade for trade. We then tested 3.7 years of Nasdaq-100 5-minute data at 2 contracts with commissions and slippage. Settings were chosen on 2023–2024 and scored only on 2025–2026. ## What were the backtest results? {#what-were-the-backtest-results} ![Key stats for the TJR Asia sweep strategy: default vs optimized settings, in-sample 2023–2024 and out-of-sample 2025–2026](https://blog.pickmytrade.trade/wp-content/uploads/2026/09/tjr-strategy-backtest-key-stats.avif) | Out-of-sample: Jan 2025 – Sep 2026 | Default settings | Optimized settings | | --- | --- | --- | | Net profit | +$39,084 | +$137,213 | | Profit factor | 1.23 | 1.35 | | Win rate (per setup) | 46.5% | 45.7% | | Setups traded | 159 | 348 | | Average win / average loss | +$2,873 / −$2,041 | +$3,324 / −$2,070 | | Max drawdown | −$26,942 | −$29,413 | The tuned settings made money in every calendar year: +$39,011 in 2023, +$76,805 in 2024, +$50,204 in 2025, and +$87,009 in 2026 through mid-September. The defaults didn’t. They lost $578 in 2023 and made most of their money in 2026. ![Equity curve for default vs optimized TJR strategy settings, with out-of-sample period shaded and drawdown panel](https://blog.pickmytrade.trade/wp-content/uploads/2026/09/tjr-strategy-equity-curve-default-vs-optimized.avif) A win rate under 50% is normal for this kind of model. Winners average about 1.6× the size of losers, because half of every winning trade runs toward 4R. ![Monthly net P&L of the optimized TJR strategy from 2023 to 2026](https://blog.pickmytrade.trade/wp-content/uploads/2026/09/tjr-strategy-monthly-pnl-2023-2026.avif) About two-thirds of months finished positive. The worst was August 2026 at roughly −$20,000. That month overlaps our TradingView validation window, so if you load the script today, the first thing you’ll see is a losing stretch. We left it in. > **In short:** On 2025–2026 data the tuned TJR sweep model never saw, it returned +$137,213 on 348 setups at 2 NQ contracts, with a profit factor of 1.35 and a 45.7% win rate. Every calendar year from 2023 to 2026 was profitable, but the worst drawdown reached $29,413. ## Two settings changed everything {#two-settings-changed-everything} We started from the script’s original defaults. Two changes made most of the difference, and one of them surprised us. 1. **We turned off the 4-hour EMA bias filter.** Our original script only allowed longs above the 50-period 4H EMA and shorts below it. Removing the filter roughly doubled the number of setups, and the extra trades were profitable. 2. **We moved the runner target from 3R to 4R.** The first half still exits at 1.5R. That first result cuts against what a lot of smart-money traders believe. A sweep is, by definition, a move against the recent push. A trend filter blocks many of the best reversals right when they start. We also didn’t want to just trust the single best combination. So we plotted every one of the 384 combinations: how it did in 2023–2024 against how it did on the unseen data. ![All 384 TJR strategy setting combinations: in-sample vs out-of-sample profit factor, and out-of-sample profit factor by take-profit targets](https://blog.pickmytrade.trade/wp-content/uploads/2026/09/tjr-strategy-robustness-384-settings.avif) Every combination stayed above a profit factor of 1.0 on the unseen data. The worst scored 1.13. The heatmap shows why the 4R runner won: wider runner targets did better at every TP1 level. > **In short:** Two changes improved the TJR sweep model: dropping the 4-hour EMA trend filter and letting the runner reach 4R. All 384 tested combinations stayed profitable out of sample, so the result doesn’t depend on one lucky setting. ## Where did the profit actually come from? {#where-did-the-profit-actually-come-from} This is the part most strategy posts skip. We split the 348 out-of-sample setups three ways: Where the TJR strategy’s out-of-sample profit came from BY DIRECTION Shorts (187)+$125,333 Longs (161)+$11,880 BY SESSION New York (138)+$102,108 London (210)+$35,105 BY ENTRY ZONE Fair value gap (162)+$120,832 Order block (186)+$16,381 Out-of-sample net profit, Jan 2025 – Sep 2026, optimized settings, 2 NQ contracts. Setup counts in brackets. *Out-of-sample net profit, Jan 2025 – Sep 2026, optimized settings, 2 NQ contracts. Setup counts in brackets.