Introducing AI Deep Dive: Horse Racing Podcasts!

[AJ the Hobbyist]

▸ SALISBURY SATURDAY 23 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Audit-based structure across the Salisbury card, built from AU figs, Smart Stats and tactical forecast layers.
No tipping language, no outcome claims, just the live V15 Early Doors framework.

• AU integrity checked race-by-race against uploaded points totals
• Tactical forecasts built winner-first from the strongest AU-supported anchor
• Smart Stats folded in through hot/cold jockey and trainer handling
• Market compression used only as support, never as an AU override
• Caution markers retained for BF LTO, class drops, headgear and stable switches
• Forecast and TOTE structures bound to the same Win Pick anchor
• Overlay alignment reviewed across AU figs, racecard evidence, Smart Stats and market layers

Read the full card... Salisbury Saturday 23 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: CALL ME TOMORROW
• Race 2: CRISTO
• Race 3: GOLDEN BROWN
• Race 4: PRINCE OF INDIA
• Race 5: ARCHER'S GRACE
• Race 6: MAYAADA
• Race 7: SHAYHANA

Forecast Combos
• Race 1: CALL ME TOMORROW → SHIMMERING SUN / GOLDTRADER
• Race 2: CRISTO → GREAT DREAM / DOVEY MOON
• Race 3: GOLDEN BROWN → CHARLIE BOYO / WHITESNAKE
• Race 4: PRINCE OF INDIA → HOLLYWOOD TREASURE / ZOUM ZOUM
• Race 5: ARCHER'S GRACE → LOVERS LEAP / MOJITO
• Race 6: MAYAADA → TOO HOT TO TANGO / FAHIDI
• Race 7: SHAYHANA → SIOUXFONIC / MASTERINTHEWOODS

EW/Combo Value Inclusions
• GOLDTRADER
• DOVEY MOON
• WHITESNAKE
• HOLLYWOOD TREASURE
• ZOUM ZOUM
• MOJITO
• FAHIDI
• MASTERINTHEWOODS

TOTE Combos Recap
• Race 1: CALL ME TOMORROW + SHIMMERING SUN / GOLDTRADER
• Race 2: CRISTO + GREAT DREAM / DOVEY MOON
• Race 3: GOLDEN BROWN + CHARLIE BOYO / WHITESNAKE
• Race 4: PRINCE OF INDIA + HOLLYWOOD TREASURE / ZOUM ZOUM
• Race 5: ARCHER'S GRACE + LOVERS LEAP / MOJITO
• Race 6: MAYAADA + TOO HOT TO TANGO / FAHIDI
• Race 7: SHAYHANA + SIOUXFONIC / MASTERINTHEWOODS

Caution Marker List
• CALL ME TOMORROW – beaten favourite LTO + class-drop volatility
• CHAMPAGNE POWDER – first-time tongue strap + cold trainer
• WHITESNAKE – market weakness versus AU
• PRINCE OF INDIA – beaten favourite LTO
• WHITE LADDER – stable switch
• FAHIDI – first-time cheekpiece
• CINAMMON COCO – beaten favourite LTO + stable switch

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

AU integrity

• Race 1: AU integrity evidenced — CALL ME TOMORROW led uploaded points totals with 15pts.
• Race 2: AU integrity evidenced — CRISTO led uploaded points totals with 8pts.
• Race 3: AU integrity evidenced — GOLDEN BROWN led uploaded points totals with 17pts.
• Race 4: AU integrity evidenced — PRINCE OF INDIA led uploaded points totals with 14pts.
• Race 5: AU integrity evidenced — ARCHER'S GRACE led uploaded points totals with 10pts.
• Race 6: AU integrity evidenced — MAYAADA led uploaded points totals with 14pts.
• Race 7: AU integrity evidenced — SHAYHANA led uploaded points totals with 9pts.

SEE THE BLOG!

▸ SALISBURY SATURDAY 23 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Post-race critique now added to the Salisbury V15 Early Doors card.
Audit-based, structure-first review only — no tipping language, no outcome claims.

• AU figs checked against the original race-by-race Win Pick structure
• Smart Stats handling reviewed through hot/cold jockey and trainer layers
• Forecast zones assessed against the declared V15 anchor-and-partner framework
• TOTE structure kept bound to the original Win Pick anchor rules
• Caution markers retained for BF LTO, class drops, headgear and stable switches
• Chaos control applied where partner strength appeared without Win Pick conversion
• Model integrity separated from betting outcome under the locked critique rules

Read the full card and post-race critique: Salisbury Saturday 23 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

Cumulative Outcome Analysis

Win Picks:
• Race 1: Call Me Tomorrow — 2nd
• Race 2: Cristo — 2nd
• Race 3: Golden Brown — 1st
• Race 4: Prince Of India — 4th
• Race 5: Archer's Grace — 4th
• Race 6: Mayaada — 1st
• Race 7: Shayhana — unplaced

Win Pick outcome:
• 2 wins from 7 races.

Exacta outcome:
• Race 1: FAILED
• Race 2: FAILED
• Race 3: FAILED
• Race 4: FAILED
• Race 5: FAILED
• Race 6: FAILED
• Race 7: FAILED

Boxed Trifecta outcome:
• Race 1: FAILED
• Race 2: FAILED
• Race 3: FAILED
• Race 4: FAILED
• Race 5: FAILED
• Race 6: FAILED
• Race 7: FAILED

Structured doubles:
• Mayaada / Shayhana — LOST
• Prince Of India / Mojito — LOST

Overall:
• Win-pick layer landed twice.
• Exacta layer did not land.
• Boxed trifecta layer did not land.
• Structured doubles did not land.
• No TOTE payout or P/L bracket printed because no qualifying V15 Exacta or Boxed Trifecta landed under the locked rules.

