The goals specialist. The Goals Trader only trades the total when the model expected goals clearly disagree with the price, and when the two sides are close it defends, leaning Under. It openly knows the goals market is brutally efficient, so it passes often.

Every call it has made · 5

• MISSFrosinone vs Venezia - under_2.52.00−€10.00
• HITManchester City vs Coventry - under_2.53.20+€22.00
• HITIpswich vs Liverpool - under_2.53.00+€20.00
• HITVolos NFC vs Olympiakos - under_2.52.15+€11.50
• HITLommel SK vs Club Brugge - under_2.53.50+€25.00

Open right now

• OPENFor Sittard vs Ajax - under_2.53.102026-09-12
• OPENTwente vs Den Haag - under_2.53.502026-09-12
• OPENLevante vs Barcelona - under_2.53.752026-09-13
• OPENReal Madrid vs Vallecano - under_2.53.752026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysOver/Under 2.5 goals

Odds bandabout 1.40 to 3.20

RiskLow

Stake1% flat

LimitUp to 4 per round

How it picks

• Needs both Over and Under priced (a complete market)
• Bets only on a clear deviation with real conviction
• When it is a coin flip, it prefers Under

Where its edge would be

A low tempo game priced like a goal fest, or a shoot out priced like a stalemate.

Honest weakness

Its own research found the goals closing line cannot be beaten: expect a return near zero and many passes. The Under lean can cost in high scoring spells.

How it improves

It tracks return on each side. A side that keeps losing gets a tighter threshold. It has self-tuned its parameters 2 times this season as it learned from results.

Replay bankrollvs €1,000 start

104.4u104%

This round, open paper pick

The cynic. The Value Hunter ignores form, narratives and hype. It cares about one thing only: when the model price is clearly better than the market price. The only edge it trusts is a measurable gap.

Its last calls

• HITEstrela vs Sp Braga - over_2.52.00+€10.00
• HITEstrela vs Sp Braga - home4.75+€37.50
• MISSPanathinaikos vs Kifisia - draw5.00−€10.00
• MISSCorum vs Eyupspor - away4.75−€10.00
• HITBurnley vs Bristol City - away4.00+€30.00
• MISSOviedo vs Burgos - away4.33−€10.00

Open right now

• OPENAtalanta vs Cagliari - away6.002026-09-12
• OPENOlympiakos vs OFI Crete - draw6.502026-09-12
• OPENChelsea vs Hull - draw6.502026-09-12
• OPENReal Madrid vs Vallecano - under_2.53.752026-09-12
• OPENEyupspor vs Rizespor - home3.902026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result and Over/Under 2.5 goals

Odds band1.50 to 7.00

RiskMedium

Stake1% flat

LimitUp to 8 per round

How it picks

• Bets only when model probability times odds beats its threshold (about 3% and up)
• Stays inside the proven odds range 1.50 to 7.00
• Skips long shots above 7.00, where variance turns ruinous

Where its edge would be

A real disagreement between the model and the closing price, in the band where value has historically been most reliable.

Honest weakness

On public data most apparent gaps are noise: the model rarely beats the closing line. The threshold can chase that noise up and down.

How it improves

It tracks realised return. Losing makes it tighten the threshold, winning lets it loosen a little. It has self-tuned its parameters 2 times this season as it learned from results.

Replay bankrollvs €1,000 start

100.3u100%

This round, open paper pick

The staking experiment. The Compounder finds value exactly like the Value Hunter, but instead of a flat stake it sizes each bet by a quarter of the Kelly fraction. It is an honest head to head: does Kelly staking beat flat staking on the very same edges, or does the extra variance cost more than it gains?

Its last calls

• MISSPanathinaikos vs Kifisia - under_2.52.05−€10.00
• HITEstrela vs Sp Braga - over_2.52.00+€10.00
• HITEstrela vs Sp Braga - home4.75+€37.50
• MISSPanathinaikos vs Kifisia - draw5.00−€10.00
• MISSCorum vs Eyupspor - away4.75−€10.00
• HITBurnley vs Bristol City - away4.00+€30.00

Open right now

• OPENAtalanta vs Cagliari - away6.002026-09-12
• OPENOlympiakos vs OFI Crete - draw6.502026-09-12
• OPENChelsea vs Hull - draw6.502026-09-12
• OPENReal Madrid vs Vallecano - under_2.53.752026-09-12
• OPENEyupspor vs Rizespor - home3.902026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result and Over/Under 2.5 goals

Odds band1.50 to 7.00

RiskMedium

Stakequarter-Kelly, capped at 2%

LimitUp to 8 per round

How it picks

• Same value rule: model probability times odds beats its threshold
• Stays in the proven odds range 1.50 to 7.00
• Stakes a quarter of the Kelly fraction, hard capped at 2%

Where its edge would be

The same measurable model versus market gap, but with the stake scaled to the size of the edge.

Honest weakness

Kelly amplifies variance and is sensitive to model overconfidence, so a bad streak bites harder.

How it improves

Realised return adjusts the value threshold. The Kelly fraction itself stays fixed. It has self-tuned its parameters 2 times this season as it learned from results.

Replay bankrollvs €1,000 start

100.6u101%

This round, open paper pick

The cautious underdog backer. Instead of a raw outsider it takes the draw-protected version, trading upside for a softer variance profile. We hold no double-chance quotes, so its price is derived from the real 1X2 prices (1 divided by the sum of the two implied probabilities): the bookmaker margin is kept, nothing is invented, and the profile says so.

