Poly Research & Robotics
PR&R / Trader Report
Home / Reports / B1C58
Polymarket / On-chain

B1C58

On-chain analysis of Polymarket trader B1C58. Active over 29 days with 143,297 trades across 7,548 markets, netting +$27,984 at +1.6% ROI.

Published May 07, 2026 ~9 min read By PR&R Research View on Polymarket →
Volume traded
$1.76M
29-day window
Realized return
+1.6%
Cash-flow accounting
Top category share
100%
Crypto of total volume
Both-sides rate
99.0%
Market-maker shape
// 001 / Analysis

The portfolio shape, and where the edge appears to come from.

Wallet activity across 29 days, every fill mapped, profile traced.

This wallet ran for 29 active days (29 calendar days in window) and logged 143,297 trades - 143,297 BUY tickets and 0 SELL tickets - across 7,548 distinct markets and 7,547 events. Gross turnover for the window: $1.76M, averaging 4,941 trades/day.

On the resolved-BUY book, the wallet has 143,291 closed positions with a 50.3% win rate (72,100 wins / 71,191 losses), netting +$27,984 on $1.76M deployed - a realized return of +1.59%.

Classical market maker, crypto-focused: 99.0% both-sides participation, top vertical Crypto at 100.0% of capital.

Sizing & execution profile

The median fill is $10.13, the mean is $12.31, P95 is $28.50, P99 is $36.48 and the largest single fill in the window was $54.37. The top 5% of trades carry 13.6% of total capital - a sizing curve where most fills are small probes and a small handful of larger fills express conviction.

Intra-market consecutive fill gap is 10s median with 45.4% of pairs firing in under ten seconds - a latency signature consistent with automated execution.

Where the edge lives

With 99.0% of markets seeing both YES and NO bought, the book has the surface profile of a market maker. Median paired cost is $0.9893; 60.3% of paired markets close under $1.00. The conviction curve and dominance buckets in the quantitative section break down which side of the spread carries the realized P/L.

Of 29 rolling 15-day windows, 29 close positive (100.0% green-window rate). This is the consistency profile that determines how reproducible the strategy is over time.

// 002 / Figure

Cumulative P/L over the window.

The line is daily cumulative net P/L. Mouse along it for daily detail. The dashed grey trace, when present, is cumulative BUY notional deployed.

// 003 / Reverse-engineering report

Reverse-engineering report

Every fill mapped, the asymmetric profile traced, the math behind the edge.

Reverse-engineering report not authored for this run. Re-queue the job to regenerate.

// 004 / Quantitative breakdown

Quantitative breakdown

Phase-by-phase statistical report. Methodology, distributions, per-bucket P/L.

Wallet: 0x60889af507ec3f9da136f38b1c58080bec32f361 Window: 2026-04-07 → 2026-05-05 (29 active / 29 calendar days) Methodology: Cash-flow P/L = -buy_usdc + sell_usdc + remaining_share_payout. Resolved shares settle at $1 (win) / $0 (loss); open positions marked at last price.


Phase 1 - Trader Profile

Scale

MetricValue
Total trades143,297
BUY trades143,297
SELL trades0 (0.0% of all)
Unique markets7,548
Unique events7,547
Active calendar days29 of 29
Trades per active day4,941
BUY notional$1,763,466
SELL notional$0
Gross turnover$1,763,466

Trade-size distribution (USDC per fill)

MetricValue
median$10.13
mean$12.31
p95$28.50
p99$36.48
max$54.37
Top 5% share of capital13.6%

Inter-trade gap, same (market, outcome)

MetricValue
Median (s)10.0
Mean (s)20.3
P10 (s)0.0
P90 (s)52.0
% under 1s0.0%
% under 10s45.4%
% under 60s92.3%

Phase 2 & 3 - Both-Sides Participation, Dominance Curve

  • Both-sides rate: 99.03% (7,475 of 7,548 markets)
  • Median paired cost: $0.9893
  • Mean paired cost: $0.9870
  • Paired cost % under $1.00: 60.3%
  • Paired cost % under $0.97: 34.3%
  • Median 2nd-side hedge lag: 10s

Dominance buckets

BucketMarketsDom WRMean PairedAvg Mkt P/L
1.0–1.5x2,40961.7%$0.9882-
1.5–2.0x1,56274.8%$0.9946-
2.0–3.0x1,81484.3%$0.9917-
3.0x+1,69095.7%$0.9733-

