Master the on-chain metrics that historically mark Bitcoin bottoms — MVRV, SOPR, NUPL and more — plus Python code to automate your bottom-detection signal chain.
Every Bitcoin cycle has a bottom. The problem isn't that bottoms don't exist — it's that they only look obvious in hindsight. By the time CNBC runs the 'crypto is dead' segment, smart money has already been accumulating for weeks. On-chain data is how you see that accumulation happening in real time, before the price confirms it.
Bitcoin's blockchain is a public ledger. Every transaction, every wallet movement, every coin that hasn't moved in years — it's all there, readable by anyone with the right tools. The traders who consistently buy cycle bottoms aren't psychic. They're reading bitcoin bottom on chain signals that the broader market ignores.
Before diving into specific metrics, it helps to understand what a signal chain actually is in this context. What is a signal chain? In trading, a signal chain is the pipeline that transforms raw data into actionable decisions: raw blockchain data → processed metrics → signal conditions → trade execution. Each layer filters noise and adds meaning. On-chain analysis operates at the very first layer of that pipeline — the source of truth.
Unlike price charts, which show you what the market decided, on-chain data shows you what individual participants are actually doing with their coins. Are long-term holders selling? Are miners capitulating? Are coins flowing onto exchanges? These behaviors tell a fundamentally different story than candlesticks. A falling price with coins leaving exchanges to cold wallets is a completely different situation than a falling price with coins flooding in.
Traders sometimes debate analysis methodologies the way others compare ball chain vs cable chain — debating the format when what actually matters is whether the tool produces edge in practice. Technical analysis and on-chain analysis aren't mutually exclusive. The strongest bottom signals in Bitcoin history have happened when both converge: on-chain capitulation metrics flashing buy zones while price action forms reversal structures.
On-chain metrics are most powerful at market extremes — cycle bottoms and tops. During trending markets they produce more noise. Use them as a macro context filter, not a standalone entry trigger.
There are dozens of on-chain metrics but a handful have shown consistent reliability across multiple Bitcoin cycles. These are the ones worth building your signal chain around — each one capturing a different dimension of market psychology and participant behavior.
No single metric is sufficient. MVRV might flash a buy signal months before the actual bottom. SOPR can give premature recovery signals. The power comes from convergence — when four or five of these metrics align simultaneously, the probability of being near a major accumulation zone increases substantially.
Glassnode is the industry-standard provider for professional on-chain data. Their API covers every major metric discussed here. The following module wraps the most important bottom indicators into a single reusable interface:
import requests
GLASSNODE_API_KEY = 'your_api_key_here'
BASE_URL = 'https://api.glassnode.com/v1/metrics'
def fetch_metric(endpoint, asset='BTC', interval='24h'):
url = f'{BASE_URL}/{endpoint}'
params = {
'a': asset,
'api_key': GLASSNODE_API_KEY,
'f': 'JSON',
'i': interval
}
response = requests.get(url, params=params)
response.raise_for_status()
data = response.json()
return data[-1]['v'] if data else None
def get_mvrv_zscore():
return fetch_metric('market/mvrv_z_score')
def get_sopr():
return fetch_metric('indicators/sopr')
def get_nupl():
return fetch_metric('indicators/nupl')
def get_puell_multiple():
return fetch_metric('mining/puell_multiple')
if __name__ == '__main__':
mvrv = get_mvrv_zscore()
sopr = get_sopr()
nupl = get_nupl()
puell = get_puell_multiple()
print(f'MVRV Z-Score: {mvrv:.2f}')
print(f'SOPR: {sopr:.4f}')
print(f'NUPL: {nupl:.4f}')
print(f'Puell Multiple: {puell:.2f}')Raw metric values are only useful if you have a framework for interpreting them together. The following scorer assigns weighted points to each metric based on historical bottom thresholds. A score of 7 or above has historically correlated with major Bitcoin accumulation zones across the 2015, 2018, 2020, and 2022 cycle lows:
def calculate_bottom_confluence(mvrv_z, sopr, nupl, puell):
score = 0
signals = []
# MVRV Z-Score: below 0 is the historically strongest buy zone
if mvrv_z < 0:
score += 3
signals.append(f'MVRV Z-Score {mvrv_z:.2f} — deep capitulation zone')
elif mvrv_z < 1:
score += 1
signals.append(f'MVRV Z-Score {mvrv_z:.2f} — below fair value')
# SOPR below 1 means coins are moving at a realized loss
if sopr < 0.97:
score += 3
signals.append(f'SOPR {sopr:.4f} — capitulation confirmed')
elif sopr < 1.0:
score += 1
signals.append(f'SOPR {sopr:.4f} — borderline capitulation')
# NUPL negative means aggregate unrealized loss across all holders
if nupl < 0:
score += 2
signals.append(f'NUPL {nupl:.4f} — market in net loss')
# Puell below 0.5 signals miners selling reserves to survive
if puell < 0.5:
score += 2
signals.append(f'Puell {puell:.2f} — miner capitulation zone')
elif puell < 0.8:
score += 1
signals.append(f'Puell {puell:.2f} — miners under pressure')
if score >= 7:
verdict = 'STRONG BUY ZONE'
elif score >= 4:
verdict = 'WATCH — accumulation possible'
else:
verdict = 'NO SIGNAL'
return {'score': score, 'max': 10, 'verdict': verdict, 'signals': signals}
# Example: metrics resembling the November 2022 bottom
result = calculate_bottom_confluence(
mvrv_z=-0.31,
sopr=0.963,
nupl=-0.14,
puell=0.41
)
score_val = result['score']
verdict_val = result['verdict']
print(f'Confluence Score: {score_val}/10 — {verdict_val}')
for sig in result['signals']:
print(f' + {sig}')On-chain signals give you the macro picture. Exchange data tells you when the market is actually starting to move. The most effective bottom-detection setups combine both layers. On Binance and Bybit you can monitor funding rates and open interest via API — deeply negative funding combined with on-chain bottom signals has preceded some of the sharpest BTC recoveries on record. When longs are being liquidated and on-chain says the market is in capitulation, that is the setup.