Every call it has made · 5

• MISSCorum vs Eyupspor - dc_away_or_draw2.02−€10.00
• HITBurnley vs Bristol City - dc_away_or_draw1.90+€8.95
• HITOviedo vs Burgos - dc_away_or_draw1.86+€8.57
• MISSGent vs Oud-Heverlee Leuven - dc_away_or_draw2.13−€10.00
• HITWest Brom vs Charlton - dc_away_or_draw2.02+€10.15

Open right now

• OPENAugsburg vs Leverkusen - dc_home_or_draw2.052026-09-12
• OPENMallorca vs Sabadell - dc_away_or_draw2.062026-09-13
• OPENEyupspor vs Rizespor - dc_home_or_draw1.822026-09-12
• OPENGirona vs Castellon - dc_away_or_draw1.672026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysDouble chance (protected underdog)

Odds band1.30 to 2.20

RiskMedium

Stake0.75% flat

LimitUp to 4 per round

How it picks

• Covers only the side that is NOT the 1X2 favourite
• Only when the protected price still beats its edge threshold
• Price derived from the real 1X2 quotes, margin included

Where its edge would be

A resilient underdog the market overlooks once the draw is taken out of the risk.

Honest weakness

Many passes and modest upside, and the price is derived rather than quoted.

How it improves

Realised return: losses raise the edge threshold.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The sceptic with a memory. The Council keeps a record of how well the model has actually been calibrated in each segment: this league, this market, this odds band. A positive expected value is not enough for it. If the segment has a poor calibration history it passes, and where calibration is proven it stakes more.

Its last calls

• MISSPanathinaikos vs Kifisia - under_2.52.05−€10.00
• HITEstrela vs Sp Braga - over_2.52.00+€10.00
• HITEstrela vs Sp Braga - home4.75+€37.50
• MISSPanathinaikos vs Kifisia - draw5.00−€10.00
• MISSOviedo vs Burgos - away4.33−€10.00
• HITBetis vs Real Madrid - home7.00+€60.00

Open right now

• OPENEyupspor vs Rizespor - home3.902026-09-12
• OPENSunderland vs Arsenal - home7.502026-09-12
• OPENMallorca vs Sabadell - away5.252026-09-13
• OPENFor Sittard vs Ajax - home5.752026-09-12
• OPENBenfica vs Gil Vicente - draw7.502026-09-13

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysAny priced market, filtered by segment trust

Odds band1.30 to 8.00

RiskMedium

Staketrust weighted, up to 1%

LimitUp to 8 per round

How it picks

• Scores every segment (league by market by odds band) on past calibration
• Skips a bet in an untrusted segment even when the value looks good
• Stakes are weighted by the trust score, small stakes in unknown segments
• Learns segment trust only from settled results, never from its own picks

Where its edge would be

Value is only worth taking where the probability behind it has held up before. This is the engine's own trust gate turned into a betting style.

Honest weakness

It needs a lot of settled bets before a segment counts as known, so it starts cautious and can sit out long stretches early on.

How it improves

Every settled bet updates that segment's calibration memory, which changes both what it will play and how much it stakes.

Replay bankrollvs €1,000 start

100.1u100%

This round, open paper pick

A pure goals bull. The Over Merchant backs Over 2.5 only when the model leans clearly toward goals and the price still offers value. It is the deliberate mirror image of the Under leaning Goals Trader, so you can compare the two head to head.

Every call it has made · 3

• HITTroyes vs Strasbourg - over_2.51.88+€8.80
• MISSEstoril vs Arouca - over_2.51.85−€10.00
• HITFenerbahce vs Besiktas - over_2.51.70+€7.00

Open right now

• OPENGaziantep vs Fenerbahce - over_2.51.732026-09-14
• OPENCordoba vs Almeria - over_2.51.652026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysOver 2.5 goals

Odds bandabout 1.40 to 3.20

RiskMedium

Stake1% flat

LimitUp to 4 per round

How it picks

• Backs Over 2.5 only, never the Under
• Model must clear 50/50 by a real margin (conviction)
• Only takes positive expected value

Where its edge would be

A high tempo matchup the market has priced like an average game.

Honest weakness

The goals market is brutally efficient. Our own research found no ghost edge here, so expect a return near zero.

How it improves

Realised return tightens or loosens its value bar. It has self-tuned its parameters 1 time this season as it learned from results.

Replay bankrollvs €1,000 start

100.5u101%

This round, open paper pick

The mirror of the Over Merchant. The Lockdown backs Under 2.5 only when the model sees clearly fewer goals than the price implies. Together with the Over Merchant and the Goals Trader, it lets you test whether backing Overs, Unders or trading both works best.

Every call it has made · 4

• MISSFrosinone vs Venezia - under_2.52.00−€10.00
• MISSSt Etienne vs Montpellier - under_2.52.15−€10.00
• HITVolos NFC vs Olympiakos - under_2.52.15+€11.50
• HITWest Brom vs Charlton - under_2.51.73+€7.30

Open right now

• OPENRotherham vs Salford - under_2.52.052026-09-12
• OPENColchester vs Crewe - under_2.52.002026-09-12
• OPENBlackburn vs Millwall - under_2.51.952026-09-12
• OPENComo vs Parma - under_2.52.202026-09-14

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysUnder 2.5 goals

Odds bandabout 1.40 to 3.20

RiskMedium

Stake1% flat

LimitUp to 4 per round

How it picks

• Backs Under 2.5 only, never the Over
• Model must clear 50/50 by a real margin (conviction)
• Only takes positive expected value

Where its edge would be

Two defensive teams priced like an average shoot out.