Phase 4 - Entry-Price Analysis

BandBUY tradesResolvedWinsWRCapitalP/LROI
$0.00–$0.106,95404156.0%$12.6K-$2,093-16.66%
$0.10–$0.2011,51101,67414.5%$43.2K-$1,218-2.82%
$0.20–$0.3013,28803,47326.1%$83.0K+$3,784+4.56%
$0.30–$0.4016,70106,00836.0%$143.5K+$2,233+1.56%
$0.40–$0.5022,779010,67846.9%$255.5K+$10,690+4.18%
$0.50–$0.6026,354014,61355.4%$353.3K+$5,916+1.67%
$0.60–$0.7016,190010,52165.0%$254.9K+$1,954+0.77%
$0.70–$0.8011,90708,97775.4%$219.9K+$1,879+0.85%
$0.80–$0.9010,17408,70985.6%$219.1K+$1,850+0.84%
$0.90–$1.007,43307,03294.6%$178.6K+$2,990+1.67%

Phase 5 - Category & Vertical Breakdown

CategoryBUY tradesBUY $ResolvedWRP/LROI
Crypto143,288$1.76M143,28850.3%+$27,996+1.59%
Other7$3330.0%-$12-100.00%
Soccer1$300.0%+$00.00%
Politics1$100.0%+$00.00%

Phase 6 - Timing & Execution

Net P/L by hour (UTC)

HourP/LWR
00:00+$1,22950.4%
01:00+$1,12050.1%
02:00+$1,48650.5%
03:00+$85450.3%
04:00+$44550.3%
05:00+$38650.4%
06:00+$94950.5%
07:00+$67950.2%
08:00+$1,10650.5%
09:00+$1,13350.6%
10:00+$1,33750.3%
11:00+$1,40550.3%
12:00+$1,48650.5%
13:00+$1,09049.9%
14:00+$1,72450.2%
15:00+$1,05350.1%
16:00+$1,55650.2%
17:00+$92850.3%
18:00+$1,34450.3%
19:00+$1,38950.2%
20:00+$1,57250.5%
21:00+$1,98650.6%
22:00+$87050.0%
23:00+$85750.7%

Phase 8 - Rolling Window Consistency

  • Rolling 7-day windows green: 27 of 29 (93.1%)
  • Rolling 7-day P/L range: -$2,877 → +$13,096
  • Rolling 15-day windows green: 29 of 29 (100.0%)
  • Rolling 15-day P/L range: +$1,851 → +$23,141

Weekly P/L

WeekSpanTradesWRP/LCumulative
W152026-04-07 → 2026-04-1230,62750.2%+$9,039+$9,039
W162026-04-13 → 2026-04-1927,38450.2%+$10,027+$19,066
W172026-04-20 → 2026-04-2639,07650.5%+$10,535+$29,602
W182026-04-27 → 2026-05-0332,96150.4%+$152+$29,753
W192026-05-04 → 2026-05-0513,24350.0%-$1,769+$27,984

Phase 9 - P/L Decomposition

MetricValue
BUY USDC out-$1,763,466
SELL USDC in+$0
Theoretical spread P/L+$18,449
Hedge-tax outflow$676.5K
Net realized P/L+$27,984
Net ROI on BUY notional+1.59%

Phase 10 - Top Markets by Volume

MarketTradesVolumeResolvedP/L
Bitcoin Up or Down - May 5, 11:30AM-11:35AM ET37$74437+$16
Bitcoin Up or Down - May 5, 3:50PM-3:55PM ET35$74135-$21
Bitcoin Up or Down - May 5, 6:35PM-6:40PM ET36$73536-$15
Bitcoin Up or Down - May 3, 7:25PM-7:30PM ET36$73536-$15
Bitcoin Up or Down - May 5, 10:10AM-10:15AM ET36$72836-$8
Bitcoin Up or Down - May 5, 1:35AM-1:40AM ET36$72636-$6
Bitcoin Up or Down - May 5, 2:05AM-2:10AM ET36$72336-$3
Bitcoin Up or Down - May 5, 1:25AM-1:30AM ET34$72234-$42
Bitcoin Up or Down - May 5, 7:20PM-7:25PM ET34$72134-$41
Bitcoin Up or Down - May 5, 6:45PM-6:50PM ET35$72035-$40