* Three things stand out: - Shorts did almost all the work. They made $125,333. Longs made $11,880 on nearly as many trades. Nasdaq rose over this period, so this isn’t a bear-market effect. Downside sweeps, where stops above the Asia high get run, simply followed through harder. - New York beat London. London produced more setups (210 vs 138) but far less profit. It also had a lower win rate: 42.9% vs 50.0%. - Fair value gaps beat order blocks. FVG retests made seven times what order-block retests made. We didn’t re-optimize around those splits. Filtering to “New York shorts off FVG only” would look great on paper. It would also be exactly the kind of curve-fitting the walk-forward test is meant to prevent. Treat them as things to watch in your own trading, not rules to add. > **In short:** Out of sample, short setups produced $125,333 of the TJR model’s $137,213 profit. New York-session trades made $102,108 against London’s $35,105, and fair value gap entries made $120,832 against $16,381 from order blocks. ## Does it survive real trading costs? {#does-it-survive-real-trading-costs} A backtest that only works at one tick of slippage isn’t worth much. So we reran the out-of-sample period with worse fills and double commissions. ![Stress test of the TJR strategy with 1, 2 and 4 ticks of slippage and $2.50 or $5.00 commission](https://blog.pickmytrade.trade/wp-content/uploads/2026/09/tjr-strategy-slippage-commission-stress-test.avif) At 4 ticks of slippage per stop and market fill and $5.00 commission per side, net profit dropped to +$116,108 and the profit factor to 1.29. The median stop was about 49 points (roughly 196 ticks). A few extra ticks barely dent a trade that size. > **In short:** Raising slippage from 1 to 4 ticks and commission from $2.50 to $5.00 per side cut the out-of-sample profit factor only from 1.35 to 1.29. The median stop was about 49 NQ points, so execution costs were small relative to each trade’s risk. ## What the numbers don’t show {#what-the-numbers-dont-show} Here’s what the numbers above don’t show. - The drawdown is large. At 2 NQ contracts, the worst peak-to-trough drop was $29,413, and the longest losing streak was 8 setups in a row. Scale down to 2 MNQ contracts (1/10 the size, [per CME specs](https://www.cmegroup.com/markets/equities/nasdaq/micro-e-mini-nasdaq-100.contractSpecs.html)) and that’s about $2,900. That’s still more than the trailing drawdown on many $50K prop evaluations. See our guide to [trailing drawdown for automated traders](https://blog.pickmytrade.trade/trailing-drawdown-explained-for-automated-futures-traders/) before running this in a challenge. - Stop sizes vary a lot. The median stop was 49 points, but the widest 10% were over 135 points. The script trades a fixed contract count, so a wide-stop trade risks far more than a tight one. The largest single loss was $11,790. - The data is a proxy. The long test used Nasdaq-100 CFD prices, not NQ futures. The one month we could compare directly matched closely, but futures fills around rollover and the daily halt can differ. - It’s our interpretation. TJR’s live trading involves judgment calls this script can’t make: which liquidity matters, when to skip a session, how news changes the plan. - Past results aren’t a forecast. 