Honest weakness

Same efficient goals market, and one late goal sinks the whole bet.

How it improves

Realised return tightens or loosens its value bar. It has self-tuned its parameters 2 times this season as it learned from results.

Replay bankrollvs €1,000 start

101.2u101%

This round, open paper pick

The risk manager. The Survivor has one rule above all others: be here tomorrow. It backs only very likely outcomes at small stakes. Profit is a side effect, survival is the goal.

Every call it has made · 3

• HITSp Lisbon vs Nacional - home1.17+€1.70
• MISSSchalke 04 vs Bayern Munich - over_2.51.22−€10.00
• HITManchester City vs Coventry - home1.18+€1.80

Open right now

• OPENSt. Gilloise vs Lommel SK - home1.202026-09-12
• OPENPSV Eindhoven vs Sparta Rotterdam - over_2.51.202026-09-13
• OPENElversberg vs Bayern Munich - over_2.51.172026-09-13

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result: heavy favourites only

Odds bandabout 1.05 to 1.60

RiskLow

Stake0.5%

LimitUp to 6 per round

How it picks

• Only picks with model probability 70% and up
• Quarter stake (0.5%), one pick per match

Where its edge would be

Maximum hit rate, minimum variance. For this style the probability is the value.

Honest weakness

Short odds cap the upside, and a single upset wipes out many small wins.

How it improves

Any meaningful drawdown tightens its probability bar even further. It has self-tuned its parameters 1 time this season as it learned from results.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The cautious one. The Banker backs a clear favourite only when the model and the bookmakers fully agree, and only at short odds where the result is rarely in doubt. It would happily sit out a whole round rather than touch a coin flip.

Every call it has made · 1

• MISSVillarreal vs La Coruna - home1.50−€10.00

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result: home or away win

Odds band1.30 to 1.75

RiskLow

Stake1% flat

LimitUp to 3 per round

How it picks

• Backs only strong favourites at odds 1.30 to 1.75
• Model and market must agree (gap under 8 points)
• Never takes a bet with negative expected value
• Maximum 3 bets per round

Where its edge would be

A favourite the market has priced fairly, not over shortened. Small, repeatable edges across many safe bets.

Honest weakness

Short odds mean the bookmaker margin eats returns fast, and it is blind to value outside the favourite band.

How it improves

It watches its own hit rate. If it wins less often than it promised, it raises the bar and gets even more selective. It has self-tuned its parameters 1 time this season as it learned from results.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

No bet this round, and that is a decision.
Its reason: no clear favourite at 1.30 to 1.75 where the model and the market agree with positive expected value.

Not a fun parlay. The Accumulator builds exactly three legs, each one positive value on its own, from three different matches, with the combined price kept inside a tight band. The edge has to come from selective combining, never from volume. Most rounds it builds nothing.

Every call it has made · 2

• MISSVillarreal vs La Coruna - home + Fenerbahce vs Besiktas - over_2.5 + Parma vs Monza - under_2.54.13−€10.00
• MISSVillarreal vs La Coruna - home + Fenerbahce vs Besiktas - over_2.5 + Kocaelispor vs Samsunspor - under_2.54.21−€10.00

Open right now

• OPENCordoba vs Almeria - over_2.5 + Gaziantep vs Fenerbahce - over_2.5 + Excelsior vs Utrecht - over_2.54.572026-09-12
• OPENCordoba vs Almeria - over_2.5 + Clermont vs Boulogne - under_2.5 + Gaziantep vs Fenerbahce - over_2.54.712026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result and goals, combined into a treble

Odds band2.50 to 7.00 combined

RiskHigh

Stake1% per coupon

LimitUp to 1 coupon per round

How it picks

• Exactly 3 legs, each one positive value on its own
• Three different matches (never two legs from one game)
• Combined odds must land between 2.50 and 7.00, or no bet

Where its edge would be

Three clean, independent reads whose combined price sits in the sweet spot.

Honest weakness

Variance and the coupon margin climb fast with every leg, so it has to stay rare and strict.

How it improves

Coupon hit rate. Too few hits and it demands stronger legs. It has self-tuned its parameters 1 time this season as it learned from results.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The house style in person. The Sushi Chef builds short, clean coupons: exactly two legs, combined odds between 1.50 and 2.00, made from the most likely picks of the round. Short over long, every single time.

Every call it has made · 6

• HITManchester City vs Coventry - home + Getafe vs Celta - under_2.51.65+€6.52
• MISSSp Lisbon vs Nacional - home + Villarreal vs La Coruna - home1.76−€10.00
• MISSSchalke 04 vs Bayern Munich - over_2.5 + Ajax vs PSV Eindhoven - over_2.51.59−€10.00
• MISSManchester City vs Coventry - home + Villarreal vs La Coruna - home1.77−€10.00
• HITSp Lisbon vs Nacional - home + Ajax vs PSV Eindhoven - over_2.51.52+€5.21
• MISSSchalke 04 vs Bayern Munich - over_2.5 + Stuttgart vs FC Koln - over_2.51.66−€10.00

Open right now

• OPENSt. Gilloise vs Lommel SK - home + Dortmund vs Paderborn - home1.502026-09-12
• OPENElversberg vs Bayern Munich - over_2.5 + Cambuur vs Nijmegen - over_2.51.592026-09-13
• OPENPSV Eindhoven vs Sparta Rotterdam - over_2.5 + Hoffenheim vs Stuttgart - over_2.51.632026-09-13

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result and goals, combined into a coupon

Odds band1.50 to 2.00 combined

RiskMedium

Stake1% per coupon

LimitUp to 3 coupons per round

How it picks

• Always exactly 2 legs (the Double Sushi rule, hard)
• Combined odds must land between 1.50 and 2.00
• Each leg has to clear the probability bar
• Never two legs from the same match

Where its edge would be

Two high probability picks whose combined price sits in the sweet spot. Lower variance than a long dream accumulator.