Top 10 winners by P/L

MarketVolumeNet P/L
Bitcoin Up or Down - April 20, 10:05AM-10:10AM ET$672+$88
Bitcoin Up or Down - April 20, 9:05AM-9:10AM ET$176+$64
Bitcoin Up or Down - April 24, 6:05AM-6:10AM ET$305+$55
Bitcoin Up or Down - April 23, 10:00PM-10:05PM ET$669+$51
Bitcoin Up or Down - April 24, 2:10PM-2:15PM ET$270+$50
Bitcoin Up or Down - April 23, 6:45AM-6:50AM ET$252+$48
Bitcoin Up or Down - April 12, 1:45PM-1:50PM ET$192+$48
Bitcoin Up or Down - April 19, 12:45AM-12:50AM ET$313+$47
Bitcoin Up or Down - April 11, 12:30PM-12:35PM ET$73+$47
Bitcoin Up or Down - April 9, 9:25PM-9:30PM ET$206+$46

Top 10 losers by P/L

MarketVolumeNet P/L
Bitcoin Up or Down - May 1, 8:45PM-8:50PM ET$84-$84
Bitcoin Up or Down - May 5, 3:25AM-3:30AM ET$546-$66
Bitcoin Up or Down - May 5, 12:50AM-12:55AM ET$702-$62
Bitcoin Up or Down - May 5, 8:30AM-8:35AM ET$697-$57
Bitcoin Up or Down - May 5, 10:15AM-10:20AM ET$696-$56
Bitcoin Up or Down - May 5, 2:30PM-2:35PM ET$414-$54
Bitcoin Up or Down - May 3, 5:15PM-5:20PM ET$218-$53
Bitcoin Up or Down - May 1, 8:15PM-8:20PM ET$201-$51
Bitcoin Up or Down - April 12, 12:25PM-12:30PM ET$409-$49
Bitcoin Up or Down - April 23, 9:10PM-9:15PM ET$568-$48

Report generated 2026-05-07 05:48 UTC.

// 005 / Filter strategy

Filter strategy

Which standard filters move the needle on this trader, and which destroy the edge.

Each filter is applied to the resolved-BUY set. Cash-flow P/L is allocated per-trade using pnl = frac × (sell_usdc + remaining_payout) − usdc. ROI is measured against BUY notional within the filter. Baseline P/L: +$27,984 on $1.76M deployed.


Filter results table

FilterTradesWRCapitalP/LROIΔ vs baseline
Unfiltered baseline143,29150.3%$1.76M+$27,984+1.59%-
Price band 0.30–0.7083,26251.3%$1.03M+$20,764+2.02%-$7.2K
High-conviction (dom≥2×)33,88788.7%$637.7K+$125,302+19.65%+$97.3K
Top category (Crypto)143,28850.3%$1.76M+$27,996+1.59%+$12
Exclude worst hours116,77550.4%$1.43M+$23,851+1.67%-$4.1K
Combined stack67,64451.4%$830.9K+$18,689+2.25%-$9.3K

Filter-by-filter commentary

The single most productive filter is High-conviction (dom≥2×) - it adds $97.3K of P/L over baseline (+19.65% ROI on the surviving 33,887 trades).

The most destructive standard filter is Combined stack: it costs $9.3K versus baseline. This usually means the filter is selecting against the actual mechanism the trader is using.

// 006 / Replication playbook

Replication playbook

Where the edge is portable, and where it isn't.

Archetype: Classical market maker, crypto-focused. Reference book: $1.76M BUY notional → +$27,984 net P/L → +1.59% ROI in 29 active days.


Where the edge appears reproducible

  • Spread capture. Both-sides participation at 99.0% suggests the wallet is paying for the spread on most markets. A market-making bot that mimics the entry pattern on the same vertical can absorb the same paired-cost dynamics with similar capital efficiency.
  • Category specialization. Capital is concentrated in Crypto; reproduce by gating on the same vertical and matching the price-band distribution in the quantitative section.
  • Sizing discipline. Median ticket $10.13 with P95 at $28.50 - the modal clip size is small and the tail is bounded. Fixed-clip sizing inside this envelope matches the wallet's risk profile and avoids tail-blowup.

Where the edge probably is not reproducible

  • Off-platform signal. Any directional view (sports info, model-driven conviction, cross-venue arbitrage) cannot be reconstructed from on-chain trade data alone. Linear approximations to the wallet's sizing curve typically recover only a fraction of the realized return.
  • Latency edge. If the wallet's intra-market gap distribution is sub-second for a meaningful share of fills, the strategy is competing on infrastructure as well as logic. Replicating that requires co-located RPC and a non-trivial engineering build.

