May 1, 202612 min read

gopfan2 Polymarket: The $343K Tail-Buying Strategy Explained

If ColdMath is Polymarket weather’s most famous story, gopfan2 is its most profitable. Over $343,000 in net profit at the top of the all-time leaderboard — generated with a strategy described as almost absurdly simple: buy below $0.15, sell NO above $0.45, risk approximately $1 per position, repeat thousands of times.

Who Is gopfan2?

gopfan2 is a Polymarket wallet address at the top of the public weather leaderboard. The wallet’s trade history is visible on-chain: tens of thousands of individual positions across Polymarket weather markets, with cumulative net profit exceeding $343,000 as of early 2026.

A related wallet, gopfan (without the “2”), sits at #4 on the same leaderboard with approximately $118,000 in profit — suggesting either the same operator with multiple wallets, or related traders using similar strategies.

Community analysis — primarily from Medium posts studying the on-chain data — attributes the strategy to a systematic price-threshold approach. The $2 million figure sometimes associated with gopfan2 appears to be a misattribution or cross-wallet confusion. The verifiable weather-specific leaderboard figure is $343,000+.

The Strategy: What “Buy Below $0.15” Actually Means

A temperature bucket priced at $0.12 means the market believes there is a 12% probability the daily high falls in that range. Buying YES at $0.12 is a bet that the true probability is higher than 12%.

The gopfan2 strategy says: systematically buy any bucket priced below $0.15. The implied claim is that temperature buckets priced in the $0.05–$0.15 range are systematically underpriced — their true probability is higher than the market reflects.

Is This Claim True?

The favorite-longshot bias is a well-documented phenomenon across prediction markets. In parts of the Polymarket temperature market ecosystem, low-probability tail buckets appear to be underpriced — the market systematically assigns too low a probability to them. Why?

Retail concentration on modal outcomes. When a forecast shows 68°F, retail traders pile into the 67–68°F and 68–69°F buckets. Adjacent buckets get far less attention, even though they represent non-trivial probability mass given forecast uncertainty. A strategy that systematically buys underattended tails profits from this attention imbalance.
Forecast uncertainty is under-reflected. Retail traders often treat a 5-day forecast as close to definitive. In reality, 5-day surface temperature forecasts have standard deviations of 4–6°F — meaning the 71–72°F bucket has meaningful probability even when the forecast says 68°F.
Volume thins at extremes. Bucket prices near $0.05–$0.15 have thin order books. A single systematic buyer can hold a significant portion of total long interest at that price level.

The NO-Above-$0.45 Component

Buying NO on buckets priced above $0.45 is the mirror image. NO costs $0.45 and pays $1.00 if any other bucket wins. The strategy only works if high-priced buckets are genuinely overpriced — and this is more controversial than the tail-buying component. In highly liquid, well-contested markets, the modal bucket is usually fairly priced because it attracts the most algorithmic attention.

Why $1 Per Position?

The tiny position size is a deliberate architecture choice:

Capital efficiency at scale. $1 per trade × 10,000 trades = $10,000 total deployed. If the strategy wins 20% of its YES bets (returning $1.00) on average underpriced tails, and loses 80%:

EV = 0.20 × $0.88 − 0.80 × $0.12 = $0.176 − $0.096 = $0.08 per trade

$0.08 × 10,000 trades = $800 on $10,000 deployed. With daily resolution and fast capital turnover, the annualized return on deployed capital is extremely high.

Thin market access. Temperature bucket markets often have limited liquidity at tail prices. A $1 position can be established at the published price. A $100 position might move the market, worsening entry price and reducing edge.

Risk distribution. Thousands of $1 bets have lower variance than dozens of $100 bets with the same expected value. The law of large numbers applies more aggressively at the micro-bet level.

Does the Strategy Require a Forecast Model?

The publicly described version — price thresholds, no model — does not appear to require a calibrated meteorological forecast. The bet is purely that tail prices are systematically mispriced across weather markets in aggregate.

This is the strategy’s most attractive feature for non-technical traders. But it has limits.

The mispricing claim requires empirical validation. “Tails are underpriced” is a hypothesis, not a fact. The gopfan2 P&L is evidence it held for this wallet over its trading history. That doesn’t guarantee it holds for a new entrant today.

Without a model, you can’t assess edge size. The price-threshold approach treats all sub-$0.15 buckets equally. But some sub-$0.15 buckets are correctly priced at 10% probability; others might genuinely have 25% probability because the forecast is uncertain and retail isn’t paying attention. A hybrid approach — using a simple forecast model to filter for sub-$0.15 buckets where the model assigns at least 18–20% probability — likely produces better risk-adjusted returns.

What's Required to Execute This

The gopfan2 strategy at $343K in profit across thousands of trades was almost certainly automated. The characteristics — consistent small sizing across dozens of simultaneous markets, response at all hours, high trade frequency — are incompatible with manual execution.

At minimum, executing this requires a bot that:

Monitors all active Polymarket weather markets via the Gamma API.
Places $1 limit orders on all YES buckets priced below $0.15.
Simultaneously places NO buys on all YES buckets priced above $0.45.
Manages open orders, cancels stale ones, and runs 24 hours a day.

A basic version of this bot can be built in Python in a few days using the official Polymarket py-clob-client SDK. The core loop is ~100–150 lines of code.

The Strategy Today vs. 18 Months Ago

The gopfan2 strategy has been public knowledge in the prediction market community since at least mid-2025. A publicly known strategy attracts imitators, and imitators narrow the edge.

What’s likely changed: More bots are running price-threshold strategies on Polymarket weather buckets. Sub-$0.15 buckets in high-volume markets (Tokyo, NYC, London) are more frequently contested, compressing the edge per trade. The pure price-threshold approach without a forecast component is more likely to buy “correctly cheap” and “incorrectly cheap” buckets at roughly equal rates.

What hasn’t changed: Retail traders still concentrate on modal buckets and undervalue tail uncertainty. New weather markets open regularly, and new cities have initial periods of lower bot saturation. The fundamental mechanism is unlikely to disappear entirely.

The Lessons Worth Taking Forward

The specific rules (buy <$0.15, sell >$0.45, $1 per bet) may not be as durable as they were. But the principles they embody are:

Tail mispricing is systematic in retail-driven markets. Understanding why tails are underpriced gives you the ability to monitor whether the mechanism still holds.
Volume beats size. Tiny bets repeated thousands of times outperforms large bets repeated dozens of times when edge is genuine but modest.
Automation enables the strategy. Without a bot, you can’t execute this. The first-order requirement isn’t meteorological expertise — it’s operational execution capability.
Passive diversification is a feature. Trading all available weather markets simultaneously gives breadth that an event-specific strategy can’t replicate.
Public strategies degrade. When the gopfan2 approach spread in the community, imitators followed. The next durable edge in weather markets probably looks like a refinement, not a replication.

Request Access to WeatherCaster