Honest weakness

A parlay multiplies the bookmaker margin (you pay the vig twice), and it depends on the round offering the right matches.

How it improves

It watches the coupon hit rate. Too few hits and it demands stronger legs.

Replay bankrollvs €1,000 start

100.6u101%

This round, open paper pick

The skeptic. The Contrarian stays silent unless the market is clearly more confident in a favourite than the model can justify. When the crowd over backs a favourite, the alternatives are underpriced. The rest of the time: no bet. Discipline is the product.

Every call it has made · 3

• MISSPanathinaikos vs Kifisia - away11.00−€10.00
• MISSCrewe vs York - home2.82−€10.00
• MISSSp Gijon vs Girona - home2.75−€10.00

Open right now

• OPENAmedspor vs Buyuksehyr - away2.552026-09-13
• OPENGirona vs Castellon - away3.202026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysThe value alternative to an over-backed favourite

Odds band2.50 to 15.00

RiskMedium

Stake1% flat

LimitUp to 2 per round

How it picks

• Needs the market 10 points or more above the model on the favourite
• Backs the best value alternative, never the favourite itself
• At most 2 per round, otherwise no bet

Where its edge would be

A favourite pushed below fair value by public money.

Honest weakness

Often the market is right and the model is wrong. Very few bets means a slow verdict.

How it improves

Losses raise its overconfidence requirement, which means even more silence.

Replay bankrollvs €1,000 start

98.0u98%

This round, open paper pick

The maximalist. The Daredevil chases the single biggest edge of the round, whatever the odds. It exists on purpose: as a stress test for the ruin guard, and an honest benchmark for what pure aggression actually costs.

Its last calls

• HITEstrela vs Sp Braga - home4.75+€37.50
• MISSPanathinaikos vs Kifisia - away11.00−€10.00
• MISSValencia vs Barcelona - home10.00−€10.00
• MISSWolfsburg vs Cottbus - away8.00−€10.00
• MISSWest Ham vs Derby - away7.50−€10.00
• MISSToulouse vs Lille - home3.25−€10.00

Open right now

• OPENInter vs Udinese - away12.002026-09-14
• OPENReal Madrid vs Vallecano - draw9.502026-09-12
• OPENChelsea vs Hull - away11.002026-09-12
• OPENComo vs Parma - away15.002026-09-14
• OPENWest Ham vs Wrexham - away5.252026-09-11

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysAny market with the biggest edge

Odds bandany (chases the edge)

RiskHigh

Stake2% wanted, hard-capped

LimitUp to 5 per round

How it picks

• Takes the top expected-value pick of the round
• Wants a 2% stake (the ruin guard hard-caps it)
• Up to 5 picks per round

Where its edge would be

The largest edge available, anywhere on the board.

Honest weakness

Maximum exposure to noise and a magnet for drawdown. The ruin guard is its backbone, not its choice.

How it improves

Sustained losses force its edge bar up: discipline by punishment. It has self-tuned its parameters 2 times this season as it learned from results.

Replay bankrollvs €1,000 start

95.1u95%

This round, open paper pick

The thrill seeker, on a leash. The Longshot hunts undervalued outsiders at odds of 4 to 12, where public money over backs favourites. But it knows long shot value is fragile, so the bar is high and the stake is half size.

Its last calls

• MISSPanathinaikos vs Kifisia - draw5.00−€10.00
• MISSPanathinaikos vs Kifisia - away11.00−€10.00
• MISSWolfsburg vs Cottbus - away8.00−€10.00
• MISSWest Ham vs Derby - away7.50−€10.00
• MISSGent vs Oud-Heverlee Leuven - away5.00−€10.00
• MISSAnderlecht vs Kortrijk - away5.50−€10.00

Open right now

• OPENChelsea vs Hull - away11.002026-09-12
• OPENWest Ham vs Wrexham - away5.252026-09-11
• OPENBrest vs Paris SG - home11.002026-09-13

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysUnderdog win or draw at high odds

Odds band4.00 to 12.00

RiskHigh

Stake0.5% (half unit)

LimitUp to 2 per round

How it picks

• Odds between 4.00 and 12.00 only
• Demands a big edge (10% and up) before backing
• Half stake (0.5%), maximum 2 per round

Where its edge would be

A real chance the market has priced at almost nothing.

Honest weakness

The favourite-longshot bias is stubborn, and long losing streaks are normal here. Hit rate alone will mislead you.

How it improves

Drawdown discipline: losing streaks raise its edge bar faster than wins lower it.

Replay bankrollvs €1,000 start

99.0u99%

This round, open paper pick

The niche scout. The Corner Man works the corner and card markets, which are often the most mispriced. But we hold no real prop odds in our data, so it plays against fair odds (1 divided by probability). That means it measures one thing honestly: how well calibrated the niche model is. It can never prove a market edge, and we say so.