Platforms like OKX and Coinbase provide institutional-grade order flow data. When large block trades start appearing on Coinbase Advanced at the same time MVRV and SOPR enter capitulation territory, that is the convergence every accumulator is waiting for. Bitget and Gate.io are also worth monitoring for perpetual futures liquidation heatmaps — mass liquidation cascades on leveraged longs often mark the final exhaustion move before a meaningful bottom.
import ccxt
import time
# Connect to Binance for real-time price and funding rate data
exchange = ccxt.binance({
'apiKey': 'your_api_key',
'secret': 'your_secret',
'enableRateLimit': True
})
def get_funding_rate():
info = exchange.fetch_funding_rate('BTC/USDT:USDT')
return info['fundingRate']
def get_btc_price():
ticker = exchange.fetch_ticker('BTC/USDT')
return ticker['last']
def run_bottom_detector(interval_hours=6):
print('Bitcoin bottom detector running...')
while True:
try:
mvrv = get_mvrv_zscore()
sopr = get_sopr()
nupl = get_nupl()
puell = get_puell_multiple()
funding = get_funding_rate()
price = get_btc_price()
result = calculate_bottom_confluence(mvrv, sopr, nupl, puell)
# Negative funding = longs being punished = forced selling
funding_note = ' + negative funding' if funding < -0.0005 else ''
ts = time.strftime('%Y-%m-%d %H:%M')
score_val = result['score']
verdict_val = result['verdict']
print(f'[{ts}] BTC ${price:,.0f} | Score {score_val}/10 — {verdict_val}{funding_note}')
if result['score'] >= 7:
# Plug in your alert here: Telegram bot, email, VoiceOfChain webhook
print('>>> BOTTOM SIGNAL ACTIVE — review position sizing and consider scaling in')
time.sleep(interval_hours * 3600)
except Exception as err:
print(f'Error: {err}')
time.sleep(300)
run_bottom_detector()For traders who don't want to maintain their own data infrastructure, VoiceOfChain aggregates on-chain signals and exchange data into a unified real-time feed. Instead of polling Glassnode and Binance separately, you receive consolidated alerts the moment multiple bottom indicators align simultaneously — including the exact metric values driving the signal and the overall confluence score.
The biggest mistake is treating a single metric entering buy territory as a bottom confirmation. Bitcoin spent most of 2014 with MVRV in capitulation territory. Being early and being right are two very different risk positions — especially if you're using leverage.
The NVT Signal (Network Value to Transactions ratio) is worth monitoring alongside the core metrics. When NVT is low while price is falling, the network is processing significant economic activity despite declining price — a sign of genuine utility and a supporting bottom indicator.
Bitcoin bottom on chain signals don't give you certainty — nothing in trading does. What they give you is a probabilistic edge grounded in actual participant behavior rather than candlestick patterns or sentiment surveys. When MVRV goes negative, SOPR confirms capitulation, NUPL turns red, and the Puell Multiple signals miner distress simultaneously, history strongly suggests you're in an accumulation window worth taking seriously.
The Python tools above give you the foundation to build your own monitoring pipeline. Pair that with exchange data from Binance and OKX — funding rates, liquidation data, order flow — and you have a genuinely comprehensive bottom-detection framework. For traders who want this analysis without the infrastructure overhead, VoiceOfChain delivers real-time on-chain signal alerts the moment confluence builds, so you can focus on managing positions rather than wrangling APIs.