BUILD IN PUBLIC · EDGE SCOREBOARD

Does DDN SI actually have edge?

We're building this in public. Here's whether DDN SI actually has edge — measured, not claimed. SI is a name, not a performance claim. Every lane starts in Shadow: DDN SI logs what it would bet and we score it against the closing price and the final result, without risking money. A lane only earns real stakes when the numbers below say so.

Shadow intents appear only after their event starts. Live bets appear only after they fill. Nothing here is published before then. Updated 2026-11-01 23:28 UTC.

Where we started (2026-10-05)

MeasureThenTarget
Realized edge, 90d, clean+0.2pp (MLB, 99 bets)> +2pp with 95% CI-low > 0
Claimed edge vs realized12.3pp claimed, 0.0pp realizedCalibration slope 0.8-1.2
CLV beat rate35.7%> 55%, avg CLV > 0
Model edge to result slope0.06>= 0.7

A 2026-10-07 backtest found that, priced correctly, DDN SI's model has about no edge over Kalshi on NHL puck lines/totals and MLB run lines, and MLB totals run market-only. Expect bets to be rare. That is the point of measuring first.

Every lane at a glance

Shadow lanes are measured on every candidate without money; maker and sharp-lag lanes show whether they are live or demoted. Realized edge uses live bets when a lane has any settled, otherwise shadow candidates. Negative numbers are shown at the same size as positive ones.

LaneTypeModeShadow nLive nBeat close (shadow)Beat close (live)Realized edgeMaker P&L
MLB · Moneyline Shadow + live Shadow 17 0 11.8% n=17 n/a n=0 -0.2pp shadow
MLB · Spread Shadow + live Shadow 18 0 33.3% n=18 n/a n=0 +9.6pp shadow
MLB · Total Shadow + live Shadow 12 0 50.0% n=12 n/a n=0 -8.4pp shadow
NHL · Spread Shadow + live Shadow 282 0 35.1% n=282 n/a n=0 -3.8pp shadow
Serie A · Moneyline Shadow + live Shadow 0 0 n/a n=0 n/a n=0 n/a shadow
Serie A · Spread Shadow + live Shadow 0 0 n/a n=0 n/a n=0 n/a shadow
Serie A · Total Shadow + live Shadow 0 0 n/a n=0 n/a n=0 n/a shadow
MLB · Sharp price lag Sharp lag Sharp lag · shadow 29 0 48.3% n=29 n/a n=0 +2.2pp shadow
NHL · Sharp price lag Sharp lag Sharp lag · shadow 463 0 41.7% n=463 n/a n=0 -5.2pp shadow
MLB · Run line maker (model -1.5) Maker Maker · live 3 0 100.0% n=3 n/a n=0 -25.0pp shadow $+0.00
MLB · Moneyline maker Maker Maker · shadow 0 0 n/a n=0 n/a n=0 n/a shadow $+0.00
NHL · Puck line maker Maker Maker · live 174 18 79.3% n=174 50.0% n=18 -4.4pp live $-1.96
NHL · Moneyline maker Maker Maker · shadow 174 0 75.3% n=174 n/a n=0 +5.7pp shadow $+0.00

Week by week

Is it getting better? Each point is one week (Monday start, by game start). Weeks without enough data are gaps, not zeros. Shadow = every measured candidate; live = real fills.

Beat-the-close rate by week

Beat-the-close rate by week Share of intents whose entry beat the closing price, per week, with 95% intervals. The dashed line is 50%, a coin flip against the close. The table below the charts lists every value. 0% 25% 50% 75% 100% 08-1008-3109-2110-12 Shadow, week of 2026-10-05: 49% (n=192), below the line Shadow, week of 2026-10-12: 52% (n=345) Shadow, week of 2026-10-19: 54% (n=312) Shadow, week of 2026-10-26: 46% (n=323), below the line Live, week of 2026-10-05: 50% (n=2) Live, week of 2026-10-12: 67% (n=3) Live, week of 2026-10-19: 56% (n=9) Live, week of 2026-10-26: 25% (n=4), below the line
Shadow (square) Live (circle) Dashed: 50%