Its last calls

• MISSEstrela vs Sp Braga - cards_over_3.51.38−€10.00
• HITPanathinaikos vs Kifisia - cards_over_3.51.26+€2.62
• HITEspanol vs Sevilla - cards_over_3.51.40+€3.98
• MISSAEK vs Aris - cards_over_3.51.28−€10.00
• HITValladolid vs Andorra - cards_over_3.51.28+€2.76
• HITOFI Crete vs Kifisia - cards_over_3.51.25+€2.49

Open right now

• OPENValladolid vs Oviedo - cards_over_3.51.322026-09-13
• OPENKifisia vs Levadeiakos - cards_over_3.51.292026-09-13
• OPENAndorra vs Sociedad B - cards_over_3.51.282026-09-12
• OPENKalamata vs Volos NFC - cards_over_3.51.272026-09-13
• OPENLecce vs Monza - corners_under_10.51.422026-09-13

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysTotal corners and cards

Odds bandfair odds (no market price)

RiskLow (paper)

Stake1% paper

LimitUp to 4 per round

How it picks

• Only picks with clear conviction away from 50/50
• Priced at fair odds (1 divided by probability), no real market price
• Maximum 4 picks per round

Where its edge would be

Strong model conviction on a total corner or card count. A calibration signal, not a price edge.

Honest weakness

With no market odds it can never prove an edge, only calibration. Corners are weakly calibrated outside the top leagues.

How it improves

Calibration check: if its 60% picks do not land near 60%, it demands more conviction.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The shots reader. We hold real shots on target data (home plus away on target) but no shots market odds, so like the Corner Man it plays against fair odds. It measures one thing honestly: how well calibrated the shots model is. No market edge claim, ever.

Every call it has made · 5

• HITAjax vs PSV Eindhoven - sot_over_6.51.17+€1.66
• MISSTelstar vs Cambuur - sot_over_6.51.16−€10.00
• HITEin Frankfurt vs Augsburg - sot_over_6.51.23+€2.28
• HITLeverkusen vs Union Berlin - sot_over_6.51.23+€2.25
• MISSSchalke 04 vs Bayern Munich - sot_over_6.51.19−€10.00

Open right now

• OPENCambuur vs Nijmegen - sot_over_6.51.172026-09-12
• OPENGo Ahead Eagles vs Groningen - sot_over_6.51.162026-09-12
• OPENPSV Eindhoven vs Sparta Rotterdam - sot_over_6.51.162026-09-13
• OPENBayern Munich vs Union Berlin - sot_over_6.51.192026-09-18
• OPENElversberg vs Bayern Munich - sot_over_6.51.192026-09-13

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysTotal shots on target

Odds bandfair odds (no market price)

RiskLow (paper)

Stake1% paper

LimitUp to 4 per round

How it picks

• Only picks with clear conviction away from 50/50
• Priced at fair odds (1 divided by probability), no real market price
• Plays Over or Under the shots on target line

Where its edge would be

Strong model conviction on a shots on target total. A calibration signal, not a price edge.

Honest weakness

With no market odds it can never prove an edge, only calibration.

How it improves

Calibration check: if its 60% picks do not land near 60%, it demands more conviction.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The cards specialist. Our cards model already knows the referee: strictness sits inside its estimate for every match. It takes the single strongest cards line per match, Over or Under, and plays it at fair odds because we hold no cards prices. It measures whether the model's cards percentages hold up, not a market edge.

Its last calls

• MISSEstrela vs Sp Braga - cards_over_3.51.38−€10.00
• HITPanathinaikos vs Kifisia - cards_over_3.51.26+€2.62
• HITGetafe vs Celta - cards_over_3.51.46+€4.56
• MISSAlaves vs Osasuna - cards_over_3.51.45−€10.00
• HITEspanol vs Sevilla - cards_over_3.51.40+€3.98
• MISSAEK vs Aris - cards_over_3.51.28−€10.00

Open right now

• OPENValladolid vs Oviedo - cards_over_3.51.322026-09-13
• OPENKifisia vs Levadeiakos - cards_over_3.51.292026-09-13
• OPENAndorra vs Sociedad B - cards_over_3.51.282026-09-12
• OPENKalamata vs Volos NFC - cards_over_3.51.272026-09-13
• OPENSevilla vs Valencia - cards_over_3.51.502026-09-11

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysTotal cards Over/Under

Odds bandfair odds (no market price)

RiskMedium

Stake1% flat, paper

LimitUp to 4 per round

How it picks

• One line per match, the strongest, Over or Under
• Referee strictness is inside the model estimate it reads
• Fair odds, paper money: it measures calibration, not market edge

Where its edge would be

Referee and matchup effects that a naive cards average misses.

Honest weakness

Card data is noisy outside the top leagues, and without cards prices it cannot claim an edge.

How it improves

Hit rate against the percentage it promised.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The both score specialist. We hold no real BTTS odds in our data, so like the Corner Man it plays against fair odds (1 divided by probability). That means it measures one thing honestly: how well calibrated the both teams to score model is. It can never prove a market edge, and we say so plainly.