2026 was the strongest year in the sample, and a quieter market could look very different. One practical warning if you run the script on TradingView. On our first run it placed zero trades and showed no error. Two NQ contracts at around 29,500 carry about $1.18 million in notional value. With 5% margin that needs about $59,000, more than the $50,000 test account, so every order was rejected silently. [TradingView cancels orders when equity is below required margin](https://www.tradingview.com/support/solutions/43000717375-how-to-simulate-trading-with-leverage-in-pine-script/). Our script sets margin to 1%, a conservative stand-in for intraday futures margin. ## TJR strategy settings {#tjr-strategy-settings} These are the tested defaults in TJR Trades Strategy : | Setting | Default | What it does | | --- | --- | --- | | Asia session | 20:00–23:59 ET | Builds the liquidity range | | London / NY sessions | 03:00–08:00 / 09:30–12:00 ET | Windows where sweeps and entries are allowed | | 4H bias filter | Off | Optional trend filter (hurt results in testing) | | Swing lookback (BOS) | 8 bars | Swing that must break to confirm structure | | Min sweep size | 0.3 × ATR | Filters out tiny wicks | | Require strong BOS | On | Break candle body must exceed 0.5 × ATR | | Require confirm candle | On | Entry candle must close in trade direction | | OB + FVG both required | Off | Either zone type can trigger an entry | | TP1 (exits 50%) | 1.5R | First partial target | | TP2 (runner) | 4R | Final target for the remaining half | | Stop buffer | 0.5 × ATR | Distance beyond the sweep wick | | Breakeven after TP1 | On | Moves the runner’s stop to entry | | Max trades / losses per day | 2 / 2 | Daily guardrails | | Flatten after NY session | On | No afternoon or overnight exposure | Want the chart-only version? The ](https://www.tradingview.com/script/AB5MKN7i-TJR-Trades-Indicator-PickMyTrade/) is free on TradingView. It marks the Asia range, sweeps, breaks of structure, order blocks and FVGs, but doesn’t place trades. If you’re converting a chart tool yourself, see [how to turn a TradingView indicator into a strategy](https://blog.pickmytrade.trade/convert-tradingview-indicator-to-strategy/). ## How do you automate the TJR strategy with PickMyTrade? {#how-do-you-automate-the-tjr-strategy-with-pickmytrade} Every step in this model is a rule the script can check without judgment. That’s what makes it automatable. A basic setup takes four steps: 1. Add TJR Trades Strategy to an NQ1! or MNQ1! 5-minute chart. 2. Create a TradingView alert on the strategy with the PickMyTrade webhook URL. 3. Paste the PickMyTrade JSON message using `{{strategy.order.action}}` and `{{strategy.order.contracts}}`, so the scale-out and runner exits route correctly. 4. Pick your broker in PickMyTrade: Tradovate, Rithmic, TradeStation, Interactive Brokers, or ProjectX-based prop firms. If you’re on Tradovate, read [how to automate without the API add-on fee](https://blog.pickmytrade.trade/tradovate-automation-skip-the-api-fee-and-cme-license/) first. And run it on a demo or evaluation account before real money. The first live week is where you find out whether your broker’s fills look like the backtest’s. > **In short:** The TJR sweep strategy is fully rule-based, so TradingView alerts can send its entries, partial exits and breakeven stops to a broker through PickMyTrade. Start on a demo or evaluation account to compare live fills with the backtest. ## Get the full script {#get-the-full-script} The complete TJR Trades Strategy script is available in our Discord community. It includes the scale-out logic, daily guardrails and the corrected margin settings. You’ll also find scripts for other traders we’ve covered the same way: [Fabio Valentini’s order-flow scalper](https://blog.pickmytrade.trade/fabio-valentini-pro-scalper-nasdaq-scalping-strategy/), [Andrea Cimitan’s opening range breakout](https://blog.pickmytrade.trade/andrea-cimitan-opening-range-breakout-strategy/), [Marci Silfrain’s gold strategy](https://blog.pickmytrade.trade/marci-silfrain-strategy-gold/) and [Inna Rosputnia’s 18-day system](https://blog.pickmytrade.trade/inna-rosputnia-strategy-the-18-day-system-from-2026s-1-futures-trader/). [Join the PickMyTrade Discord](https://discord.com/invite/FkJ7g7498S) ## FAQ {#faq} ### What is the TJR trading strategy? {#what-is-the-tjr-trading-strategy} It’s a smart-money approach built on market structure, liquidity sweeps, order blocks and fair value gaps. The Asia session sweep version waits for London or New York to take out the overnight range. It enters after a break of structure, on a retest of the order block or gap. ### Who is TJR? {#who-is-tjr} TJR is Tyler Riches, founder of TJR LLC and a trading educator with more than 1.8 million YouTube subscribers. He’s known for live-streaming his trades, wins and losses included. ### Does the TJR strategy actually work? {#does-the-tjr-strategy-actually-work} In our test, a coded version made +$137,213 on NQ from January 2025 to September 2026, at a 1.35 profit factor with 2 contracts. The settings were chosen only on 2023–2024 data. The default settings were much weaker, and every version had drawdowns of $25,000 or more at that size. ### What broker does TJR use? {#what-broker-does-tjr-use} We couldn’t find a reliable primary source confirming TJR’s personal broker, so we won’t guess. The strategy itself isn’t tied to a broker. Through PickMyTrade it runs on Tradovate, Rithmic, TradeStation, Interactive Brokers and ProjectX-based prop firms. ### What time frame does the TJR Asia sweep strategy use? {#what-time-frame-does-the-tjr-asia-sweep-strategy-use} Our version runs on a 5-minute NQ chart. The Asia range builds from 8:00 PM to midnight New York time, and trades are allowed from 3:00–8:00 AM and 9:30 AM–12:00 PM. ### Can I use the TJR strategy in a prop firm challenge? {#can-i-use-the-tjr-strategy-in-a-prop-firm-challenge} You can, but size it carefully. At 2 MNQ contracts the backtest’s worst drawdown was about $2,900, which exceeds many $50K evaluation limits. Test on a demo first, and consider 1 contract or tighter daily loss limits. --- _Educational content, not financial advice. TJR Trades and Tyler Riches are not affiliated with PickMyTrade; this guide is an independent interpretation of publicly taught concepts, with an independent Pine Script implementation and backtest. All figures are hypothetical backtest results with modeled commission and slippage, using Nasdaq-100 CFD price data as a proxy for NQ futures. Past performance does not guarantee future results. Futures trading carries substantial risk of loss._ ## Sources {#sources} - [TJR Trades: About TJR](https://www.jointjrtrades.com/about) - [SnappChart: TJR’s ICT Trading Strategy, What TJR Actually Teaches](https://www.snappchart.app/blog/beginner-playbook/tjr-ict-trading-strategy) - [TradingView: TJR Asia session sweep + FVG entry (community script)](https://www.tradingview.com/script/2XfMBwdR/) - [CME Group: E-mini Nasdaq-100 futures contract specs](https://www.cmegroup.com/markets/equities/nasdaq/e-mini-nasdaq-100.contractSpecs.html) - [CME Group: Micro E-mini Nasdaq-100 futures contract specs](https://www.cmegroup.com/markets/equities/nasdaq/micro-e-mini-nasdaq-100.contractSpecs.html) - [TradingView Help: Broker emulator](https://www.tradingview.com/support/solutions/43000786181-broker-emulator/) - [TradingView Help: Simulating leverage in Pine Script](https://www.tradingview.com/support/solutions/43000717375-how-to-simulate-trading-with-leverage-in-pine-script/) - [TradingView Help: Bar Magnifier backtesting mode](https://www.tradingview.com/support/solutions/43000669285-what-is-bar-magnifier-backtesting-mode/) - [Dukascopy: Historical data feed](https://www.dukascopy.com/swiss/english/marketwatch/historical/) For AI tools & developers:[View Markdown →](https://blog.pickmytrade.trade/tjr-trading-strategy-asia-sweep-backtest.md)