Calibration slope by week

Calibration slope by week Logistic calibration slope of DDN SI per week (1.00 = well scaled, under 1 = overconfident). Needs 30 settled results in a week; the dashed line is 1.00. The table below the charts lists every value. 0.00 0.50 1.00 1.50 2.00 08-1008-3109-2110-12 Shadow, week of 2026-10-05: 0.85 (n=192), below the line Shadow, week of 2026-10-12: 1.40 (n=345) Shadow, week of 2026-10-19: 1.24 (n=312) Shadow, week of 2026-10-26: 1.19 (n=310)
Shadow (square) Live (circle) Dashed: 1.00
Charts as a table
Week ofShadow beat closenShadow slopeLive beat closenLive slope
2026-08-10n/a0n/an/a0n/a
2026-08-17n/a0n/an/a0n/a
2026-08-24n/a0n/an/a0n/a
2026-08-31n/a0n/an/a0n/a
2026-09-07n/a0n/an/a0n/a
2026-09-14n/a0n/an/a0n/a
2026-09-21n/a0n/an/a0n/a
2026-09-28n/a0n/an/a0n/a
2026-10-0549.0%1920.8550.0%2n/a
2026-10-1252.5%3451.4066.7%3n/a
2026-10-1953.5%3121.2455.6%9n/a
2026-10-2646.4%3231.1925.0%4n/a

Lanes

A lane is one sport and market type. Live means real filled bets. Shadow means what DDN SI would have bet (cleared its edge gate), with no money behind it.

MLB · Moneyline

Shadow + live Shadow Logged only. No money.
Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Market scan15013.3% n=15-1.27pp15-6.2pp
Model pick200.0% n=2-2.00pp2+45.0pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

MLB · Moneyline reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.240, market Brier 0.245; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 36%, happened 100% (n=1) Market: priced 44%, happened 0% (n=3) Market: priced 53%, happened 62% (n=8) Market: priced 64%, happened 60% (n=5) Model: said 47%, happened 25% (n=4) Model: said 55%, happened 62% (n=8) Model: said 64%, happened 60% (n=5)
Model (circle) Market price (square)

Brier: model 0.240 vs market 0.245 · slope n/a · n=17

Chart as a table
SourceBinPredictedHappenedn
Model40-50%47.4%25.0%4
Model50-60%55.0%62.5%8
Model60-70%63.7%60.0%5
Market30-40%35.8%100.0%1
Market40-50%43.6%0.0%3
Market50-60%53.1%62.5%8
Market60-70%64.5%60.0%5
What it takes to reach Probation
  • Not yet: Shadow intents with a measured close >= 50 (now: 0)
  • Not yet: Shadow CLV beat rate > 50% (now: n/a)

MLB · Spread

Shadow + live Shadow Logged only. No money.
Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Market scan18033.3% n=18+1.89pp18+9.6pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

MLB · Spread reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.272, market Brier 0.273; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 35%, happened 80% (n=5) Market: priced 42%, happened 33% (n=3) Market: priced 58%, happened 80% (n=5) Market: priced 65%, happened 40% (n=5) Model: said 35%, happened 80% (n=5) Model: said 42%, happened 33% (n=3) Model: said 59%, happened 80% (n=5) Model: said 65%, happened 40% (n=5)
Model (circle) Market price (square)

Brier: model 0.272 vs market 0.273 · slope n/a · n=18

Chart as a table
SourceBinPredictedHappenedn
Model30-40%35.4%80.0%5
Model40-50%41.7%33.3%3
Model50-60%58.9%80.0%5
Model60-70%65.5%40.0%5
Market30-40%35.2%80.0%5
Market40-50%42.3%33.3%3
Market50-60%58.2%80.0%5
Market60-70%64.7%40.0%5
What it takes to reach Probation
  • Not yet: Shadow intents with a measured close >= 50 (now: 0)
  • Not yet: Shadow CLV beat rate > 50% (now: n/a)

MLB · Total

Shadow + live Shadow Logged only. No money.
Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Market scan12050.0% n=12+4.00pp12-8.4pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

MLB · Total reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.251, market Brier 0.251; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 48%, happened 43% (n=7) Market: priced 51%, happened 40% (n=5) Model: said 48%, happened 43% (n=7) Model: said 51%, happened 40% (n=5)
Model (circle) Market price (square)

Brier: model 0.251 vs market 0.251 · slope n/a · n=12

Chart as a table
SourceBinPredictedHappenedn
Model40-50%48.2%42.9%7
Model50-60%51.3%40.0%5
Market40-50%48.2%42.9%7
Market50-60%51.3%40.0%5
What it takes to reach Probation
  • Not yet: Shadow intents with a measured close >= 50 (now: 0)
  • Not yet: Shadow CLV beat rate > 50% (now: n/a)

NHL · Spread

Shadow + live Shadow Logged only. No money.
Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Market scan278135.2% n=278-0.29pp274-3.2pp
Model pick4025.0% n=4-0.25pp4-46.5pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close1 / 1
Beat the close0.0%95% CI 0.0% to 79.3%
Average CLV-2.00pp
Realized edge+43.0pp
Win rate vs avg price100.0% vs 57¢
Flat-unit ROI (after fees)69.5%

Last 30 days: beat the close 0.0% (n=1), realized edge +43.0pp (n=1).