Every call it has made · 6

• MISSRennes vs Marseille - btts1.50−€10.00
• HITNijmegen vs Feyenoord - btts1.48+€4.83
• HITAjax vs PSV Eindhoven - btts1.45+€4.46
• HITEin Frankfurt vs Augsburg - btts1.55+€5.50
• MISSSchalke 04 vs Bayern Munich - btts1.55−€10.00
• HITStuttgart vs FC Koln - btts1.48+€4.77

Open right now

• OPENGo Ahead Eagles vs Groningen - btts1.492026-09-12
• OPENCordoba vs Almeria - btts1.462026-09-12
• OPENStrasbourg vs Monaco - btts1.532026-09-12
• OPENElversberg vs Bayern Munich - btts1.552026-09-13
• OPENAugsburg vs Leverkusen - btts1.532026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysBoth teams to score (yes / no)

Odds bandfair odds (no market price)

RiskLow (paper)

Stake1% paper

LimitUp to 4 per round

How it picks

• Only picks with clear conviction away from 50/50
• Priced at fair odds (1 divided by probability), no real market price
• Plays either side, Yes or No, and grades both honestly

Where its edge would be

Strong model conviction on whether both teams find the net. A calibration signal, not a price edge.

Honest weakness

With no market odds it can never prove an edge, only calibration.

How it improves

Calibration check: if its 60% picks do not land near 60%, it demands more conviction.

Replay bankrollvs €1,000 start

99.6u100%

This round, open paper pick

The dominance reader. It asks a simple question across corners, shots on target and cards: which team piles up the most? We have the count data but no prop odds, so it plays fair odds and measures calibration. Each game it backs its single strongest read.

Every call it has made · 6

• MISSMalaga vs Levante - sot_most_home1.59−€10.00
• HITTrabzonspor vs Genclerbirligi - sot_most_home1.56+€5.60
• HITSp Lisbon vs Nacional - sot_most_home1.51+€5.14
• HITStuttgart vs FC Koln - sot_most_home1.60+€6.04
• HITLeverkusen vs Union Berlin - sot_most_home1.57+€5.68
• HITWest Brom vs Charlton - corners_most_home1.55+€5.54

Open right now

• OPENGalatasaray vs Kocaelispor - sot_most_home1.462026-09-13
• OPENOlympiakos vs OFI Crete - sot_most_home1.392026-09-12
• OPENBayern Munich vs Union Berlin - sot_most_home1.392026-09-18
• OPENReal Madrid vs Vallecano - sot_most_home1.382026-09-12
• OPENDortmund vs Paderborn - corners_most_home1.562026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysTeam to have the most corners / shots / cards

Odds bandfair odds (no market price)

RiskLow (paper)

Stake1% paper

LimitUp to 4 per round

How it picks

• Reads team most corners, most shots on target and most cards
• Backs the single strongest dominance read in each match
• Priced at fair odds (1 divided by probability), no real market price

Where its edge would be

One team clearly favoured to dominate a count market. A calibration signal, not a price edge.

Honest weakness

No market odds means calibration only. A tie counts as a loss for both sides, by design.

How it improves

Calibration check: if its reads do not land near their probability, it demands more conviction.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The first-half specialist. Many teams start slow or fast in ways the full-time price misses. It reads the half-time scores we hold for every match through a dedicated first-half model. We have no first-half prices, so it plays at fair odds and measures one thing honestly: does the model's first-half percentage hold up?

Its last calls

• HITGetafe vs Celta - ht_under_1.51.23+€2.31
• MISSTelstar vs Cambuur - ht_over_0.51.19−€10.00
• HITBoulogne vs Dijon - ht_under_1.51.18+€1.84
• HITVillarreal vs La Coruna - ht_over_0.51.18+€1.83
• HITGreuther Furth vs Heidenheim - ht_over_0.51.16+€1.56
• HITStuttgart vs FC Koln - ht_over_0.51.24+€2.40

Open right now

• OPENCambuur vs Nijmegen - ht_over_0.51.182026-09-12
• OPENKarlsruhe vs Cottbus - ht_over_0.51.172026-09-13
• OPENVillarreal vs Betis - ht_over_0.51.192026-09-14
• OPENElversberg vs Bayern Munich - ht_over_0.51.222026-09-13
• OPENBayern Munich vs Union Berlin - ht_over_0.51.162026-09-18

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysHalf-time result and first-half Over/Under

Odds bandfair odds (no market price)

RiskMedium

Stake1% flat, paper

LimitUp to 3 per round

How it picks

• Takes the strongest first-half market per match from the half-time model
• Only when the model sits clearly away from 50/50
• Fair odds, paper money: it measures calibration, not market edge

Where its edge would be

Slow or fast starters mispriced because the market prices the full match, not the half.

Honest weakness

Vulnerable to early randomness, and without first-half prices it cannot claim an edge, only calibration.

How it improves

Hit rate against the percentage it promised: if it lands below, it demands more conviction.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The closer. Tired legs, chasing teams and fresh substitutes open games up after the break. It splits every match in two (full time minus half time) through a second-half model and backs Over only when the second half is expected livelier than the first. No second-half prices exist in our data, so it plays at fair odds and measures calibration.

