DEV Community

Gunnar Thorderson
Gunnar Thorderson

Posted on • Originally published at getregime.com

How to Add Market Regime Filtering to Any Crypto Trading Strategy

#strategy #regimefilter #tutorial #tradingbot

How to Add Market Regime Filtering to Any Crypto Trading Strategy

Every backtest looks great until you run it live through a choppy market. The strategy that returned 40% in a trending quarter suddenly bleeds 15-30% of those gains when the regime shifts to sideways price action. This is the single most common failure mode in algorithmic crypto trading, and the fix is surprisingly simple.

The Problem: Strategies Are Regime-Dependent

Most trading strategies — momentum, mean-reversion, breakout, crossover — are implicitly designed for a specific market condition. A trend-following system prints money during sustained bull or bear moves. But when the market enters a chop regime, that same system generates false signal after false signal, each one taking a small bite out of your capital.

The math is unforgiving. If your strategy gains 25% during a trending phase and then loses 18% during the subsequent chop, your net return is barely 2.5%. Multiply that across several regime transitions per year and you understand why most live strategies underperform their backtests.

The root cause is not the strategy itself. It is the absence of a regime filter — a pre-trade check that asks: "Is this the right environment for my strategy to operate in?"

What Regime Filtering Actually Means

Regime filtering is a gate that sits before your entry logic. Before your strategy evaluates any signal, it checks the current market regime classification:

  • Bull: Sustained uptrend with confirming signals across momentum, funding, and macro indicators.
  • Bear: Sustained downtrend with risk-off positioning and negative momentum.
  • Chop: No clear directional bias. Price oscillates within a range. Most signals are noise.

The filter logic is straightforward. During a bull regime, allow long entries. During a bear regime, allow shorts or hedges. During chop, skip all entries and hold cash. This single gate eliminates the majority of false signals that cause drawdowns in regime transitions.

You do not need to build a regime detector from scratch. That requires aggregating dozens of on-chain, derivatives, and macro signals, weighting them correctly, and validating against historical regime transitions. Instead, you can call an API.

The Regime API Call

The Regime API provides a real-time market regime classification based on 10+ signals including funding rates, open interest shifts, order book imbalance, macro indicators (DXY, VIX), and on-chain flows.

Free endpoint (no authentication required):

GET https://getregime.com/api/v1/market/regime
Enter fullscreen mode Exit fullscreen mode

Example response:

{
  "regime": "BULL",
  "confidence": 0.74,
  "signals": {
    "momentum": "bullish",
    "funding": "neutral",
    "openInterest": "rising",
    "macro": "risk-on",
    "fearGreed": 62,
    "dxyTrend": "weakening"
  },
  "timestamp": "2026-04-20T14:30:00Z"
}
Enter fullscreen mode Exit fullscreen mode

The regime field gives you the classification. The confidence field (0 to 1) tells you how strong the signal is. The signals object breaks down the individual components so you can see what is driving the classification.

For Pro users, the intelligence brief endpoint provides a full synthesis including crowd positioning, divergences, and risk assessment:

GET https://getregime.com/api/v1/intelligence/brief
Authorization: Bearer YOUR_API_KEY
Enter fullscreen mode Exit fullscreen mode

Implementation in Three Languages

Python

The most common setup for quant strategies. Add this check before any entry decision:

import requests

def get_regime():
    r = requests.get("https://getregime.com/api/v1/market/regime")
    return r.json()

def should_enter_long():
    regime = get_regime()
    if regime["regime"] == "CHOP":
        return False
    if regime["regime"] == "BEAR":
        return False
    if regime["confidence"] < 0.4:
        return False  # low conviction — stay flat
    return True

# In your strategy loop:
if signal_detected and should_enter_long():
    execute_trade()
Enter fullscreen mode Exit fullscreen mode

JavaScript / Node.js

For bot builders and dashboard integrations:

async function getRegime() {
  const res = await fetch("https://getregime.com/api/v1/market/regime");
  return res.json();
}

async function shouldTrade(direction) {
  const { regime, confidence } = await getRegime();

  if (regime === "CHOP") return false;
  if (regime === "BEAR" && direction === "long") return false;
  if (regime === "BULL" && direction === "short") return false;
  if (confidence < 0.4) return false;

  return true;
}

// In your bot logic:
if (await shouldTrade("long")) {
  placeLongOrder();
}
Enter fullscreen mode Exit fullscreen mode

Using the npm SDK

If you prefer a typed client with built-in retries:

import { Regime } from "getregime";

const client = new Regime(); // free tier, no key needed

const { regime, confidence } = await client.market.regime();

if (regime === "BULL" && confidence > 0.6) {
  // proceed with long entry
}
Enter fullscreen mode Exit fullscreen mode

Install with npm install getregime. Full SDK docs at getregime.com/quickstart.

Strategy Rules: A Simple Decision Matrix

Here is a concrete rule set you can adapt to any strategy:

Regime Confidence Action
BULL > 60% Take long entries normally
BULL 40-60% Reduce position size by 50%
BEAR > 60% Only short or hedge existing longs
BEAR 40-60% Reduce short size by 50%
CHOP Any Skip all entries, hold cash
Any < 40% Reduce position size by 50% or sit out

The confidence threshold matters. A BULL regime with 45% confidence is not the same as one with 80% confidence. The former might be an early or fading trend — worth participating in, but with reduced size. The latter is a high-conviction environment where your trend-following strategy should perform well.

Real Results: What the Data Shows

We have tracked 148+ regime transitions since the system went live. The data is publicly available at getregime.com/track-record with timestamped snapshots proving accuracy.

Key findings from the dataset:

  • Regime shifts typically lead price by 24-48 hours. The classification picks up on funding rate shifts, order book changes, and macro moves before they manifest as price action.
  • Strategies using regime filtering avoided an average of 60-70% of false signals during chop periods compared to unfiltered versions.
  • The CHOP classification, specifically, is the highest-value signal. Knowing when not to trade is worth more than knowing when to trade.

The regime detector aggregates data from Binance, Hyperliquid, CoinGecko, DeFiLlama, and FRED (Federal Reserve Economic Data). It is not a single-indicator system. It cross-references derivatives positioning, on-chain flows, and traditional macro to produce a composite classification.

Get Started in 30 Seconds

The regime endpoint is free and requires no authentication. Test it right now:

curl https://getregime.com/api/v1/market/regime
Enter fullscreen mode Exit fullscreen mode

That single call gives you the current market classification. Integrate it into your strategy as a pre-trade gate, and you have regime filtering running in production.

For the full interactive playground, API documentation, and code examples in additional languages, visit getregime.com/quickstart.

The best trade is often no trade at all. Let the regime tell you when to sit on your hands.

Try Regime Intelligence

Regime is a real-time crypto market regime detection API. One endpoint tells you if the market is bull, bear, or chop — so your bot only trades when conditions match your strategy.

Free API access → | See pricing → | API docs →

Top comments (0)