Calibration · all candidates

NHL · Spread reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.225, market Brier 0.225; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 29%, happened 33% (n=27) Market: priced 34%, happened 28% (n=83) Market: priced 43%, happened 35% (n=26) Market: priced 56%, happened 55% (n=42) Market: priced 65%, happened 68% (n=81) Market: priced 71%, happened 47% (n=19) Model: said 29%, happened 31% (n=26) Model: said 34%, happened 29% (n=85) Model: said 43%, happened 32% (n=22) Model: said 56%, happened 55% (n=42) Model: said 65%, happened 67% (n=84) Model: said 71%, happened 47% (n=19)
Model (circle) Market price (square)

Brier: model 0.225 vs market 0.225 · slope 1.03 · n=278

Chart as a table
SourceBinPredictedHappenedn
Model20-30%29.2%30.8%26
Model30-40%33.9%29.4%85
Model40-50%42.6%31.8%22
Model50-60%56.4%54.8%42
Model60-70%65.3%66.7%84
Model70-80%70.6%47.4%19
Market20-30%29.1%33.3%27
Market30-40%34.0%27.7%83
Market40-50%43.5%34.6%26
Market50-60%56.5%54.8%42
Market60-70%65.3%67.9%81
Market70-80%70.7%47.4%19
What it takes to reach Probation
  • Not yet: Shadow intents with a measured close >= 50 (now: 1)
  • Not yet: Shadow CLV beat rate > 50% (now: 0.0)

Serie A · Moneyline

Shadow + live Shadow Logged only. No money.

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

Serie A · Moneyline reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier n/a, market Brier n/a; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted No settled candidates yet
Model (circle) Market price (square)

Brier: model n/a vs market n/a · slope n/a · n=0

Chart as a table
SourceBinPredictedHappenedn
What it takes to reach Probation
  • Not yet: Shadow intents with a measured close >= 50 (now: 0)
  • Not yet: Shadow CLV beat rate > 50% (now: n/a)

Serie A · Spread

Shadow + live Shadow Logged only. No money.

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

Serie A · Spread reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier n/a, market Brier n/a; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted No settled candidates yet
Model (circle) Market price (square)

Brier: model n/a vs market n/a · slope n/a · n=0

Chart as a table
SourceBinPredictedHappenedn
What it takes to reach Probation
  • Not yet: Shadow intents with a measured close >= 50 (now: 0)
  • Not yet: Shadow CLV beat rate > 50% (now: n/a)

Serie A · Total

Shadow + live Shadow Logged only. No money.

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

Serie A · Total reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier n/a, market Brier n/a; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted No settled candidates yet
Model (circle) Market price (square)

Brier: model n/a vs market n/a · slope n/a · n=0

Chart as a table
SourceBinPredictedHappenedn
What it takes to reach Probation
  • Not yet: Shadow intents with a measured close >= 50 (now: 0)
  • Not yet: Shadow CLV beat rate > 50% (now: n/a)

MLB · Run line maker (model -1.5)

Maker Maker · live 3-contract post-only buys on the model's -1.5 run-line pick, resting at the best bid at or below the market consensus no-vig price; never crosses the spread, cancelled 10 minutes before start. Drops to shadow automatically after 30 fills if fills do not beat the close.

Maker fills · measured

Orders finished / filled2 / 0 (0.0%)
Median time to filln/a
Fill price vs fair at filln/a n=0
Adverse selection, +5 minn/a n=0
Adverse selection, +30 minn/a n=0
Adverse selection, at closen/a n=0
Beat the Kalshi close (mid)n/a avg n/a, n=0
Realized P&L$+0.00 ROI n/a, n=0

Auto-control: insufficient_fills.

Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Maker quote33100.0% n=3+12.00pp3-25.0pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close3 / 3
Beat the close100.0%95% CI 43.9% to 100.0%
Average CLV+12.00pp95% CI +1.82pp to +22.18pp
Realized edge-25.0pp95% CI -32.8pp to -17.2pp
Win rate vs avg price0.0% vs 25¢
Flat-unit ROI (after fees)-100.0%95% CI -100.0% to -100.0%

Last 30 days: beat the close 100.0% (n=3), realized edge -25.0pp (n=3).

Calibration · all candidates

MLB · Run line maker (model -1.5) reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.110, market Brier 0.110; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 28%, happened 0% (n=1) Market: priced 36%, happened 0% (n=2) Model: said 28%, happened 0% (n=1) Model: said 36%, happened 0% (n=2)
Model (circle) Market price (square)

Brier: model 0.110 vs market 0.110 · slope n/a · n=3

Chart as a table
SourceBinPredictedHappenedn
Model20-30%28.0%0.0%1
Model30-40%35.5%0.0%2
Market20-30%28.0%0.0%1
Market30-40%35.5%0.0%2

MLB · Moneyline maker

Maker Maker · shadow 2-contract post-only buys on the side our posterior prefers, resting at least 1pp below the market consensus no-vig price; never crosses the spread, cancelled 10 minutes before start. Drops to shadow automatically after 30 fills if fills do not beat the close.

Maker fills · measured

Orders finished / filled0 / 0 (n/a)
Median time to filln/a
Fill price vs fair at filln/a n=0
Adverse selection, +5 minn/a n=0
Adverse selection, +30 minn/a n=0
Adverse selection, at closen/a n=0
Beat the Kalshi close (mid)n/a avg n/a, n=0
Realized P&L$+0.00 ROI n/a, n=0

Auto-control: insufficient_fills.

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

MLB · Moneyline maker reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier n/a, market Brier n/a; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted No settled candidates yet
Model (circle) Market price (square)

Brier: model n/a vs market n/a · slope n/a · n=0

Chart as a table
SourceBinPredictedHappenedn

NHL · Puck line maker

Maker Maker · live 2-contract post-only buys on the side our posterior prefers, resting at least 0.5pp below the market consensus no-vig price; never crosses the spread, cancelled 10 minutes before start. Drops to shadow automatically after 30 fills if fills do not beat the close.

Maker fills · measured

Orders finished / filled153 / 18 (11.8%)
Median time to fill1036 min
Fill price vs fair at fill+0.39pp95% CI +0.24pp to +0.55pp n=18
Adverse selection, +5 min+0.38pp95% CI +0.17pp to +0.59pp n=18
Adverse selection, +30 min+0.25pp95% CI +0.00pp to +0.50pp n=18
Adverse selection, at close-0.11pp95% CI -0.79pp to +0.56pp n=18
Beat the Kalshi close (mid)50.0%95% CI 29.0% to 71.0% avg -0.11pp, n=18
Realized P&L$-1.96 ROI -16.6%, n=17

Auto-control: insufficient_fills.

Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Maker quote17417479.3% n=174+1.89pp172-5.9pp

Live · real fills

Settled / with close16 / 18
Beat the close50.0%95% CI 29.0% to 71.0%
Average CLV+0.39pp95% CI -0.29pp to +1.06pp
Realized edge-4.4pp95% CI -27.3pp to +18.6pp
Win rate vs avg price31.2% vs 36¢
Flat-unit ROI (after fees)-16.4%95% CI -80.5% to 47.7%

Last 30 days: beat the close 50.0% (n=18), realized edge -4.4pp (n=16).

Shadow · would have bet

Settled / with close172 / 174
Beat the close79.3%95% CI 72.7% to 84.7%
Average CLV+1.89pp95% CI +1.62pp to +2.16pp
Realized edge-5.9pp95% CI -12.9pp to +1.0pp
Win rate vs avg price40.7% vs 47¢
Flat-unit ROI (after fees)-17.1%95% CI -33.5% to -0.8%

Last 30 days: beat the close 79.3% (n=174), realized edge -5.9pp (n=172).

