Technical / Structural Regime: The Universal Overlay for Crypto Trading Bots

Why Every Crypto Trading Regime Needs a Technical / Structural Overlay

Most trading strategies fire on a state — an oversold RSI, a 200-day moving average reclaim, a Bollinger band squeeze about to break. None of those are regimes on their own; they are structural overlays that pair with whatever regime you're already in.

A 200MA reclaim during a confirmed bull regime is a high-conviction long. The same reclaim during a liquidation cascade is a bull trap. The price-structure signal is identical; the context is what makes it tradable.

The new /api/v1/indicators/technical endpoint exposes the four canonical price-structure indicators — moving averages, Bollinger squeezes, range positions, and RSI extremes — for every asset in the Hyperliquid perps universe plus the top Binance USDT spot pairs. One call, ~630 symbols, computed every 30 minutes. It's Regime #14 in our regime taxonomy: the universal overlay.

What the Technical Regime Endpoint Returns

Each asset in the response carries four indicator blocks — moving-average state, Bollinger compression state, 20-day range state, and RSI extremes — plus current price and source (binance_spot or hyperliquid_perp).

curl -H "X-API-Key: cdk_live_yourkey" \
  "https://cryptodataapi.com/api/v1/indicators/technical?bb=in_squeeze&sort=squeeze_days&order=desc&limit=5"

One symbol in the response:

{
  "symbol": "KLAY",
  "source": "binance_spot",
  "price": 0.1255,
  "ma": {
    "sma_200": 0.155928,
    "above_200": false,
    "dist_from_200_pct": -19.51,
    "last_cross_200": null
  },
  "bollinger": {
    "bandwidth_pct": 11.09,
    "squeeze_percentile": 8.3,
    "in_squeeze": true,
    "days_in_squeeze": 14,
    "expanding": false
  },
  "range_state": {
    "low": 0.1173, "high": 0.1391,
    "position_pct": 37.6, "zone": "mid"
  },
  "rsi": {
    "rsi_14_1d": 48.2, "state_1d": "neutral",
    "rsi_14_1h": 71.8, "state_1h": "overbought",
    "extreme_duration_bars": 0
  }
}

That single record tells you KLAY is below its 200-day MA by 19.5%, has been in the tightest 10% of its 90-day Bollinger bandwidth for 14 consecutive days, sits mid-range on the daily, and is overbought on the 1-hour. A volatility breakout is structurally overdue.

The Four Price-Structure Signals, Decoded

Each block answers a different structural question. Combine them or read them separately — the screener supports independent filters per indicator.

• Moving averages. sma_50, sma_100, sma_200 plus above_X booleans and dist_from_X_pct. The last_cross_200 field reports the most recent SMA-50 / SMA-200 crossover direction (golden or death) with days_since_cross_200, so an agent can detect a fresh reclaim or breakdown in one read.
• Bollinger compression. 20-period, 2-σ bands. bandwidth_pct is the current width as a percent of mid. squeeze_percentile ranks today's width in its trailing 90-day distribution — a percentile of 8.3 means today's bandwidth is tighter than 91.7% of the last 90 days. in_squeeze fires when that percentile is ≤ 20. days_in_squeeze counts consecutive bars in squeeze; expanding flips true when bandwidth rises above its value 3 bars ago.
• Range state. Rolling 20-day high/low. position_pct places current price on a 0-100 scale within the range. zone resolves to near_low, mid, or near_high using a 15% edge buffer — useful for systematic range fades.
• RSI extremes. Wilder RSI(14) on daily and 1h closes. state_1d classifies as oversold (≤30), neutral, or overbought (≥70). extreme_1d fires only on RSI > 80 or < 20. extreme_duration_bars counts consecutive recent bars sitting in the extreme zone — the longer the streak, the higher the mean-reversion probability.

How Do I Screen for Specific Technical Setups via the API?

The endpoint accepts independent filters per indicator. Stack them to build high-conviction screens.

Available sort keys: squeeze_days, rsi_1d, bb_bandwidth, dist_from_200, position_in_range, symbol, price. Default limit is 250, max 500. The full screener is computed off a 30-minute TTL cache, so repeated queries are essentially free.

Technical Regime vs SIGNUM_RGG: Which Indicator When?

Crypto Data API exposes two computed trend layers. They answer different questions and the right approach is to read both.

Practical pattern: filter your universe with SIGNUM, time entries with Technical. A green-trend asset (SIGNUM) entering a Bollinger squeeze with RSI exiting oversold (Technical) is a higher-quality long than either signal alone.

Wiring the Technical Regime Into an AI Trading Agent

Two reads is usually enough — pull the universe-wide screener for candidate scanning, then pull the per-symbol detail for any candidates you want to act on.

import httpx

HDR = {"X-API-Key": "cdk_live_yourkey"}
BASE = "https://cryptodataapi.com/api/v1/indicators/technical"

# 1. Find assets in a sustained Bollinger squeeze
squeezed = httpx.get(
    BASE, headers=HDR,
    params={"bb": "in_squeeze", "sort": "squeeze_days",
            "order": "desc", "limit": 20},
).json()["items"]

# 2. For each, pull detail with 60d history
for r in squeezed[:5]:
    detail = httpx.get(f"{BASE}/{r['symbol']}", headers=HDR).json()
    print(r['symbol'], detail['bollinger']['days_in_squeeze'],
          'in_squeeze', '|', detail['rsi']['state_1d'])

For backtests, the same summary is included in every archived daily snapshot under technical_regime — 90 days of history is already backfilled, so you can replay regime-tagged simulations without computing indicators yourself.

curl -H "X-API-Key: cdk_live_yourkey" \
  "https://cryptodataapi.com/api/v1/backtesting/daily-snapshots/2026-05-15" \
  | jq '.snapshot.technical_regime'

{
  "total": 629,
  "above_200ma": 93,
  "in_squeeze": 85,
  "overbought_1d": 13,
  "oversold_1d": 86,
  "top_in_squeeze": ["KLAY", "FET", "..."],
  "backfilled": true
}

When to Reach for the Technical Regime

This endpoint earns its keep in three concrete scenarios:

• Entry timing inside a known regime. SIGNUM tells you it's a bull. Technical tells you whether to chase the breakout or wait for a range fade. Stack the two and you cut bad-entry slippage.
• Volatility-breakout pre-positioning. Filter bb=in_squeeze&sort=squeeze_days daily. The longer the squeeze, the more asymmetric the upcoming expansion. Pair with funding-rate extremes for direction bias.
• Mean-reversion fades during chop. When SIGNUM shows grey across the majors, range_zone=near_high&rsi=overbought surfaces fade candidates with the highest hit rate. Avoid these filters during confirmed trends — fades into trending markets are how accounts die.

What this endpoint isn't: a standalone alpha signal. Technical setups fire constantly. Treating every BB squeeze as a buy is how you over-trade. Always pair with regime context — that's why we built the full regime taxonomy.