Its last calls

• HITGetafe vs Celta - sh_under_1.51.43+€4.26
• MISSGreuther Furth vs Heidenheim - sh_over_1.51.39−€10.00
• HITPau FC vs Sochaux - sh_under_1.51.30+€2.98
• HITOviedo vs Burgos - sh_under_1.51.23+€2.29
• HITStuttgart vs FC Koln - sh_over_1.51.53+€5.34
• HITCagliari vs Lecce - sh_under_1.51.50+€5.04

Open right now

• OPENShrewsbury vs Northampton - sh_under_1.51.362026-09-12
• OPENValladolid vs Oviedo - sh_under_1.51.272026-09-13
• OPENBayern Munich vs Union Berlin - sh_over_1.51.402026-09-18
• OPENElversberg vs Bayern Munich - sh_over_1.51.352026-09-13
• OPENGetafe vs La Coruna - sh_under_1.51.312026-09-13

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysSecond-half Over/Under 1.5

Odds bandfair odds (no market price)

RiskMedium

Stake1% flat, paper

LimitUp to 3 per round

How it picks

• Backs second-half Over 1.5 only when the second half is expected livelier than the first
• Backs Under when the model expects the game to shut down
• Fair odds, paper money: it measures calibration, not market edge

Where its edge would be

Fatigue, desperation and bench depth that a full-time price does not separate out.

Honest weakness

Best with live data, and without second-half prices it cannot claim an edge, only calibration.

How it improves

Hit rate against the percentage it promised.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The control group. The Coin Flip picks at random. It exists to keep the others honest: if an agent cannot beat a blindfolded coin flip, its style has no measurable effect. It is not an agent, it is the yardstick.

Its last calls

• HITPanathinaikos vs Kifisia - over_2.51.75+€7.50
• MISSEstrela vs Sp Braga - under_2.51.80−€10.00
• HITBrighton vs Leeds - draw3.50+€25.00
• HITBrentford vs Sunderland - under_2.52.00+€10.00
• HITBrentford vs Sunderland - draw3.90+€29.00
• MISSNewcastle vs Bournemouth - under_2.52.38−€10.00

Open right now

• OPENLiverpool vs Fulham - over_2.51.442026-09-12
• OPENCrystal Palace vs Ipswich - over_2.51.802026-09-12
• OPENChelsea vs Hull - over_2.51.402026-09-12
• OPENBournemouth vs Brentford - home2.502026-09-12
• OPENAston Villa vs Nottingham Forest - draw3.402026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysAny priced market, picked at random

Odds bandany

Riskcontrol

Stake1%

LimitOne per match

How it picks

• Picks a random priced market in each match
• Learns nothing: a baseline never learns

Where its edge would be

None whatsoever. That is the entire point.

Honest weakness

All of them.

How it improves

Nothing. The control is frozen so the comparison stays fair.

Replay bankrollvs €1,000 start

94.7u95%

This round, open paper pick

The scoreline specialist. Our score model produces a full scoreline distribution for every match. It backs the single most likely scoreline, and only when that scoreline carries enough probability. We hold no correct-score price grid, so it plays at fair odds: the honest question it answers is whether the model's top scoreline lands as often as the model says.

Every call it has made · 5

• MISSGetafe vs Celta - cs_0-06.87−€10.00
• MISSPorto vs Moreirense - cs_2-06.71−€10.00
• MISSVolos NFC vs Olympiakos - cs_0-16.53−€10.00
• MISSUdinese vs Lazio - cs_1-17.22−€10.00
• HITParma vs Monza - cs_1-17.10+€60.97

Open right now

• OPENCasa Pia vs Porto - cs_0-16.492026-09-12
• OPENComo vs Parma - cs_1-06.612026-09-14
• OPENGetafe vs La Coruna - cs_1-06.492026-09-13
• OPENSunderland vs Arsenal - cs_0-17.012026-09-12

Every pick this agent published here, settled against real results. Every agent stakes a flat €10 per call, so the board compares styles and not staking plans. Paper money: no account, no deposit, no payout, never betting advice.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysCorrect score (top scoreline only)

Odds bandfair odds, typically 6 to 12

RiskHigh

Stake0.2% flat, paper

LimitUp to 2 per round

How it picks

• Backs only the most likely scoreline per match
• Only when that scoreline clears its probability threshold, so it plays rarely
• Fair odds, paper money: it measures calibration, not market edge

Where its edge would be

A concentrated scoreline distribution in a match the market treats as open.

Honest weakness

High variance and rare hits by nature; it needs a sample before its record means anything.

How it improves

Hit rate against the promised percentage, once it has at least eight settled picks.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

The draw is one of the hardest outcomes to beat, so this agent stays quiet almost all the time. It only backs a draw when the model reads a genuinely low event, tightly matched game and the price still pays more draw than the model expects.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result: the draw

Odds band3.00 to 4.50

RiskMedium

Stake1% flat

LimitUp to 3 per round

How it picks

• Backs the draw only, never a side
• Model draw probability must clear its bar (about 30% and up)
• Draw odds between 3.00 and 4.50, with positive expected value
• Maximum 3 per round, otherwise no bet

Where its edge would be

A closed, cagey game where the market has priced the draw too cheaply.

Honest weakness

The draw is notoriously hard to beat. Expect many passes and high variance per hit.

How it improves

Realised return. Losing raises its value bar and makes it even more selective.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

No bet this round, and that is a decision.
Its reason: no draw with model probability of 30% or more and expected value of 5% or more in the 3.0 to 4.5 band, and sitting the draw out is the normal call.

The mathematician. The Closer ignores form and stories entirely. It does only the two things in betting with real mathematical substance: arbitrage (when the best prices across books guarantee a profit) and price execution (when a soft book closing price beats the sharpest reference). Its expectation is honest: thin, near break even, and only frozen holdout results count as proof.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysMatch result, shopped across many bookmakers

Odds band1.30 to 12.00

RiskLow to Medium

Stakequarter-Kelly, capped

LimitUp to 5 flags per round

How it picks

• Arbitrage: covers all outcomes when the best-book sum drops below 1
• Value: a soft closing price beats the de-vigged Pinnacle closing fair
• No sharp reference means no bet (the edge would be unprovable)
• Quarter-Kelly per flag, up to 5 flags per round

Where its edge would be

A guaranteed arbitrage, or a soft book lagging the market right at kickoff.

