Retail Sentiment Contrarian
When Naver searches and Telegram chatter triple on a coin that already ran, retail is late — fade the frenzy.
- SignalMentions ≥ 3× · +10%
- EntryShort perp
- Exit< 1.5× or −8%
How it works
One direction · holds 1–3 days- naver_trend20/hSearch interest per coin
- telegram_messages20/hChannel mentions
- ticker10/h24h price change
- premium10/hIs the KRW side overheating?
- SignalMentions ≥ 3× · +10%Naver search and Telegram mentions vs their 7-day average
- EntryShort perpPrice +10% in 24h and KRW premium ≥ 2%
- Exit< 1.5× or −8%Mentions under 1.5×, or price −8% from entry
- RiskStop +6%2% of book per coin; skip events inside 48h
The signal
XRP · mentions vs 7-day average · sample| Hour (KST) | Mentions × |
|---|---|
| Nov 16, 00:00 | 1.26 × |
| Nov 16, 01:00 | 1.44 × |
| Nov 16, 02:00 | 1.41 × |
| Nov 16, 03:00 | 1.43 × |
| Nov 16, 04:00 | 1.57 × |
| Nov 16, 05:00 | 1.53 × |
| Nov 16, 06:00 | 1.34 × |
| Nov 16, 07:00 | 1.31 × |
| Nov 16, 08:00 | 1.38 × |
| Nov 16, 09:00 | 1.29 × |
| Nov 16, 10:00 | 1.2 × |
| Nov 16, 11:00 | 1.35 × |
| Nov 16, 12:00 | 1.5 × |
| Nov 16, 13:00 | 1.46 × |
| Nov 16, 14:00 | 1.47 × |
| Nov 16, 15:00 | 1.61 × |
| Nov 16, 16:00 | 1.6 × |
| Nov 16, 17:00 | 1.4 × |
| Nov 16, 18:00 | 1.31 × |
| Nov 16, 19:00 | 1.32 × |
| Nov 16, 20:00 | 3.95 × |
| Nov 16, 21:00 | 3.47 × |
| Nov 16, 22:00 | 3.24 × |
| Nov 16, 23:00 | 3.1 × |
| Nov 17, 00:00 | 2.81 × |
| Nov 17, 01:00 | 2.61 × |
| Nov 17, 02:00 | 2.6 × |
| Nov 17, 03:00 | 2.51 × |
| Nov 17, 04:00 | 2.23 × |
| Nov 17, 05:00 | 2.06 × |
| Nov 17, 06:00 | 1.99 × |
| Nov 17, 07:00 | 1.75 × |
| Nov 17, 08:00 | 1.44 × |
| Nov 17, 09:00 | 1.33 × |
| Nov 17, 10:00 | 1.32 × |
| Nov 17, 11:00 | 1.16 × |
| Nov 17, 12:00 | 1.07 × |
| Nov 17, 13:00 | 1.2 × |
| Nov 17, 14:00 | 1.3 × |
| Nov 17, 15:00 | 1.22 × |
| Nov 17, 16:00 | 1.22 × |
| Nov 17, 17:00 | 1.34 × |
| Nov 17, 18:00 | 1.29 × |
| Nov 17, 19:00 | 1.09 × |
| Nov 17, 20:00 | 1.06 × |
| Nov 17, 21:00 | 1.11 × |
| Nov 17, 22:00 | 0.99 × |
| Nov 17, 23:00 | 0.9 × |
Sample backtest
Sample · illustrative 120D · not real performance| Date (UTC) | Equity |
|---|---|
| Jul 21 | 0.19% |
| Jul 22 | 0.66% |
| Jul 23 | 0.93% |
| Jul 24 | 1.36% |
| Jul 25 | 1.47% |
| Jul 26 | 1.35% |
| Jul 27 | 1.6% |
| Jul 28 | 1.79% |
| Jul 29 | 2.21% |
| Jul 30 | 2.8% |
| Jul 31 | 2.91% |
| Aug 1 | 2.98% |
| Aug 2 | 2.9% |
| Aug 3 | 2.63% |
| Aug 4 | 2.83% |
| Aug 5 | 3% |
| Aug 6 | 3.07% |
| Aug 7 | 3.21% |
| Aug 8 | 2.82% |
| Aug 9 | 2.41% |
| Aug 10 | 2.22% |
| Aug 11 | 1.94% |
| Aug 12 | 2.09% |
| Aug 13 | 2.23% |
| Aug 14 | 2.04% |
| Aug 15 | 1.96% |
| Aug 16 | 1.61% |
| Aug 17 | 1.36% |
| Aug 18 | 1.61% |
| Aug 19 | 1.78% |
| Aug 20 | 2.15% |
| Aug 21 | 2.49% |
| Aug 22 | 2.37% |
| Aug 23 | 2.46% |
| Aug 24 | 2.56% |
| Aug 25 | 2.73% |
| Aug 26 | 3.38% |
| Aug 27 | 3.78% |
| Aug 28 | 4.02% |
| Aug 29 | 4.24% |
| Aug 30 | 4.01% |
| Aug 31 | 4.05% |
| Sep 1 | 4.29% |
| Sep 2 | 4.41% |
| Sep 3 | 4.79% |
| Sep 4 | 4.8% |
| Sep 5 | 4.46% |
| Sep 6 | 4.28% |
| Sep 7 | 3.89% |
| Sep 8 | 3.77% |
| Sep 9 | 3.99% |
| Sep 10 | 3.89% |
| Sep 11 | 3.86% |
| Sep 12 | 3.63% |
| Sep 13 | 3.11% |
| Sep 14 | 3.02% |
| Sep 15 | 3.01% |
| Sep 16 | 3.14% |
| Sep 17 | 3.57% |
| Sep 18 | 3.58% |
| Sep 19 | 3.55% |
| Sep 20 | 3.59% |
| Sep 21 | 3.47% |
| Sep 22 | 3.86% |
| Sep 23 | 4.38% |
| Sep 24 | 4.71% |
| Sep 25 | 5.17% |
| Sep 26 | 5.19% |
| Sep 27 | 5.11% |