The reproducible parts give you a reasonable bot. The non-reproducible parts tell you what to keep working on. Use the quantitative section as the spec - every threshold and distribution there is what you would seed into a backtest, not the high-level summary above.

// 001 / Analysis

The portfolio shape, and where the edge appears to come from.

Wallet activity across 29 days, every fill mapped, profile traced.

This wallet ran for 29 active days (29 calendar days in window) and logged 143,297 trades - 143,297 BUY tickets and 0 SELL tickets - across 7,548 distinct markets and 7,547 events. Gross turnover for the window: $1.76M, averaging 4,941 trades/day.

On the resolved-BUY book, the wallet has 143,291 closed positions with a 50.3% win rate (72,100 wins / 71,191 losses), netting +$27,984 on $1.76M deployed - a realized return of +1.59%.

Classical market maker, crypto-focused: 99.0% both-sides participation, top vertical Crypto at 100.0% of capital.

Sizing & execution profile

The median fill is $10.13, the mean is $12.31, P95 is $28.50, P99 is $36.48 and the largest single fill in the window was $54.37. The top 5% of trades carry 13.6% of total capital - a sizing curve where most fills are small probes and a small handful of larger fills express conviction.

Intra-market consecutive fill gap is 10s median with 45.4% of pairs firing in under ten seconds - a latency signature consistent with automated execution.

Where the edge lives

With 99.0% of markets seeing both YES and NO bought, the book has the surface profile of a market maker. Median paired cost is $0.9893; 60.3% of paired markets close under $1.00. The conviction curve and dominance buckets in the quantitative section break down which side of the spread carries the realized P/L.

Of 29 rolling 15-day windows, 29 close positive (100.0% green-window rate). This is the consistency profile that determines how reproducible the strategy is over time.

// 002 / Figure

Cumulative P/L over the window.

The line is daily cumulative net P/L. Mouse along it for daily detail. The dashed grey trace, when present, is cumulative BUY notional deployed.

// 003 / Reverse-engineering report

Reverse-engineering report

Every fill mapped, the asymmetric profile traced, the math behind the edge.

Reverse-engineering report not authored for this run. Re-queue the job to regenerate.

// 004 / Quantitative breakdown

Quantitative breakdown

Phase-by-phase statistical report. Methodology, distributions, per-bucket P/L.

Wallet: 0x60889af507ec3f9da136f38b1c58080bec32f361 Window: 2026-04-07 → 2026-05-05 (29 active / 29 calendar days) Methodology: Cash-flow P/L = -buy_usdc + sell_usdc + remaining_share_payout. Resolved shares settle at $1 (win) / $0 (loss); open positions marked at last price.


Phase 1 - Trader Profile

Scale

MetricValue
Total trades143,297
BUY trades143,297
SELL trades0 (0.0% of all)
Unique markets7,548
Unique events7,547
Active calendar days29 of 29
Trades per active day4,941
BUY notional$1,763,466
SELL notional$0
Gross turnover$1,763,466

Trade-size distribution (USDC per fill)

MetricValue
median$10.13
mean$12.31
p95$28.50
p99$36.48
max$54.37
Top 5% share of capital13.6%

Inter-trade gap, same (market, outcome)

MetricValue
Median (s)10.0
Mean (s)20.3
P10 (s)0.0
P90 (s)52.0
% under 1s0.0%
% under 10s45.4%
% under 60s92.3%

Phase 2 & 3 - Both-Sides Participation, Dominance Curve

  • Both-sides rate: 99.03% (7,475 of 7,548 markets)
  • Median paired cost: $0.9893
  • Mean paired cost: $0.9870
  • Paired cost % under $1.00: 60.3%
  • Paired cost % under $0.97: 34.3%
  • Median 2nd-side hedge lag: 10s

Dominance buckets

BucketMarketsDom WRMean PairedAvg Mkt P/L
1.0–1.5x2,40961.7%$0.9882-
1.5–2.0x1,56274.8%$0.9946-
2.0–3.0x1,81484.3%$0.9917-
3.0x+1,69095.7%$0.9733-