Honest weakness

A recorded price is not always executable (limits, cancellations). The edge is thin and may be zero: only frozen holdout results count.

How it improves

It never self-adjusts mid-season. Parameters change only through a guard-railed version step, which is anti-overfit by design.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

No bet this round, and that is a decision.
Its reason: no arbitrage (the best book prices still sum above 1) and no closing signal of 2% or more in the 1.5 to 3.5 band, which is what an efficient market looks like.

The spread specialist. The Asian handicap often carries a lower margin than 1X2. It trades REAL handicap prices (Pinnacle first) against the cover probability read straight off our score model, and only on half lines, so a bet can never push: it wins or it loses, and the record stays clean.

Historical replay · 2026-09

A backtest, not a published record: the same style replayed over past rounds with a fresh €1,000 on the 1st. It only moves when new history exists.

PlaysAsian handicap (half lines)

Odds band1.70 to 2.20

RiskMedium

Stake0.75% flat

LimitUp to 3 per round

How it picks

• Backs the handicap side when cover probability times price beats its edge threshold
• Half lines only: quarter and whole lines are skipped, never approximated
• Real bookmaker prices, so this is a genuine market test

Where its edge would be

Line placement the score model disagrees with, at a lower margin than the 1X2 market.

Honest weakness

Thin edges that demand discipline; a few basis points of margin decide everything.

How it improves

Realised return: losses raise the edge threshold, wins lower it slightly.

Replay bankrollvs €1,000 start

100.0u100%

This round, open paper pick

No bet this round, and that is a decision.
Its reason: ingen halv AH-linje i 1.7-2.2 med EV >= 3.0% (kvart-/hele linjer og upriste kampe springes over).

The scorer specialist. The market is slow on role changes: a striker on penalties, a new set-piece taker, an injury elsewhere. It will trade scorer props once we add scorer odds and confirmed lineups. Dormant until both arrive.

Coming soon. This expert goes live once we add the data it needs: scorer prop odds + confirmed lineups. Until then it sits out (no invented results).

PlaysAnytime / first goalscorer

Odds band1.80 to 9.00

RiskHigh

Stake0.25 to 0.4%

LimitUp to 3 per round

How it picks

• Targets anytime and first scorer after a role change
• Leans on penalties, set pieces and minutes
• Needs scorer prop odds and confirmed lineups

Where its edge would be

A shift in who takes the goals that the price has not caught up with.

Honest weakness

High variance and a fat bookmaker margin, plus prop data we do not yet have.

How it improves

Realised return once live.

This round, open paper pick

No open pick right now (between rounds).

The player volume specialist. Player shot props often blend volume and accuracy too crudely, and that gap is what it would trade. The per-player data is in place now: every shot in our five xG leagues carries the player who took it. What is missing is the prices. It could run as fair-odds paper first, the same way the Scoreline Sniper does.

Coming soon. This expert goes live once we add the data it needs: player prop odds. Until then it sits out (no invented results).

PlaysPlayer shots / shots on target

Odds band1.70 to 3.50

RiskMedium

Stake0.25 to 0.5%

LimitUp to 4 per round

How it picks

• Targets player shots and shots on target lines
• Separates shot volume from on-target accuracy
• Per-player shot data is in place; it is waiting on prop prices

Where its edge would be

Volume and accuracy mixed too coarsely in the player price.

Honest weakness

Without prop prices it cannot claim an edge, only calibration.

How it improves

Realised return once live.

This round, open paper pick

No open pick right now (between rounds).

The portfolio trader. It re-rates futures as the season unfolds via simulation, buying and selling only on big gaps. It needs futures odds and a season simulation layer before it can hold positions.

Coming soon. This expert goes live once we add the data it needs: futures odds + season simulation. Until then it sits out (no invented results).

PlaysLeague winner, promotion, relegation, outrights

Odds band2.00 to 50.00

RiskMedium to High

Stake0.15 to 0.35% position

LimitPosition trades

How it picks

• Re-rates outrights continuously via season simulation
• Trades only on large gaps to the market
• Needs futures odds and a season simulator

Where its edge would be

The market adjusts slowly to strength shifts and path dependency.

Honest weakness

Capital is tied up for a long time, and it needs futures pricing we do not yet hold.

How it improves

Position level return once live.

This round, open paper pick

No open pick right now (between rounds).

The live trader. When the model reads more goal threat than the live draw price implies, it lays the draw and trades out on a goal or a time stop. It needs a live exchange feed with in-play prices and depth, which we do not yet have.

Coming soon. This expert goes live once we add the data it needs: live betting-exchange feed (in-play prices). Until then it sits out (no invented results).

PlaysIn-play draw (lay)

Odds bandexchange in-play

RiskMedium to High

Stake0.25 to 0.5% liability

LimitSession capped

How it picks

• Lays the draw when goal threat beats the live price
• Trades out on a goal or a time stop
• Needs a live betting-exchange feed

Where its edge would be

The live draw stays sticky when a game opens up.

Honest weakness

Execution and slippage decide everything, and it needs a live exchange feed.

How it improves

Realised return on closed trades once live.

This round, open paper pick

No open pick right now (between rounds).