| Sep 28 | 5.33% |
| Sep 29 | 5.44% |
| Sep 30 | 5.84% |
| Oct 1 | 6.26% |
| Oct 2 | 6.19% |
| Oct 3 | 6.15% |
| Oct 4 | 5.88% |
| Oct 5 | 5.54% |
| Oct 6 | 5.68% |
| Oct 7 | 5.69% |
| Oct 8 | 5.71% |
| Oct 9 | 5.74% |
| Oct 10 | 5.25% |
| Oct 11 | 4.91% |
| Oct 12 | 4.72% |
| Oct 13 | 4.52% |
| Oct 14 | 4.82% |
| Oct 15 | 4.98% |
| Oct 16 | 4.9% |
| Oct 17 | 4.94% |
| Oct 18 | 4.66% |
| Oct 19 | 4.64% |
| Oct 20 | 5.03% |
| Oct 21 | 5.31% |
| Oct 22 | 5.83% |
| Oct 23 | 6.16% |
| Oct 24 | 6.09% |
| Oct 25 | 6.25% |
| Oct 26 | 6.32% |
| Oct 27 | 6.55% |
| Oct 28 | 7.16% |
| Oct 29 | 7.42% |
| Oct 30 | 7.59% |
| Oct 31 | 7.62% |
| Nov 1 | 7.26% |
| Nov 2 | 7.24% |
| Nov 3 | 7.33% |
| Nov 4 | 7.35% |
| Nov 5 | 7.62% |
| Nov 6 | 7.43% |
| Nov 7 | 7.03% |
| Nov 8 | 6.78% |
| Nov 9 | 6.35% |
| Nov 10 | 6.34% |
| Nov 11 | 6.54% |
| Nov 12 | 6.48% |
| Nov 13 | 6.54% |
| Nov 14 | 6.32% |
| Nov 15 | 5.95% |
| Nov 16 | 6.04% |
| Nov 17 | 6.14% |
Generated to show the shape of the strategy, not measured on real fills. Ask your agent to backtest on DataMaxi+ history before trading.
Backtest this strategy
One prompt: your agent pulls the history below, tests these rules with fees and slippage, and charts the equity curve and drawdown. Run it before you trust the sample above.
- Period
- As far back as naver_trend returns, daily
- History calls
- 20
- Credits
- 100
- Cost per run
- ≈ $0.01
Backtest the "Retail Sentiment Contrarian" strategy on DataMaxi+ history. Thesis: When Naver searches and Telegram chatter triple on a coin that already ran, retail is late — fade the frenzy. Use the DataMaxi+ MCP server at https://mcp.datamaxiplus.com/mcp. Pull the history once, cache it, and run the backtest in your own code sandbox (Python is fine). ## History to pull (split the period with from/to; never refetch a range you have) 1. naver_trend(symbol="XRP") — Naver search interest per coin; about 10 calls 2. cex_candle(exchange="upbit", market="spot", symbol="XRP-KRW", interval="1d") — OHLCV, up to 1,000 candles per call, history from Jan 2024; about 10 calls Universe: the 10 most searched coins on Naver, Upbit KRW price, Binance perp for the short Period: as far back as naver_trend returns, daily ## Rules to test Signal: Naver search interest and Telegram mentions vs their 7-day average (× multiple), with KRW premium and 24h price change. Entry: Mentions ≥ 3× the 7-day average, price +10% in 24h and KRW premium ≥ 2%: short the global perp. Exit: Mentions back under 1.5×, or price −8% from entry. Risk: Stop +6%. Max 2% of book per coin. Skip coins with a listing or unlock event inside 48h. ## Assumptions Costs: 0.05% perp taker on entry and exit, 0.1% slippage Data: Telegram mentions and the KRW premium have no history here, so test on Naver search and price alone and say so. Use only data stamped before each decision (no look-ahead). Say plainly where the data cannot test a rule. ## Report Net return, annualised return, Sharpe, max drawdown, number of trades, win rate and average hold, next to buy-and-hold over the same period. Chart the equity curve and the drawdown, and list the 5 worst trades. ## Budget About 20 history calls ≈ 100 credits ≈ $0.01 from my prepaid DataMaxi+ balance. Tell me how many calls you made. If DataMaxi+ is not connected, ask me to [connect DataMaxi+](https://claude.ai/customize/connectors?modal=add-custom-connector) using https://mcp.datamaxiplus.com/mcp before continuing.