Model Reflection & Refinement Notes

The strongest part of the build was the winner-first discipline on Golden Brown and Mayaada.

The main structural weakness was forecast-partner accuracy. Race 3 and Race 6 both found the winner, but neither converted into the anchored Exacta because the 2nd horse sat outside the forecast pair.

Race 1 and Race 2 showed ordering exposure. The structure contained the winner and Win Pick in the first two places, but the Win Pick did not win, so the Exacta correctly failed under the enforced rule.

Race 7 showed partner strength but anchor failure. Masterinthewoods and Siouxfonic filled the first two positions, but Shayhana failed as the Win Pick, so the model structure was exposed at the anchor level.

Race 4 and Race 5 were direct anchor failures. Prince Of India and Archer's Grace did not convert AU leadership into the win slot.

Carry-forward refinement:
• Keep winner-first discipline.
• Do not over-credit partner success when the Win Pick fails.
• Treat partner-first outcomes as structural evidence, not betting success.
• Tighten forecast partner selection where the Win Pick is strong but the second-place risk is broader than the selected pair.
• Keep TOTE payout discipline strict: no landed structure, no payout, no P/L bracket.

 

[AJ the Hobbyist]

▸ CURRAGH 24 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Curragh handled through the V15 Early Doors audit frame.
Structure first, with AU alignment, Smart Stats, market shape, and caution control kept separate.

• Eight-race Curragh card built from uploaded racecard, Smart Stats, AU-style points, and market layers
• AU integrity retained race by race, with Win Picks tied back to uploaded points totals
• Market prices used only as compression evidence, not as an override of AU alignment
• Tactical forecast structure built winner-first, then outward into Partner A and Partner B
• Caution markers isolated where market weakness versus AU or weaker AU support was evidenced
• TOTE Exacta and Trifecta structures anchored to the same Win Pick as the main V15 selection
• No result logic, simulation, or unsupported upgrade used in the live structure
• Charter discipline maintained: tactical overlay system, not a tipping service

Read the full card... Curragh 24 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: SUN GODDESS
• Race 2: GREAT BARRIER REEF
• Race 3: PREVALENCE
• Race 4: JANCIS
• Race 5: MINNIE HAUK
• Race 6: TRUE LOVE
• Race 7: SPANISH TEMPTRESS
• Race 8: STORM HEART

Forecast Combos
• Race 1: SUN GODDESS → LAN DE GHAIRE / LADY PATRONA
• Race 2: GREAT BARRIER REEF → CARRY THE FLAG / IMMORTAL GUARD
• Race 3: PREVALENCE → GREEN CARRERA / ANUSHKA
• Race 4: JANCIS → CITY OF MEMPHIS / ONE LOOK
• Race 5: MINNIE HAUK → BAY CITY ROLLER / ALMAQAM
• Race 6: TRUE LOVE → PRECISE / ABASHIRI
• Race 7: SPANISH TEMPTRESS → HIGHBURY SEE SEE / SUMMER ISLAND
• Race 8: STORM HEART → KINGSTONIAN / RAGMANS CORNER

EW/Combo Value Inclusions
• LAN DE GHAIRE
• LADY PATRONA
• CARRY THE FLAG
• IMMORTAL GUARD
• GREEN CARRERA
• ANUSHKA
• CITY OF MEMPHIS
• ONE LOOK
• BAY CITY ROLLER
• ALMAQAM
• PRECISE
• ABASHIRI
• HIGHBURY SEE SEE
• SUMMER ISLAND
• KINGSTONIAN
• RAGMANS CORNER

TOTE Combos Recap
• Race 1: SUN GODDESS + LAN DE GHAIRE / LADY PATRONA
• Race 2: GREAT BARRIER REEF + CARRY THE FLAG / IMMORTAL GUARD
• Race 3: PREVALENCE + GREEN CARRERA / ANUSHKA
• Race 4: JANCIS + CITY OF MEMPHIS / ONE LOOK
• Race 5: MINNIE HAUK + BAY CITY ROLLER / ALMAQAM
• Race 6: TRUE LOVE + PRECISE / ABASHIRI
• Race 7: SPANISH TEMPTRESS + HIGHBURY SEE SEE / SUMMER ISLAND
• Race 8: STORM HEART + KINGSTONIAN / RAGMANS CORNER

Caution Marker List
• LAN DE GHAIRE – Market weakness versus AU evidenced from uploaded market layer
• IMMORTAL GUARD – Market weakness versus AU evidenced from uploaded market layer
• ANUSHKA – Market weakness versus AU evidenced from uploaded market layer
• BAY CITY ROLLER – Market weakness versus AU evidenced from uploaded market layer
• ABASHIRI – Weaker AU points support versus market position evidenced from uploaded layers
• RAGMANS CORNER – Market weakness versus AU evidenced from uploaded market layer

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

AU integrity

• Race 1: AU integrity evidenced — SUN GODDESS led uploaded points totals with 13pts.
• Race 2: AU integrity evidenced — GREAT BARRIER REEF led uploaded points totals with 14pts.
• Race 3: AU integrity evidenced — PREVALENCE led uploaded points totals with 10pts.
• Race 4: AU integrity evidenced — JANCIS led uploaded points totals with 11pts.
• Race 5: AU integrity evidenced — MINNIE HAUK led uploaded points totals with 18pts.
• Race 6: AU integrity evidenced — TRUE LOVE led uploaded points totals with 16pts.
• Race 7: AU integrity evidenced — SPANISH TEMPTRESS led uploaded points totals with 13pts.
• Race 8: AU integrity evidenced — STORM HEART led uploaded points totals with 15pts. SEE THE BLOG!

