Smart Money Tracking Explained: Why Win-Rate Matters
Not every wallet is alpha. We curate to >65% historical win-rate over 100 trades. Here's how that filter dramatically changes the signal-to-noise ratio.
Smart-money tracking became a feature of every crypto dashboard in 2024 — usually badly. This article explains the win-rate filter we apply to the Pulsentric wallet list, why it matters, and what we trade away to keep it tight.
The naive approach
The naive version of smart-money tracking is: take a list of "famous" wallets (whales from Etherscan rich list, named DeFi traders, foundation multisigs) and surface every trade they make. The signal looks great in a screenshot. In practice it's noise: a foundation rebalance is not alpha, and a known whale exiting a position is the opposite of a buy signal.
Win-rate as a hard filter
We compute win-rate per wallet over the last 100 closed positions, using realized PnL on token sells. A position counts as a win if exit price > entry price net of gas. The 65% cutoff was chosen empirically: above 65% the wallet's buy signal correlates with future price action; below 65% it's noise.
Why not 80% or 90%?
Two reasons. First, sample size: very few wallets have 100+ closed positions and 80%+ win-rate at the same time, and those that do are often arbitrage bots that don't hold long enough for retail to follow. Second, recall: we'd rather track 100 wallets at 65% than 12 wallets at 80% — diversification of the signal itself reduces wallet-specific drift.
Rotation
Wallets that fall below 65% for 30 consecutive days get rotated out. Win-rate isn't a one-shot vote — markets change, traders take breaks, strategies decay. The list is dynamic and surveillance-driven.
What this enables
When a wallet from the curated list buys a token *and* sentiment for that token from Telegram/X is >0.7, we generate an ELITE SNIPER signal. The cross-reference is what makes the alert actionable: smart-money buy alone is interesting; smart-money buy + retail sentiment turning bullish is a setup.
ELITE SNIPER alerts dispatch in <30 seconds (Alchemy webhook → Redis queue → QStash → Telegram + email + in-app) on Pro plans only. The methodology stays public. The wallet list does not.
Related posts
Telegram Alpha Channels: Signal vs Noise
Most Telegram alpha is paid shilling, recycled CT screenshots, or pure FOMO. Here's how to tell signal from noise — and how AI sentiment helps.
How to Spot a Rug Pull in 60 Seconds (2026 Guide)
A no-fluff checklist to vet any token in under a minute: liquidity lock, holder concentration, ownership, honeypot simulation. With real on-chain examples.
The 7 Red Flags of a Memecoin Honeypot
Honeypots are getting smarter. Here are the 7 telltale patterns that show up across the latest generation of trap contracts on BSC, Base and Solana.