Phase 4 - Entry-Price Analysis

BandBUY tradesResolvedWinsWRCapitalP/LROI
$0.00–$0.106,95404156.0%$12.6K-$2,093-16.66%
$0.10–$0.2011,51101,67414.5%$43.2K-$1,218-2.82%
$0.20–$0.3013,28803,47326.1%$83.0K+$3,784+4.56%
$0.30–$0.4016,70106,00836.0%$143.5K+$2,233+1.56%
$0.40–$0.5022,779010,67846.9%$255.5K+$10,690+4.18%
$0.50–$0.6026,354014,61355.4%$353.3K+$5,916+1.67%
$0.60–$0.7016,190010,52165.0%$254.9K+$1,954+0.77%
$0.70–$0.8011,90708,97775.4%$219.9K+$1,879+0.85%
$0.80–$0.9010,17408,70985.6%$219.1K+$1,850+0.84%
$0.90–$1.007,43307,03294.6%$178.6K+$2,990+1.67%

Phase 5 - Category & Vertical Breakdown

CategoryBUY tradesBUY $ResolvedWRP/LROI
Crypto143,288$1.76M143,28850.3%+$27,996+1.59%
Other7$3330.0%-$12-100.00%
Soccer1$300.0%+$00.00%
Politics1$100.0%+$00.00%

Phase 6 - Timing & Execution

Net P/L by hour (UTC)

HourP/LWR
00:00+$1,22950.4%
01:00+$1,12050.1%
02:00+$1,48650.5%
03:00+$85450.3%
04:00+$44550.3%
05:00+$38650.4%
06:00+$94950.5%
07:00+$67950.2%
08:00+$1,10650.5%
09:00+$1,13350.6%
10:00+$1,33750.3%
11:00+$1,40550.3%
12:00+$1,48650.5%
13:00+$1,09049.9%
14:00+$1,72450.2%
15:00+$1,05350.1%
16:00+$1,55650.2%
17:00+$92850.3%
18:00+$1,34450.3%
19:00+$1,38950.2%
20:00+$1,57250.5%
21:00+$1,98650.6%
22:00+$87050.0%
23:00+$85750.7%

Phase 8 - Rolling Window Consistency

  • Rolling 7-day windows green: 27 of 29 (93.1%)
  • Rolling 7-day P/L range: -$2,877 → +$13,096
  • Rolling 15-day windows green: 29 of 29 (100.0%)
  • Rolling 15-day P/L range: +$1,851 → +$23,141

Weekly P/L

WeekSpanTradesWRP/LCumulative
W152026-04-07 → 2026-04-1230,62750.2%+$9,039+$9,039
W162026-04-13 → 2026-04-1927,38450.2%+$10,027+$19,066
W172026-04-20 → 2026-04-2639,07650.5%+$10,535+$29,602
W182026-04-27 → 2026-05-0332,96150.4%+$152+$29,753
W192026-05-04 → 2026-05-0513,24350.0%-$1,769+$27,984

Phase 9 - P/L Decomposition

MetricValue
BUY USDC out-$1,763,466
SELL USDC in+$0
Theoretical spread P/L+$18,449
Hedge-tax outflow$676.5K
Net realized P/L+$27,984
Net ROI on BUY notional+1.59%

Phase 10 - Top Markets by Volume

MarketTradesVolumeResolvedP/L
Bitcoin Up or Down - May 5, 11:30AM-11:35AM ET37$74437+$16
Bitcoin Up or Down - May 5, 3:50PM-3:55PM ET35$74135-$21
Bitcoin Up or Down - May 5, 6:35PM-6:40PM ET36$73536-$15
Bitcoin Up or Down - May 3, 7:25PM-7:30PM ET36$73536-$15
Bitcoin Up or Down - May 5, 10:10AM-10:15AM ET36$72836-$8
Bitcoin Up or Down - May 5, 1:35AM-1:40AM ET36$72636-$6
Bitcoin Up or Down - May 5, 2:05AM-2:10AM ET36$72336-$3
Bitcoin Up or Down - May 5, 1:25AM-1:30AM ET34$72234-$42
Bitcoin Up or Down - May 5, 7:20PM-7:25PM ET34$72134-$41
Bitcoin Up or Down - May 5, 6:45PM-6:50PM ET35$72035-$40