 

[AJ the Hobbyist]

▸ CURRAGH 24 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Curragh reviewed through the V15 Early Doors audit frame.
Structure first, with AU evidence, Smart Stats, market shape, and caution handling kept separate.

• AU figs remained the primary driver across the race-by-race review
• Smart Stats support was checked against the original forecast zones
• Forecast structures were reviewed through Win Pick, Partner A, and Partner B only
• TOTE Exacta and Trifecta logic was kept tied to the declared pre-race structure
• Caution markers were retained where AU strength and market position needed separation
• Chaos control stayed focused on evidence discipline, not hindsight reshaping
• Betting outcome and model integrity were treated as separate audit layers
• Charter discipline held: tactical overlay system, not a tipping service

Read the full card and post-race critique: Curragh 24 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

Cumulative Outcome Analysis

Win Pick performance:
• Race 1: SUN GODDESS won.
• Race 2: GREAT BARRIER REEF won.
• Race 3: PREVALENCE unplaced.
• Race 4: JANCIS unplaced.
• Race 5: MINNIE HAUK unplaced.
• Race 6: TRUE LOVE finished 2nd.
• Race 7: SPANISH TEMPTRESS unplaced.
• Race 8: STORM HEART unplaced.

Win Pick strike:
• 2 wins from 8 races.

Exacta performance:
• Race 2 exacta landed.
• All other exactas failed.

Trifecta performance:
• Race 2 boxed trifecta landed.
• Race 6 boxed trifecta landed.
• All other boxed trifectas failed.

Structured bet outcome:
• Yankee lost.
• Treble lost.
• Total uploaded settled bet return: £0.00.

Model integrity:
• Race 1 held at Win Pick level only.
• Race 2 was a clean structural hit.
• Race 3 exposed anchor failure despite two forecast horses placing in the top three.
• Race 4 exposed anchor failure and missed outsider disruption.
• Race 5 exposed anchor failure but retained meaningful partner relevance.
• Race 6 delivered full boxed trifecta structure but failed winner-first ordering.
• Race 7 failed at anchor and forecast level except for Summer Island placing.
• Race 8 failed completely.

Model Reflection & Refinement Notes

What held structurally:
• AU-led short-price anchors held in Race 1 and Race 2.
• Race 2 produced the cleanest structure, with Win Pick, exacta, and boxed trifecta all landing.
• Race 6 showed strong three-runner structural containment, with all forecast horses filling the first three positions.
• Green Carrera was correctly retained inside the Race 3 structure and also won as a structured bet-slip leg.
• Almaqam was correctly retained inside the Race 5 forecast structure and won.

What failed structurally:
• Winner-first ordering failed in Races 3, 4, 5, 6, 7, and 8.
• Race 3 selected PREVALENCE over GREEN CARRERA, exposing anchor discipline.
• Race 4 selected JANCIS over CITY OF MEMPHIS, exposing AU-versus-market/fitness ordering.
• Race 5 selected MINNIE HAUK over ALMAQAM, exposing anchor failure despite partner relevance.
• Race 6 selected TRUE LOVE over PRECISE, producing a boxed trifecta return but failing the win-pick-anchored exacta rule.
• Race 8 produced no structural containment.

Betting outcome versus model integrity:
• The settled bet slip failed.
• The model did not fail uniformly.
• Race 2 was clean.
• Race 6 was structurally strong for boxed trifecta logic but failed winner-first discipline.
• Race 3 and Race 5 showed meaningful shortlist quality but incorrect anchor ordering.
• Race 8 was the clearest full structural miss.

Refinement notes:
• Market weakness versus AU must remain a harder caution in competitive handicaps.
• AU points leadership should not automatically dominate where market compression and partner evidence identify a stronger live win candidate.
• In big-field handicaps, a low-AU but market-compressed runner should not be forced above a stronger AU runner without direct support, but AU-led anchors with market drift require stricter caution.
• The system should preserve Race 2 structure as the clean model standard.
• Race 6 confirms that boxed trifecta value can hold even where winner-first selection fails.
• V15 should continue to separate Win Pick discipline from forecast/TOTE containment.

 

[AJ the Hobbyist]

▸ REDCAR MONDAY 25 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

A clean V15 Early Doors structure for Redcar, built from uploaded AU layers, Smart Stats, market shape and caution evidence.
Audit-based, structure-first, and not a tipping post.

• AU integrity checked across all seven races, with each Win Pick tied to uploaded points evidence
• Tactical forecast structure built winner-first, then outward through Partner A and Partner B
• Smart Stats integrated for hot/cold jockeys, hot/cold trainers, headgear, class droppers and weighted-to-win runners
• Market compression used as support only, not as an override to AU alignment
• Caution markers retained where evidenced, including beaten favourites, headgear, class-drop volatility and market weakness versus AU
• TOTE Exacta and Trifecta structures anchored to the same V15 Win Pick in every race
• Charter discipline maintained throughout: no simulation, no hindsight, no unsupported upgrades

Read the full card... Redcar Monday 25 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: MARDY BUM
• Race 2: ROMANTICIZING
• Race 3: PERFIDIA
• Race 4: RIFFA SPIRIT
• Race 5: RAINBOW NEBULA
• Race 6: WINGED MESSENGER
• Race 7: PRINCE HECTOR