Calibration · all candidates

NHL · Puck line maker reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.219, market Brier 0.219; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 29%, happened 29% (n=17) Market: priced 34%, happened 23% (n=57) Market: priced 44%, happened 31% (n=13) Market: priced 56%, happened 48% (n=31) Market: priced 65%, happened 63% (n=51) Market: priced 71%, happened 33% (n=3) Model: said 29%, happened 29% (n=17) Model: said 34%, happened 23% (n=57) Model: said 44%, happened 31% (n=13) Model: said 56%, happened 48% (n=31) Model: said 65%, happened 62% (n=50) Model: said 71%, happened 50% (n=4)
Model (circle) Market price (square)

Brier: model 0.219 vs market 0.219 · slope 1.18 · n=172

Chart as a table
SourceBinPredictedHappenedn
Model20-30%28.8%29.4%17
Model30-40%33.7%22.8%57
Model40-50%43.7%30.8%13
Model50-60%55.7%48.4%31
Model60-70%64.8%62.0%50
Model70-80%70.6%50.0%4
Market20-30%28.8%29.4%17
Market30-40%33.7%22.8%57
Market40-50%43.7%30.8%13
Market50-60%55.7%48.4%31
Market60-70%64.9%62.7%51
Market70-80%70.8%33.3%3

NHL · Moneyline maker

Maker Maker · shadow 2-contract post-only buys on the side our posterior prefers, resting at least 1pp below the market consensus no-vig price; never crosses the spread, cancelled 10 minutes before start. Drops to shadow automatically after 30 fills if fills do not beat the close.

Maker fills · measured

Orders finished / filled0 / 0 (n/a)
Median time to filln/a
Fill price vs fair at filln/a n=0
Adverse selection, +5 minn/a n=0
Adverse selection, +30 minn/a n=0
Adverse selection, at closen/a n=0
Beat the Kalshi close (mid)n/a avg n/a, n=0
Realized P&L$+0.00 ROI n/a, n=0

Auto-control: insufficient_fills.

Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Maker quote17417475.3% n=174+1.84pp172+5.7pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close172 / 174
Beat the close75.3%95% CI 68.4% to 81.1%
Average CLV+1.84pp95% CI +1.53pp to +2.15pp
Realized edge+5.7pp95% CI -1.4pp to +12.8pp
Win rate vs avg price57.0% vs 51¢
Flat-unit ROI (after fees)7.3%95% CI -7.4% to 21.9%

Last 30 days: beat the close 75.3% (n=174), realized edge +5.7pp (n=172).

Calibration · all candidates

NHL · Moneyline maker reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.226, market Brier 0.226; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 26%, happened 0% (n=3) Market: priced 36%, happened 35% (n=17) Market: priced 45%, happened 50% (n=38) Market: priced 55%, happened 56% (n=70) Market: priced 64%, happened 74% (n=39) Market: priced 75%, happened 100% (n=5) Model: said 26%, happened 0% (n=3) Model: said 36%, happened 35% (n=17) Model: said 45%, happened 50% (n=38) Model: said 55%, happened 56% (n=70) Model: said 64%, happened 74% (n=39) Model: said 75%, happened 100% (n=5)
Model (circle) Market price (square)

Brier: model 0.226 vs market 0.226 · slope 1.64 · n=172

Chart as a table
SourceBinPredictedHappenedn
Model20-30%26.5%0.0%3
Model30-40%36.1%35.3%17
Model40-50%45.3%50.0%38
Model50-60%54.7%55.7%70
Model60-70%64.0%74.4%39
Model70-80%74.6%100.0%5
Market20-30%26.5%0.0%3
Market30-40%36.1%35.3%17
Market40-50%45.3%50.0%38
Market50-60%54.7%55.7%70
Market60-70%64.0%74.4%39
Market70-80%74.6%100.0%5

MLB · Sharp price lag

Sharp lag Sharp lag · shadow $5 taker bets (max 3% of bankroll) when the Kalshi ask plus fee is at least 2.5pp below the market consensus no-vig price on the identical market. Drops to shadow automatically when the last 30 fired candidates beat the close under 45% of the time, or average CLV is below zero with its 95% CI under zero.
Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Sharp price lag29048.3% n=29+0.48pp29+2.2pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

MLB · Sharp price lag reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.243, market Brier 0.243; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 36%, happened 67% (n=6) Market: priced 46%, happened 33% (n=9) Market: priced 54%, happened 50% (n=10) Market: priced 66%, happened 75% (n=4) Model: said 36%, happened 67% (n=6) Model: said 46%, happened 33% (n=9) Model: said 54%, happened 50% (n=10) Model: said 66%, happened 75% (n=4)
Model (circle) Market price (square)

