RN1

RN1 is best understood not as a gambler trying to guess winners one market at a time, but as a highly automated inventory trader operating inside a binary prediction market. The strategy is closer to a wholesaler buying discounted claim tickets than to a traditional bettor firing takes.

In a binary market, one side eventually pays $1.00 and the other side pays $0.00. If a trader can acquire exposure to both sides for less than $1.00 combined, the matched portion of that position behaves like buying a dollar for ninety-six or ninety-seven cents.

Executive verdict

• The raw logs overwhelmingly support a both-sides accumulation strategy.
• The strategy is bursty, automated, and inventory-aware.
• The 2.0x dominance ratio is the key side-weight threshold.
• The core edge is spread capture plus selective directional leaning.
• The packet’s “zero sells” claim is directionally right but literally false.

Sources used

• All ten Markdown packet documents in the folder.
• Six monthly trade logs: October 2025 through March 2026.
• Direct statistical aggregation over trade size, price, timing, side usage, and market-level allocation.

The report intentionally separates hard facts from inferences. Facts are things the CSVs can prove directly. Inferences are things that are strongly suggested by trade behavior but cannot be established with certainty because order book state, external odds history, and final resolved outcomes are not fully included in this folder.

Evidence hierarchy

1. Highest confidence: direct CSV facts like trade size, timestamps, price bands, market counts, and buy/sell frequencies.
2. Medium confidence: reconstructed behavior like paired-cost math, DCA patterns, or side-switch timing.
3. Lower confidence: any claim requiring missing information such as book depth at the instant of entry, external odds feeds, or official market resolution labels.

Why binary markets matter

Polymarket’s binary structure makes this whole strategy possible. A two-outcome market is algebraically simple: one side will be worth $1, the other side will be worth $0, and the two together represent the complete state space of the event.

That means every trade can be reframed as a question about probability, inventory, and settlement value. RN1 appears to exploit this structure relentlessly.

Key terms

Measured baseline

1. Market discovery

RN1 appears to scan a wide universe and select markets that are binary, active, liquid enough, and worth watching relative to likely fair value.

2. Probe and test

He opens many markets with small trades, almost like tapping a microphone before beginning the speech. The first clip tests whether the market will accept more size cleanly.

3. Accumulate in bursts

Once a market passes the test, RN1 keeps returning in short machine bursts. He is not dropping one rock into the pond; he is skipping many small stones in sequence.

4. Pair the position

In a large share of markets, the second side arrives quickly. That is the mechanical heart of the strategy. The goal is not merely to own the favorite; the goal is to own enough of both sides cheaply enough that the paired portion behaves like discounted settlement inventory.

5. Lean when conviction justifies it

RN1 is not always perfectly balanced. The dominance ratio shows frequent asymmetric allocation. That means the strategy is not pure arbitrage. It is better described as spread capture with directional bias layered on top.

The 2.0x ratio threshold is the most useful operational divider: below it, the position behaves more like neutral inventory harvesting; above it, the trade starts to look like informed leaning.

Observed filters

• Binary structure: no traded condition in the raw logs shows three or more bought outcomes.
• Practical pairability: the strategy clearly prefers markets where the second side can be obtained within the same operating window.
• Fast capital recycling: the packet’s sports-heavy claim lines up with the short holding-cycle behavior visible in the logs.
• Liquidity preference: the market concentration statistics imply RN1 repeatedly chooses markets that can absorb meaningful size.

Why sports likely dominates

Sports markets are attractive for three reasons:

1. The event timetable is known.
2. External odds are easy to obtain.
3. Settlement usually happens quickly enough to recycle capital.

Probe-first behavior

Median first trade per market is only $16.60. Even in markets where RN1 ultimately deploys over $1,000, the median first trade is only $47.58.

That is the statistical fingerprint of a trader who explores before committing. It is what a good market operator does when slippage matters and book depth is uncertain.

Burst cadence

The same market+outcome median inter-trade gap is 10 seconds, and the 25th percentile is 2 seconds. Same-outcome run length within a market has a median of 2.

This looks like repeated aggressive clips at the touch, not slow passive posting and waiting.

Second-side acquisition

For markets where RN1 bought both outcomes, the second side usually appears quickly. Median lag is 20.4 minutes. On meaningful both-side markets where the smaller side reaches at least $100, about 68.9% receive the second side within one hour and 95.4% within one day.

Why the 40c-80c zone matters

The data shows 58.0% of capital deployed between 0.40 and 0.80. That range is strategically attractive because both sides still have enough economic significance to justify hedging, but neither side is so extreme that one leg becomes dead capital.

Capital by price band

Paired-cost reconstruction

A useful way to think about RN1’s economics is to ask: after pairing the shares bought on each side, what did those matched share pairs cost on average?

On meaningful both-sides markets where the smaller side received at least $100, the median paired cost comes out to 0.9632 dollars per matched share pair, and about 63.16% of those markets remain below $1.00 on the paired portion.