Forecast Combos
• Race 1: MARDY BUM → EL FLORIDITA / CELESTIAL FIRE
• Race 2: ROMANTICIZING → GLEN NEVIS / NORMAN INVASION
• Race 3: PERFIDIA → EY UP HE'S A STAR / MAYO COUNTY
• Race 4: RIFFA SPIRIT → FILLE UNIQUE / SOCA STAR
• Race 5: RAINBOW NEBULA → FIERCE FORTITUDE / DANGER BAY
• Race 6: WINGED MESSENGER → ARTAGNAN / BADRI
• Race 7: PRINCE HECTOR → HEARTENED / WOODLEIGH

EW/Combo Value Inclusions
• EL FLORIDITA
• CELESTIAL FIRE
• GLEN NEVIS
• NORMAN INVASION
• EY UP HE'S A STAR
• MAYO COUNTY
• FILLE UNIQUE
• SOCA STAR
• FIERCE FORTITUDE
• DANGER BAY
• ARTAGNAN
• BADRI
• HEARTENED
• WOODLEIGH

TOTE Combos Recap
• Race 1: MARDY BUM + EL FLORIDITA / CELESTIAL FIRE
• Race 2: ROMANTICIZING + GLEN NEVIS / NORMAN INVASION
• Race 3: PERFIDIA + EY UP HE'S A STAR / MAYO COUNTY
• Race 4: RIFFA SPIRIT + FILLE UNIQUE / SOCA STAR
• Race 5: RAINBOW NEBULA + FIERCE FORTITUDE / DANGER BAY
• Race 6: WINGED MESSENGER + ARTAGNAN / BADRI
• Race 7: PRINCE HECTOR + HEARTENED / WOODLEIGH

Caution Marker List
• MARDY BUM – Market weakness versus AU
• EY UP HE'S A STAR – Beaten favourite LTO
• RIFFA SPIRIT – Beaten favourite LTO
• RAINBOW NEBULA – Market weakness versus AU; headgear
• WINGED MESSENGER – Class-drop volatility; headgear
• PRINCE HECTOR – Headgear; cold jockey; cold trainer; market weakness versus AU

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

AU integrity

• Race 1: AU integrity evidenced — MARDY BUM led uploaded points totals with 8pts.
• Race 2: AU integrity evidenced — ROMANTICIZING led uploaded points totals with 16pts.
• Race 3: AU integrity evidenced — PERFIDIA led uploaded points totals with 12pts.
• Race 4: AU integrity evidenced — RIFFA SPIRIT led uploaded points totals with 11pts.
• Race 5: AU integrity evidenced — RAINBOW NEBULA led uploaded points totals with 9pts.
• Race 6: AU integrity evidenced — WINGED MESSENGER led uploaded points totals with 9pts.
• Race 7: AU integrity evidenced — PRINCE HECTOR led uploaded points totals with 12pts. SEE THE BLOG!

 

[AJ the Hobbyist]

I'm changing MY BETS from the usual Yankee to 3 x combo Tricast. These are not tips or advice! Bank Holiday madness.

 

[AJ the Hobbyist]

AJ - Do you think it is time to make a new V15 project thread and update my build prompts? I'm reluctant to interrupt your progress with the V15 ED blog.

GPT_HRE
Yes — I think it is time, but not in a way that interrupts the current V15 ED blog.
The Redcar build has exposed enough recurring prompt-drift points to justify a fresh V15 project thread and a tightened master prompt. The main reasons are:
VALIDATION & TRUST LAYER should now be hard-locked as a title before AU integrity.
The AU integrity wording is now mature and should be permanently embedded, especially the race-by-race points-total format.
The Step 04 blog build is working well structurally, but the prompt has become very long and heavy. A cleaner thread with a condensed, locked version would reduce drift.
The e/w lesson is now settled: do not turn V15 into an each-way service; use place-shape evidence for forecast, TOTE and combo structure only.
The winner-first override is now central and should sit nearer the top of the master prompt rather than being buried inside later build logic.
I would not abandon this thread mid-card. Best route:
Finish today’s Redcar workflow here.
Then start a new V15 ED project thread with an updated “MASTER DAILY BUILD PROMPT — VERSION 4” that carries forward only the hardened rules, not all the accumulated working clutter.
The new thread should probably include these permanent locks:
• ED = Human & Computer working together
• winner-first AU hierarchy
• uploaded-layer-only source discipline
• Step 04 modular output lock
• VALIDATION & TRUST LAYER title lock
• race-by-race AU integrity points format
• Smart Stats validation structure
• market weakness versus AU caution handling
• e/w evidence used only for forecast/TOTE/place-structure
• no tipping language / no simulation / no hindsight
So my judgement: yes, create the new thread after today’s Redcar card is fully closed, not before. That preserves continuity and gives us a cleaner base for tomorrow.

 

[AJ the Hobbyist]

▸ REDCAR MONDAY 25 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

A clean V15 Early Doors structure for Redcar, built from uploaded AU layers, Smart Stats, market shape and caution evidence.
Audit-based, structure-first, and not a tipping post.