Brier: model 0.243 vs market 0.243 · slope n/a · n=29

Chart as a table
SourceBinPredictedHappenedn
Model30-40%36.0%66.7%6
Model40-50%46.0%33.3%9
Model50-60%53.7%50.0%10
Model60-70%65.6%75.0%4
Market30-40%36.0%66.7%6
Market40-50%46.0%33.3%9
Market50-60%53.7%50.0%10
Market60-70%65.6%75.0%4

NHL · Sharp price lag

Sharp lag Sharp lag · shadow $5 taker bets (max 3% of bankroll) when the Kalshi ask plus fee is at least 2.5pp below the market consensus no-vig price on the identical market. Drops to shadow automatically when the last 30 fired candidates beat the close under 45% of the time, or average CLV is below zero with its 95% CI under zero.
Shadow candidates by scanner (all time)
ScannerCandidatesWould betBeat closeAvg CLVSettledRealized edge
Sharp price lag463041.7% n=463+0.22pp458-5.2pp

Live · real fills

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Shadow · would have bet

Settled / with close0 / 0
Beat the closen/a
Average CLVn/a
Realized edgen/a
Win rate vs avg pricen/a vs n/a
Flat-unit ROI (after fees)n/a

Last 30 days: beat the close n/a (n=0), realized edge n/a (n=0).

Calibration · all candidates

NHL · Sharp price lag reliability diagram Predicted probability against how often it happened, for DDN SI and for the market price. Points on the dashed diagonal are perfectly calibrated. DDN SI Brier 0.231, market Brier 0.231; lower is better. 0% 0% 25% 25% 50% 50% 75% 75% 100% 100% Predicted Market: priced 28%, happened 25% (n=16) Market: priced 35%, happened 25% (n=71) Market: priced 46%, happened 40% (n=138) Market: priced 54%, happened 49% (n=128) Market: priced 65%, happened 63% (n=93) Market: priced 72%, happened 83% (n=12) Model: said 28%, happened 25% (n=16) Model: said 35%, happened 25% (n=71) Model: said 46%, happened 40% (n=138) Model: said 54%, happened 49% (n=128) Model: said 65%, happened 63% (n=93) Model: said 72%, happened 83% (n=12)
Model (circle) Market price (square)

Brier: model 0.231 vs market 0.231 · slope 1.31 · n=458

Chart as a table
SourceBinPredictedHappenedn
Model20-30%28.0%25.0%16
Model30-40%34.7%25.4%71
Model40-50%46.5%39.9%138
Model50-60%54.0%49.2%128
Model60-70%65.2%63.4%93
Model70-80%71.6%83.3%12
Market20-30%28.0%25.0%16
Market30-40%34.7%25.4%71
Market40-50%46.5%39.9%138
Market50-60%54.0%49.2%128
Market60-70%65.2%63.4%93
Market70-80%71.6%83.3%12

Recent intents

Every row, wins and losses alike. Click a date for the full record (decision snapshot, entry, close, result, versions). CLV is the closing price minus our entry price; positive means we got a better price than the market closed at. * marks a close read just after the start or from the last trade.