Estimate from the call counts in the prompt. Your agent may pull less or more; it reports the calls it made.
Rules
| Signal | Naver search interest and Telegram mentions vs their 7-day average (× multiple), with KRW premium and 24h price change. |
|---|---|
| Entry | Mentions ≥ 3× the 7-day average, price +10% in 24h and KRW premium ≥ 2%: short the global perp. |
| Exit | Mentions back under 1.5×, or price −8% from entry. |
| Risk | Stop +6%. Max 2% of book per coin. Skip coins with a listing or unlock event inside 48h. |
Data used
≈ $0.528 / day| MCP tool · REST route | Dataset | Tier | Calls / h | Credits / day | USD / day |
|---|---|---|---|---|---|
naver_trend GET /naver-trend | Trend | History · 5 cr | 20 | 2,400 | $0.24 |
telegram_messages GET /telegram/messages | Trend | History · 5 cr | 20 | 2,400 | $0.24 |
ticker GET /ticker | Candle · ticker | Latest · 1 cr | 10 | 240 | $0.024 |
premium GET /premium | Premium | Latest · 1 cr | 10 | 240 | $0.024 |
| Total | 60 | 5,280 | $0.528 |
1 credit = $0.0001. Only successful calls are billed, from one balance shared by the API and MCP. Calls per hour are the suggested cadence; scale them to the symbols you watch.
Build it with your agent
Connect the MCP server →Open it in Claude with one click (it asks you to add the DataMaxi+ connector first if you haven't), or paste it into Cursor or any MCP client. It backtests first, then runs in paper mode and never places orders on its own.
Build the "Retail Sentiment Contrarian" strategy for me with DataMaxi+ data. Thesis: When Naver searches and Telegram chatter triple on a coin that already ran, retail is late — fade the frenzy. Connect to the DataMaxi+ MCP server at https://mcp.datamaxiplus.com/mcp (header X-DTMX-APIKEY, read from env DATAMAXI_API_KEY). ## Data (call these tools; arguments are examples — loop over the symbols and venues you need) 1. naver_trend(symbol="XRP") — Search interest per coin; about 20×/hour 2. telegram_messages(search_query="XRP", category="korean", limit="100") — Channel mentions; about 20×/hour 3. ticker(exchange="upbit", symbol="XRP-KRW", currency="KRW") — 24h price change; about 10×/hour 4. premium(asset="XRP", source_exchange="binance", target_exchange="upbit") — Is the KRW side overheating?; about 10×/hour ## Rules Signal: Naver search interest and Telegram mentions vs their 7-day average (× multiple), with KRW premium and 24h price change. Entry: Mentions ≥ 3× the 7-day average, price +10% in 24h and KRW premium ≥ 2%: short the global perp. Exit: Mentions back under 1.5×, or price −8% from entry. Risk: Stop +6%. Max 2% of book per coin. Skip coins with a listing or unlock event inside 48h. ## What to do 1. Build the signal from the tools above and backtest the rules on the history you can pull; report trades, win rate, average trade and max drawdown. 2. Then run it live in paper mode and show me coins at ≥ 3× mentions with price change, premium and a fade decision, refreshed on the cadence above. 3. Never place real orders. Propose trades and wait for my confirmation. ## Budget Every call is billed from my prepaid DataMaxi+ balance. Expected load: ~60 calls/hour ≈ 5,280 credits/day (≈ $0.53/day). Cache results, never repeat a call inside its refresh window, and print the number of calls each refresh made. If DataMaxi+ is not connected, ask me to [connect DataMaxi+](https://claude.ai/customize/connectors?modal=add-custom-connector) using https://mcp.datamaxiplus.com/mcp before continuing.
Risk notes
- Frenzies can run much further than seems reasonable; the stop has to be honoured.
- Short squeezes on hyped coins are violent — size small.
- Mentions spike on news too (hacks, delistings); read the messages before trading.
Educational example, not investment advice. Sample charts and figures are illustrative. See the live screen it builds on: Retail Trend Signals.
Keep exploring
Live screens, dashboards and the data behind this strategy.