• AU integrity checked across all seven races, with each Win Pick tied to uploaded points evidence
• Tactical forecast structure built winner-first, then outward through Partner A and Partner B
• Smart Stats integrated for hot/cold jockeys, hot/cold trainers, headgear, class droppers and weighted-to-win runners
• Market compression used as support only, not as an override to AU alignment
• Caution markers retained where evidenced, including beaten favourites, headgear, class-drop volatility and market weakness versus AU
• TOTE Exacta and Trifecta structures anchored to the same V15 Win Pick in every race
• Charter discipline maintained throughout: no simulation, no hindsight, no unsupported upgrades

Read the full card... Redcar Monday 25 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: MARDY BUM
• Race 2: ROMANTICIZING
• Race 3: PERFIDIA
• Race 4: RIFFA SPIRIT
• Race 5: RAINBOW NEBULA
• Race 6: WINGED MESSENGER
• Race 7: PRINCE HECTOR

Forecast Combos
• Race 1: MARDY BUM → EL FLORIDITA / CELESTIAL FIRE
• Race 2: ROMANTICIZING → GLEN NEVIS / NORMAN INVASION
• Race 3: PERFIDIA → EY UP HE'S A STAR / MAYO COUNTY
• Race 4: RIFFA SPIRIT → FILLE UNIQUE / SOCA STAR
• Race 5: RAINBOW NEBULA → FIERCE FORTITUDE / DANGER BAY
• Race 6: WINGED MESSENGER → ARTAGNAN / BADRI
• Race 7: PRINCE HECTOR → HEARTENED / WOODLEIGH

EW/Combo Value Inclusions
• EL FLORIDITA
• CELESTIAL FIRE
• GLEN NEVIS
• NORMAN INVASION
• EY UP HE'S A STAR
• MAYO COUNTY
• FILLE UNIQUE
• SOCA STAR
• FIERCE FORTITUDE
• DANGER BAY
• ARTAGNAN
• BADRI
• HEARTENED
• WOODLEIGH

TOTE Combos Recap
• Race 1: MARDY BUM + EL FLORIDITA / CELESTIAL FIRE
• Race 2: ROMANTICIZING + GLEN NEVIS / NORMAN INVASION
• Race 3: PERFIDIA + EY UP HE'S A STAR / MAYO COUNTY
• Race 4: RIFFA SPIRIT + FILLE UNIQUE / SOCA STAR
• Race 5: RAINBOW NEBULA + FIERCE FORTITUDE / DANGER BAY
• Race 6: WINGED MESSENGER + ARTAGNAN / BADRI
• Race 7: PRINCE HECTOR + HEARTENED / WOODLEIGH

Caution Marker List
• MARDY BUM – Market weakness versus AU
• EY UP HE'S A STAR – Beaten favourite LTO
• RIFFA SPIRIT – Beaten favourite LTO
• RAINBOW NEBULA – Market weakness versus AU; headgear
• WINGED MESSENGER – Class-drop volatility; headgear
• PRINCE HECTOR – Headgear; cold jockey; cold trainer; market weakness versus AU

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

AU integrity

• Race 1: AU integrity evidenced — MARDY BUM led uploaded points totals with 8pts.
• Race 2: AU integrity evidenced — ROMANTICIZING led uploaded points totals with 16pts.
• Race 3: AU integrity evidenced — PERFIDIA led uploaded points totals with 12pts.
• Race 4: AU integrity evidenced — RIFFA SPIRIT led uploaded points totals with 11pts.
• Race 5: AU integrity evidenced — RAINBOW NEBULA led uploaded points totals with 9pts.
• Race 6: AU integrity evidenced — WINGED MESSENGER led uploaded points totals with 9pts.
• Race 7: AU integrity evidenced — PRINCE HECTOR led uploaded points totals with 12pts. SEE THE BLOG!

▸ REDCAR MONDAY 25 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Redcar reviewed through the V15 Early Doors audit lens, with structure separated from outcome.
AU figs, Smart Stats and caution markers kept in their lanes throughout.

• AU points shaped the winner-first structure across the seven-race card
• Smart Stats supported the audit through hot/cold jockeys, trainer notes, headgear and class-drop flags
• Forecast zones were built from the Win Pick outward into Partner A and Partner B
• Market compression was used as structural evidence only, not as an override
• Caution markers were retained where evidenced, including beaten favourites, headgear and market weakness versus AU
• TOTE Exacta and Trifecta logic remained bound to the declared V15 anchor structure
• Post-race critique separates betting outcome from model integrity and structural exposure

Read the full card and post-race critique: Redcar Monday 25 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

Cumulative Outcome Analysis

Across the seven-race card:

• V15 Win Picks produced one winner: Romanticizing.
• Forecast partners produced multiple result-frame or winning placements, including El Floridita, Celestial Fire, Mayo County, Fille Unique, Danger Bay, Fierce Fortitude, Badri, Woodleigh and Heartened.
• Exacta outcomes failed across the card under the enforced Win-Pick-anchored rule.
• Boxed Trifecta outcomes failed across the card because no race had all three forecast horses finishing in the top three.
• The structured bet slip lost across all three uploaded Combination Tricast positions.

No TOTE Exacta P/L bracket printed because no Exacta was declared landed.

No TOTE Trifecta P/L bracket printed because no boxed Trifecta was declared landed.

Model Reflection & Refinement Notes

The strongest issue was not total structural blindness. The recurring issue was anchor failure versus partner strength.

Partner evidence repeatedly found live horses, but the AU-led Win Pick did not convert often enough outside Romanticizing.

Race 2 held cleanly at Win Pick level, but failed as a forecast structure because Bee Farmer was not contained.

Race 3, Race 5, Race 6 and Race 7 all show the same refinement point: when the eventual winner is already inside the partner structure, the model should not over-protect the AU points leader if that leader carries unresolved market weakness or caution pressure.

Market weakness versus AU remains a live caution, especially where the market is clearly stronger around a forecast partner.

For future builds, retain winner-first discipline, but tighten treatment of AU leaders carrying caution stacks where a partner has both market compression and clean supporting evidence.

The model was not clean enough on Win Pick strike rate. Forecast and TOTE structure had live fragments, but fragment strength does not repair failed anchors.