DateResultGameSide EntryCloseCLVTypeLane
Nov 01 PENDING Edmonton Oilers @ New York Rangers Edmonton Oilers +1.5 69¢ 68¢ -1.0pp Shadow (rejected) NHL:spread
Nov 01 PENDING Edmonton Oilers @ New York Rangers New York Rangers -1.5 33¢ 33¢ +0.0pp Shadow (rejected) NHL:spread
Nov 01 PENDING Edmonton Oilers @ New York Rangers New York R ML 52¢ 55¢ +3.0pp Shadow (would bet) NHL:moneyline_maker
Nov 01 PENDING Edmonton Oilers @ New York Rangers Over 6.5 52¢ 51¢ -1.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 PENDING Edmonton Oilers @ New York Rangers not Edmonton ML 52¢ 54¢ +2.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 PENDING Edmonton Oilers @ New York Rangers New York R -1.5 34¢ 33¢ -1.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 PENDING Edmonton Oilers @ New York Rangers New York R not -1.5 66¢ 68¢ +2.0pp Shadow (would bet) NHL:spread_maker
Nov 01 PENDING Utah Mammoth @ Philadelphia Flyers Philadelphia ML 39¢ 39¢ +0.0pp Shadow (would bet) NHL:moneyline_maker
Nov 01 PENDING Utah Mammoth @ Philadelphia Flyers Philadelphia Flyers +1.5 64¢ 63¢ -1.0pp Shadow (rejected) NHL:spread
Nov 01 PENDING Utah Mammoth @ Philadelphia Flyers Utah Mammoth -1.5 37¢ 38¢ +1.0pp Shadow (rejected) NHL:spread
Nov 01 PENDING Utah Mammoth @ Philadelphia Flyers Utah not -1.5 64¢ 63¢ -1.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 PENDING Utah Mammoth @ Philadelphia Flyers not Utah ML 41¢ 39¢ -2.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 PENDING Utah Mammoth @ Philadelphia Flyers Utah -1.5 35¢ 38¢ +3.0pp Shadow (would bet) NHL:spread_maker
Nov 01 LOSS Minnesota Wild @ Washington Capitals not Minnesota ML 50¢ 48¢ -2.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 WIN Minnesota Wild @ Washington Capitals Minnesota Wild -1.5 29¢ 31¢ +2.0pp Shadow (rejected) NHL:spread
Nov 01 WIN Minnesota Wild @ Washington Capitals Minnesota ML 49¢ 53¢ +4.0pp Shadow (would bet) NHL:moneyline_maker
Nov 01 LOSS Minnesota Wild @ Washington Capitals Washington Capitals +1.5 71¢ 70¢ -1.0pp Shadow (rejected) NHL:spread
Nov 01 WIN Minnesota Wild @ Washington Capitals Minnesota -1.5 27¢ 31¢ +4.0pp Shadow (would bet) NHL:spread_maker
Nov 01 LOSS Minnesota Wild @ Washington Capitals Minnesota not -1.5 71¢ 70¢ -1.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 LOSS Minnesota Wild @ Washington Capitals Washington Capitals -1.5 29¢ 29¢ +0.0pp Shadow (rejected) NHL:spread
Nov 01 LOSS Winnipeg Jets @ Carolina Hurricanes not Carolina ML 28¢ 28¢ +0.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 LOSS Winnipeg Jets @ Carolina Hurricanes Carolina not -1.5 50¢ 50¢ +0.0pp Shadow (would bet) NHL:spread_maker
Nov 01 LOSS Winnipeg Jets @ Carolina Hurricanes Carolina not -1.5 52¢ 50¢ -2.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 LOSS Winnipeg Jets @ Carolina Hurricanes Winnipeg Jets +1.5 53¢ 50¢ -3.0pp Shadow (rejected) NHL:spread
Nov 01 WIN Winnipeg Jets @ Carolina Hurricanes Carolina ML 69¢ 73¢ +4.0pp Shadow (would bet) NHL:moneyline_maker
Nov 01 LOSS Calgary Flames @ Colorado Avalanche Calgary Flames +1.5 43¢ 41¢ -2.0pp Shadow (rejected) NHL:spread
Nov 01 WIN Calgary Flames @ Colorado Avalanche Over 6.5 50¢ 50¢ +0.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 LOSS Calgary Flames @ Colorado Avalanche Colorado not -1.5 38¢ 41¢ +3.0pp Shadow (would bet) NHL:spread_maker
Nov 01 LOSS Calgary Flames @ Colorado Avalanche Colorado not -2.5 53¢ 55¢ +2.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 WIN Calgary Flames @ Colorado Avalanche Colorado ML 78¢ 79¢ +1.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 WIN Calgary Flames @ Colorado Avalanche Colorado ML 76¢ 79¢ +3.0pp Shadow (would bet) NHL:moneyline_maker
Nov 01 LOSS Columbus Blue Jackets @ Dallas Stars Dallas not -1.5 61¢ 61¢ +0.0pp Shadow (would bet) NHL:spread_maker