Top 10 winners by P/L

MarketVolumeNet P/L
Bitcoin Up or Down - April 20, 10:05AM-10:10AM ET$672+$88
Bitcoin Up or Down - April 20, 9:05AM-9:10AM ET$176+$64
Bitcoin Up or Down - April 24, 6:05AM-6:10AM ET$305+$55
Bitcoin Up or Down - April 23, 10:00PM-10:05PM ET$669+$51
Bitcoin Up or Down - April 24, 2:10PM-2:15PM ET$270+$50
Bitcoin Up or Down - April 23, 6:45AM-6:50AM ET$252+$48
Bitcoin Up or Down - April 12, 1:45PM-1:50PM ET$192+$48
Bitcoin Up or Down - April 19, 12:45AM-12:50AM ET$313+$47
Bitcoin Up or Down - April 11, 12:30PM-12:35PM ET$73+$47
Bitcoin Up or Down - April 9, 9:25PM-9:30PM ET$206+$46

Top 10 losers by P/L

MarketVolumeNet P/L
Bitcoin Up or Down - May 1, 8:45PM-8:50PM ET$84-$84
Bitcoin Up or Down - May 5, 3:25AM-3:30AM ET$546-$66
Bitcoin Up or Down - May 5, 12:50AM-12:55AM ET$702-$62
Bitcoin Up or Down - May 5, 8:30AM-8:35AM ET$697-$57
Bitcoin Up or Down - May 5, 10:15AM-10:20AM ET$696-$56
Bitcoin Up or Down - May 5, 2:30PM-2:35PM ET$414-$54
Bitcoin Up or Down - May 3, 5:15PM-5:20PM ET$218-$53
Bitcoin Up or Down - May 1, 8:15PM-8:20PM ET$201-$51
Bitcoin Up or Down - April 12, 12:25PM-12:30PM ET$409-$49
Bitcoin Up or Down - April 23, 9:10PM-9:15PM ET$568-$48

Report generated 2026-05-07 05:48 UTC.

// 005 / Filter strategy

Filter strategy

Which standard filters move the needle on this trader, and which destroy the edge.

Each filter is applied to the resolved-BUY set. Cash-flow P/L is allocated per-trade using pnl = frac × (sell_usdc + remaining_payout) − usdc. ROI is measured against BUY notional within the filter. Baseline P/L: +$27,984 on $1.76M deployed.


Filter results table

FilterTradesWRCapitalP/LROIΔ vs baseline
Unfiltered baseline143,29150.3%$1.76M+$27,984+1.59%-
Price band 0.30–0.7083,26251.3%$1.03M+$20,764+2.02%-$7.2K
High-conviction (dom≥2×)33,88788.7%$637.7K+$125,302+19.65%+$97.3K
Top category (Crypto)143,28850.3%$1.76M+$27,996+1.59%+$12
Exclude worst hours116,77550.4%$1.43M+$23,851+1.67%-$4.1K
Combined stack67,64451.4%$830.9K+$18,689+2.25%-$9.3K

Filter-by-filter commentary

The single most productive filter is High-conviction (dom≥2×) - it adds $97.3K of P/L over baseline (+19.65% ROI on the surviving 33,887 trades).

The most destructive standard filter is Combined stack: it costs $9.3K versus baseline. This usually means the filter is selecting against the actual mechanism the trader is using.

// 006 / Replication playbook

Replication playbook

Where the edge is portable, and where it isn't.

Archetype: Classical market maker, crypto-focused. Reference book: $1.76M BUY notional → +$27,984 net P/L → +1.59% ROI in 29 active days.


Where the edge appears reproducible

  • Spread capture. Both-sides participation at 99.0% suggests the wallet is paying for the spread on most markets. A market-making bot that mimics the entry pattern on the same vertical can absorb the same paired-cost dynamics with similar capital efficiency.
  • Category specialization. Capital is concentrated in Crypto; reproduce by gating on the same vertical and matching the price-band distribution in the quantitative section.
  • Sizing discipline. Median ticket $10.13 with P95 at $28.50 - the modal clip size is small and the tail is bounded. Fixed-clip sizing inside this envelope matches the wallet's risk profile and avoids tail-blowup.

Where the edge probably is not reproducible

  • Off-platform signal. Any directional view (sports info, model-driven conviction, cross-venue arbitrage) cannot be reconstructed from on-chain trade data alone. Linear approximations to the wallet's sizing curve typically recover only a fraction of the realized return.
  • Latency edge. If the wallet's intra-market gap distribution is sub-second for a meaningful share of fills, the strategy is competing on infrastructure as well as logic. Replicating that requires co-located RPC and a non-trivial engineering build.

The reproducible parts give you a reasonable bot. The non-reproducible parts tell you what to keep working on. Use the quantitative section as the spec - every threshold and distribution there is what you would seed into a backtest, not the high-level summary above.

Join Discord