 

[AJ the Hobbyist]

AJ - Do you think it is time to make a new V15 project thread and update my build prompts? I'm reluctant to interrupt your progress with the V15 ED blog.

GPT_HRE
Yes — I think it is time, but not in a way that interrupts the current V15 ED blog.
The Redcar build has exposed enough recurring prompt-drift points to justify a fresh V15 project thread and a tightened master prompt. The main reasons are:
VALIDATION & TRUST LAYER should now be hard-locked as a title before AU integrity.
The AU integrity wording is now mature and should be permanently embedded, especially the race-by-race points-total format.
The Step 04 blog build is working well structurally, but the prompt has become very long and heavy. A cleaner thread with a condensed, locked version would reduce drift.
The e/w lesson is now settled: do not turn V15 into an each-way service; use place-shape evidence for forecast, TOTE and combo structure only.
The winner-first override is now central and should sit nearer the top of the master prompt rather than being buried inside later build logic.
I would not abandon this thread mid-card. Best route:
Finish today’s Redcar workflow here.
Then start a new V15 ED project thread with an updated “MASTER DAILY BUILD PROMPT — VERSION 4” that carries forward only the hardened rules, not all the accumulated working clutter.
The new thread should probably include these permanent locks:
• ED = Human & Computer working together
• winner-first AU hierarchy
• uploaded-layer-only source discipline
• Step 04 modular output lock
• VALIDATION & TRUST LAYER title lock
• race-by-race AU integrity points format
• Smart Stats validation structure
• market weakness versus AU caution handling
• e/w evidence used only for forecast/TOTE/place-structure
• no tipping language / no simulation / no hindsight
So my judgement: yes, create the new thread after today’s Redcar card is fully closed, not before. That preserves continuity and gives us a cleaner base for tomorrow.

These are the initial starter prompts for V15. I hope that someone finds aspects of the starter prompt useful for their own AI-driven projects. The remaining build steps remain confidential, and copyright has been confirmed. It has taken 16 months of posting daily meeting racecards to get this far. We go again tomorrow.

Daily V15 EARLY DOORS — SUPER-LEAN OPERATOR VERSION 4

ED is Human & Computer working together. The human retains final judgement, betting authority, publication authority, and all real-world action authority.

MODEL HANDLING CLAUSE
Use the strongest available reasoning model.
This prompt is model-agnostic.
Newer models do not have permission to expand, soften, reinterpret, restructure, or embellish the V15 method.
Better model capability must only improve compliance, consistency, contradiction detection, and clean execution.
Do not alter the evidence standard, staking discipline, output order, race structure, tone, pass logic, or source lock based on model availability.
If model behaviour becomes expansive, compress back to the locked format.
If model behaviour becomes conversational, remove warmth from racing analysis sections.
If model behaviour introduces unsupported interpretation, reject it unless supported by the declared evidence set.

ED CORE LOCK
• ED means Human & Computer working together
• The model supports structure, audit, contradiction detection and clean execution
• The human retains final judgement, betting authority, publication authority and all real-world action authority
• Model ≠ Result
• Structure ≠ Outcome
• No tipping language
• No simulation
• No hindsight
• No invented data

MODEL COMPLIANCE TEST
Follow the instructions EXACTLY.

RULES:
• Do not add commentary
• Do not explain anything
• Do not reformat anything
• Print exactly what is requested
• If any character, spacing, line break, bullet, or symbol differs → FAIL TEST

Print EXACTLY:

ED V15 DAILY BUILD — TEST COURSE — TEST DATE

V15 EARLY DOORS | TACTICAL FORECASTS + SMART STATS INTEGRATION

════════════════════════════════
TEST RACE — STRUCTURE CHECK
════════════════════════════════

Win Pick — TEST HORSE A
AU Alignment: Strong
AU Source: AU proxy: panel + form + pace

Partner A — TEST HORSE B
AU Alignment: Positive
AU Source: AU proxy: panel + suitability

Partner B — TEST HORSE C
AU Alignment: Positive
AU Source: AU proxy: form + market compression

Why this works
• Bullet one
• Bullet two
• Bullet three

PRINT THE SECOND 3 RACES

END TEST

///////////////////////////////////////////

════════════════════════════════
V15 EARLY DOORS — SUPER-LEAN BUILD LOCK
════════════════════════════════

EXECUTION ORDER LOCK
• Execute steps sequentially only
• Do not skip, merge, or reorder steps
• If sequencing is broken → RESTART FROM STEP 00

SOURCE LOCK
Use only uploaded layers in this thread:
• Smart Stats
• Racecards / tactical form layers
• Market data layers
• AU / computer tips / panel layers
• Later validation / marketing prompts as supplied

If any field, runner, stat, marker, or metadata is missing or unverifiable from uploaded layers:
• do not invent it
• do not assume it
• do not simulate it
• print the approved fallback wording where supplied
• STOP OUTPUT only if a required field remains missing after re-checking uploaded layers

WINNER-FIRST OVERRIDE
• Primary objective: identify the most likely WINNER using AU hierarchy
• Prioritise decisive selection over structural coverage
• Do not dilute the Win Pick to protect forecast balance
• If AU evidence is uneven → still commit to the strongest evidenced runner
• Model success is measured primarily by Win Pick strike rate
• Forecast, TOTE and place-structure are built outward from the Win Pick

MARKET DISCIPLINE LOCK
• Market prices must not override AU alignment
• Market compression may support, confirm or caution structure
• Market weakness versus AU is a caution only when directly evidenced
• Market weakness versus AU must be treated as harder caution in big-field handicaps
• Market-weak AU anchors are not clean anchors in big-field handicaps unless the supporting layers clearly justify retention

EW / PLACE-SHAPE LOCK
• V15 is not an each-way service
• Do not force place-shape evidence into the Win Pick slot
• Use e/w-style evidence only to strengthen forecast, TOTE, combo and danger handling
• If a runner is not the cleanest Win Pick but has strong place-shape evidence, mark it as a stronger forecast / TOTE / place-structure candidate

STEP 00 — ACTIVATE CHARTER

Activate V15 LEAN Charter.