Nov 01 LOSS Columbus Blue Jackets @ Dallas Stars Over 5.5 48¢ 49¢ +1.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 LOSS Columbus Blue Jackets @ Dallas Stars Columbus Blue Jackets +1.5 62¢ 61¢ -1.0pp Shadow (rejected) NHL:spread
Nov 01 WIN Columbus Blue Jackets @ Dallas Stars Dallas ML 60¢ 63¢ +3.0pp Shadow (would bet) NHL:moneyline_maker
Nov 01 WIN Columbus Blue Jackets @ Dallas Stars Dallas Stars -1.5 38¢ 41¢ +3.0pp Shadow (rejected) NHL:spread
Nov 01 WIN Columbus Blue Jackets @ Dallas Stars Dallas -1.5 38¢ 41¢ +3.0pp Shadow (rejected) NHL:sharp_lag
Nov 01 LOSS Columbus Blue Jackets @ Dallas Stars not Dallas ML 38¢ 38¢ +0.0pp Shadow (rejected) NHL:sharp_lag
Oct 31 LOSS Pittsburgh Penguins @ Montreal Canadiens Montreal Canadiens -1.5 37¢ 33¢ -4.0pp Shadow (rejected) NHL:spread
Oct 31 LOSS Toronto Maple Leafs @ Vancouver Canucks Vancouver Canucks +1.5 62¢ 58¢ -4.0pp Shadow (rejected) NHL:spread
Oct 29 WIN New Jersey Devils @ Colorado Avalanche COL:yes 38¢ 36¢ -2.0pp Live NHL:spread_maker
Oct 28 LOSS Minnesota Wild @ Utah Mammoth UTA:yes 30¢ 31¢ +1.0pp Live NHL:spread_maker
Oct 28 LOSS Montreal Canadiens @ St. Louis Blues STL:yes 28¢ 28¢ +0.0pp Live NHL:spread_maker
Oct 27 LOSS Detroit Red Wings @ Edmonton Oilers EDM:yes 41¢ 41¢ +0.0pp Live NHL:spread_maker
Oct 25 LOSS Montreal Canadiens @ Winnipeg Jets MTL:yes 29¢ 30¢ +1.0pp Live NHL:spread_maker
Oct 25 WIN New York Islanders @ Dallas Stars DAL:yes 38¢ 38¢ +0.0pp Live NHL:spread_maker
Oct 22 LOSS Los Angeles Kings @ New York Islanders NYI:yes 27¢ 28¢ +1.0pp Live NHL:spread_maker
Oct 22 WIN Nashville Predators @ Boston Bruins BOS:yes 35¢ 36¢ +1.0pp Live NHL:spread_maker
Oct 22 PENDING Florida Panthers @ Pittsburgh Penguins FLA:yes 28¢ 30¢ +2.0pp Live NHL:spread_maker
Oct 21 WIN Carolina Hurricanes @ Vancouver Canucks CAR:yes 52¢ 52¢ +0.0pp Live NHL:spread_maker
Oct 20 LOSS Colorado Avalanche @ New Jersey Devils COL:yes 30¢ 28¢ -2.0pp Live NHL:spread_maker
Oct 20 LOSS Los Angeles Kings @ Washington Capitals WSH:yes 35¢ 37¢ +2.0pp Live NHL:spread_maker
Oct 19 LOSS Anaheim Ducks @ New York Rangers NYR:no 64¢ 64¢ +0.0pp Live NHL:spread_maker
Oct 17 LOSS New York Islanders @ Toronto Maple Leafs TOR:yes 32¢ 33¢ +1.0pp Live NHL:spread_maker
Oct 16 LOSS Utah Mammoth @ Edmonton Oilers EDM:yes 31¢ 29¢ -2.0pp Live NHL:spread_maker
Oct 13 LOSS Florida Panthers @ Columbus Blue Jackets CBJ:yes 30¢ 33¢ +3.0pp Live NHL:spread_maker
Oct 11 WIN Anaheim Ducks @ Calgary Flames ANA:yes 30¢ 32¢ +2.0pp Live NHL:spread_maker
Oct 10 PENDING Toronto Maple Leafs @ Colorado Avalanche COL:no 56¢ 55¢ -1.0pp Live NHL:spread_maker

How to read this

  • Beat the close (CLV) is the fastest honest signal. If we buy at 45¢ and the market closes at 50¢, we got a better price than the market's final view. A lane that can't beat the close over many bets doesn't have edge, whatever its record says.
  • Calibration: when DDN SI says 60%, it should happen about 60% of the time. Points on the dashed line are calibrated. Brier score is the average squared miss; if DDN SI's Brier isn't lower than the market's, DDN SI isn't adding information.
  • Realized edge is win rate minus the average price paid. A 95% confidence interval that crosses zero means we can't tell it apart from luck yet.
  • Stages: Shadow (no money) → Probation (minimum bets, after shadow beats the close on 50+ intents) → Certified (150+ clean bets, beats the close with confidence, calibrated, edge above zero). Small samples move nothing.
  • Results are past results. They do not promise anything about future ones.

See every published pick and result →   Raw data (JSON) →   Daily tamper-evident ledger →

21+ only. For entertainment and information; not financial advice. Gambling problem? Call 1-800-GAMBLER.