This is an approximation based on weighted average fill prices, not a perfect FIFO matching model. It is still directionally useful because it shows that a large share of the strategy genuinely resembles buying eventual settlement value at a discount.

Definition

Dominance ratio is simply:

dominance_ratio = capital_on_larger_side / capital_on_smaller_side

When the ratio is near 1.0x, the trader is close to balanced. When it rises above 2.0x, the trader is expressing a clear directional preference while still retaining some hedge.

Measured distribution

How to interpret the threshold

The dominance ratio behaves like a dimmer switch, not a light switch. It is not “hedged” versus “not hedged.” It is a gradual measure of how much the trade shifts from neutral inventory collection toward informed directional preference.

The packet’s resolved-market analysis claims that above 2.0x, the dominant side wins much more often. The raw logs cannot independently prove the win rate because the folder does not contain a full settlement table, but they do confirm that the 2.0x zone is a natural structural breakpoint in RN1’s capital allocation.

Power-law ticketing

RN1’s sizing is not uniform. It is a classic power-law profile:

• Many very small tickets.
• A smaller middle layer of meaningful orders.
• A thin tail of larger clips that carries much of the capital.

That is why median trade size is only $9.84 while mean trade size jumps to $93.82.

What this means operationally

The strategy should not be implemented as “decide market, place one large order.” A better mental model is:

1. Set a market budget.
2. Open with a probe.
3. Increase only when fills are clean.
4. Use many small clips to reach the target.

Market-level concentration

Hold-heavy, not hyperactive

The packet’s main characterization holds: RN1 is not constantly trading in and out. Once a position is built, it is usually kept until market resolution. The SELL exceptions are real, but tiny relative to the total footprint.

This matters because it simplifies the strategy architecture. The real complexity is front-loaded into market selection, entry timing, side weighting, and inventory math. Exit timing is comparatively simple.

Important caveat

“Never sells” should be rewritten as:

Recommended adaptation for a smaller proprietary bot

A smaller operator does not have to mimic RN1’s exit behavior perfectly. It may be rational to add an optional near-certainty recycle rule, such as allowing pre-resolution exit when a winning leg can be sold near 0.98 - 0.999 and the capital is more useful elsewhere.

Diversification as primary defense

RN1’s risk control is not elegant stop-loss engineering. It is breadth. Daily market count climbs materially over time, reaching averages near 384 in January, 383 in February, and 400 in March.

Recommended practical caps

• Maximum market exposure: 1% - 2% of daily budget.
• Maximum side exposure: 0.8% - 1.5% of daily budget.
• Max unpaired inventory age: about 24 hours unless special circumstances apply.
• Pause accumulation if projected paired cost rises above $1.00.

Why this matters

A spread-capture strategy can quietly mutate into a portfolio of disguised directional bets if the smaller side is neglected for too long. The right way to think about risk here is not only “how much can this market lose?” but also “how much of this position is still true paired inventory, and how much has become unhedged opinion?”

When RN1 is most active

The strongest block of activity is late UTC / U.S. event hours, especially roughly 15:00 - 21:00 UTC, with the highest capital shares clustering around 19:00 - 21:00 UTC.

Minimal working bot loop

# RN1-style operator loop
1. Ingest market metadata from Gamma.
2. Ingest live prices and depth from CLOB.
3. Ingest fair value from external odds or models.
4. Filter to binary, short-dated, tradeable markets.
5. Estimate paired-cost opportunity and target dominance ratio.
6. Open with a probe order.
7. Scale using short aggressive bursts.
8. Acquire second side quickly if running in both-sides mode.
9. Stop adding if paired-cost math breaks or risk caps are hit.
10. Hold to resolution by default; optionally recycle near-certain winners.

High confidence facts

• RN1 is heavily automated.
• RN1 mostly buys, with only rare sells.
• RN1 frequently buys both sides of the same market.
• RN1 prefers many small trades over single large blocks.
• RN1 often completes the second side quickly.

Medium confidence inferences

• RN1 likely uses aggressive near-ask execution.
• RN1 likely relies on external fair-value references.
• RN1 likely prioritizes scheduled event markets for recycling efficiency.
• RN1 likely uses liquidity-aware scaling after probe entries.

Open unknowns

• Exact order-book state at every fill.
• Exact external odds source and update frequency.
• Exact realized P&L by market after resolution.
• Exact reason behind each rare SELL event.

The cleanest summary is this: RN1 appears to run a high-throughput, liquidity-aware prediction-market engine that combines discounted paired inventory acquisition with selective directional leaning. The trader is not acting like a pure bettor and not acting like a pure market-neutral arbitrageur either. The strategy lives in the middle: it buys the market’s settlement structure at favorable prices, but it also uses asymmetric sizing when conviction is stronger.

If you strip away the noise, RN1’s edge looks like a disciplined combination of market microstructure exploitation, binary settlement math, fast execution, and diversified capital deployment. That is the architecture worth preserving.