Rules:
• Structural language only
• No tipping language
• No hindsight commentary
• No simulation
• Model ≠ Result
• Human retains final authority
• Uploaded evidence only

Role:
You are a UK horse racing analyst using only uploaded racecards, Smart Stats, AU layers and market layers. No invented data.

 

[AJ the Hobbyist]

▸ LEICESTER TUESDAY 26 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE
Audit-based V15 structure built from uploaded AU figs, Smart Stats, racecard layers and market data.
No tipping language — structure, caution control and evidence discipline only.
• AU hierarchy retained as the primary driver across the card
• Win Pick binding carried through Forecast Combo and TOTE Anchor
• Market compression used only as support, not override
• Smart Stats flags applied only where evidenced
• Caution markers retained for BF LTO, class drop, headgear and market weakness
• Race-by-race AU integrity checked against uploaded points totals
• Forecast and TOTE structures built outward from the Win Pick
• Charter discipline locked: no simulation, no hindsight, no invented data
Read the full card... Leicester Tuesday 26 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: VICTORY GOLD
• Race 2: ABRAHAMSEN
• Race 3: CROWNRIGHT
• Race 4: HATAMOTO
• Race 5: MYNA
• Race 6: FRANTIC
• Race 7: ALICE DE CLARE

Forecast Combos
• Race 1: VICTORY GOLD → LIBERTANGO / MIAMI TO IBIZA
• Race 2: ABRAHAMSEN → PARADING / BLUESTONE LADY
• Race 3: CROWNRIGHT → CAIM / HARLEQUIN BREEZE
• Race 4: HATAMOTO → SHE'S CRAFTY / WOODY Y FERNANDEZ
• Race 5: MYNA → HAPPY CHANDLER / BEAUTY GENERATION
• Race 6: FRANTIC → ALWAYS PERFECT / FILLY FODEN
• Race 7: ALICE DE CLARE → PENTONVILLE / ZIPWIRE

EW/Combo Value Inclusions
• LIBERTANGO
• MIAMI TO IBIZA
• PARADING
• BLUESTONE LADY
• CAIM
• HARLEQUIN BREEZE
• SHE'S CRAFTY
• WOODY Y FERNANDEZ
• HAPPY CHANDLER
• BEAUTY GENERATION
• ALWAYS PERFECT
• FILLY FODEN
• PENTONVILLE
• ZIPWIRE

TOTE Combos Recap
• Race 1: VICTORY GOLD + LIBERTANGO / MIAMI TO IBIZA
• Race 2: ABRAHAMSEN + PARADING / BLUESTONE LADY
• Race 3: CROWNRIGHT + CAIM / HARLEQUIN BREEZE
• Race 4: HATAMOTO + SHE'S CRAFTY / WOODY Y FERNANDEZ
• Race 5: MYNA + HAPPY CHANDLER / BEAUTY GENERATION
• Race 6: FRANTIC + ALWAYS PERFECT / FILLY FODEN
• Race 7: ALICE DE CLARE + PENTONVILLE / ZIPWIRE

Caution Marker List
• LION O – market weakness versus AU
• CROWNRIGHT – class-drop volatility
• HATAMOTO – beaten favourite LTO + cold jockey
• HAPPY CHANDLER – beaten favourite LTO
• FILLY FODEN – market weakness versus AU
• PENTONVILLE – first-time headgear

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

VALIDATION & TRUST LAYER

AU integrity

• Race 1: AU integrity evidenced — VICTORY GOLD led uploaded points totals with 15pts.
• Race 2: AU integrity evidenced — ABRAHAMSEN did not lead uploaded points totals; BLUESTONE LADY led with 10pts; ABRAHAMSEN retained by Rated to Win support and market compression.
• Race 3: AU integrity evidenced — CROWNRIGHT did not lead uploaded points totals; CAIM led with 15pts; CROWNRIGHT retained by Rated to Win and R&S Tips support.
• Race 4: AU integrity evidenced — HATAMOTO led uploaded points totals with 12pts.
• Race 5: AU integrity evidenced — MYNA led uploaded points totals with 12pts.
• Race 6: AU integrity evidenced — FRANTIC did not lead uploaded points totals; FILLY FODEN led with 14pts; FRANTIC retained by R&S Tips support and market compression.
• Race 7: AU integrity evidenced — ALICE DE CLARE led uploaded points totals with 13pts. SEE THE BLOG!

 

[AJ the Hobbyist]

▸ LINGFIELD TUESDAY 26 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Lingfield card built through the V15 audit lens.
AU figs, Smart Stats, market shape and caution markers kept separate.

• Six-race structure built from uploaded racecard, AU-style panels and Smart Stats layers
• AU hierarchy retained as the primary driver across the card
• Market prices used only for compression, support or caution — not override
• Caution stacks flagged where beaten-favourite, stable-switch, headgear or market weakness evidence appeared
• Forecast and TOTE structures bound to the same Win Pick anchor
• Smart Stats flags handled only where directly evidenced
• Big-field market weakness treated with harder discipline where relevant
• Tactical overlay system only — not a tipping service

Read the full card... Lingfield Tuesday 26 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: ACROSS EARTH
• Race 2: ASSET
• Race 3: MISS MAGIC DRAGON
• Race 4: ASCENDING STAR
• Race 5: THUNDEROUS LOVE
• Race 6: LUMINARE

Forecast Combos
• Race 1: ACROSS EARTH → SYNERGISM / RUSSIAN RUMOUR
• Race 2: ASSET → BINTKNIGHT / GREEN AREA
• Race 3: MISS MAGIC DRAGON → COUGAR FORCE / TICKETTOTHESTARS
• Race 4: ASCENDING STAR → WEFFAAG / STORMING POINT
• Race 5: THUNDEROUS LOVE → BELLA BISBEE / HILLTOP
• Race 6: LUMINARE → MIGHTY VEGA / VENETIAN ROMANCE

EW/Combo Value Inclusions
• RUSSIAN RUMOUR
• BINTKNIGHT
• TICKETTOTHESTARS
• STORMING POINT
• BELLA BISBEE
• VENETIAN ROMANCE

TOTE Combos Recap
• Race 1: ACROSS EARTH + SYNERGISM / RUSSIAN RUMOUR
• Race 2: ASSET + BINTKNIGHT / GREEN AREA
• Race 3: MISS MAGIC DRAGON + COUGAR FORCE / TICKETTOTHESTARS
• Race 4: ASCENDING STAR + WEFFAAG / STORMING POINT
• Race 5: THUNDEROUS LOVE + BELLA BISBEE / HILLTOP
• Race 6: LUMINARE + MIGHTY VEGA / VENETIAN ROMANCE

Caution Marker List
• ACROSS EARTH – beaten favourite LTO + stable switch + cold trainer evidenced from uploaded layers
• ASSET – cold jockey evidenced from uploaded Smart Stats layer
• MISS MAGIC DRAGON – beaten favourite LTO + market weakness versus AU in a big-field handicap evidenced from uploaded layers
• ASCENDING STAR – beaten favourite LTO + stable switch + market weakness versus AU evidenced from uploaded layers
• SUMMER HEAT – first-time tongue strap + cheek piece and cold trainer evidenced from uploaded layers
• LOHOOBB – beaten favourite LTO + market strength without leading AU points evidenced from uploaded layers

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

VALIDATION & TRUST LAYER

AU integrity

• Race 1: AU integrity evidenced — ACROSS EARTH led uploaded points totals with 12pts.
• Race 2: AU integrity evidenced — ASSET led uploaded points totals with 16pts.
• Race 3: AU integrity evidenced — MISS MAGIC DRAGON led uploaded points totals with 12pts.
• Race 4: AU integrity evidenced — ASCENDING STAR led uploaded points totals with 13pts.
• Race 5: AU integrity evidenced — THUNDEROUS LOVE led uploaded points totals with 10pts.
• Race 6: AU integrity evidenced — LUMINARE led uploaded points totals with 9pts. SEE THE BLOG!

 

[AJ the Hobbyist]

▸ LEICESTER TUESDAY 26 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE
Audit-based V15 structure built from uploaded AU figs, Smart Stats, racecard layers and market data.
No tipping language — structure, caution control and evidence discipline only.
• AU hierarchy retained as the primary driver across the card
• Win Pick binding carried through Forecast Combo and TOTE Anchor
• Market compression used only as support, not override
• Smart Stats flags applied only where evidenced
• Caution markers retained for BF LTO, class drop, headgear and market weakness
• Race-by-race AU integrity checked against uploaded points totals
• Forecast and TOTE structures built outward from the Win Pick
• Charter discipline locked: no simulation, no hindsight, no invented data
Read the full card... Leicester Tuesday 26 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: VICTORY GOLD
• Race 2: ABRAHAMSEN
• Race 3: CROWNRIGHT
• Race 4: HATAMOTO
• Race 5: MYNA
• Race 6: FRANTIC
• Race 7: ALICE DE CLARE

Forecast Combos
• Race 1: VICTORY GOLD → LIBERTANGO / MIAMI TO IBIZA
• Race 2: ABRAHAMSEN → PARADING / BLUESTONE LADY
• Race 3: CROWNRIGHT → CAIM / HARLEQUIN BREEZE
• Race 4: HATAMOTO → SHE'S CRAFTY / WOODY Y FERNANDEZ
• Race 5: MYNA → HAPPY CHANDLER / BEAUTY GENERATION
• Race 6: FRANTIC → ALWAYS PERFECT / FILLY FODEN
• Race 7: ALICE DE CLARE → PENTONVILLE / ZIPWIRE

EW/Combo Value Inclusions
• LIBERTANGO
• MIAMI TO IBIZA
• PARADING
• BLUESTONE LADY
• CAIM
• HARLEQUIN BREEZE
• SHE'S CRAFTY
• WOODY Y FERNANDEZ
• HAPPY CHANDLER
• BEAUTY GENERATION
• ALWAYS PERFECT
• FILLY FODEN
• PENTONVILLE
• ZIPWIRE

TOTE Combos Recap
• Race 1: VICTORY GOLD + LIBERTANGO / MIAMI TO IBIZA
• Race 2: ABRAHAMSEN + PARADING / BLUESTONE LADY
• Race 3: CROWNRIGHT + CAIM / HARLEQUIN BREEZE
• Race 4: HATAMOTO + SHE'S CRAFTY / WOODY Y FERNANDEZ
• Race 5: MYNA + HAPPY CHANDLER / BEAUTY GENERATION
• Race 6: FRANTIC + ALWAYS PERFECT / FILLY FODEN
• Race 7: ALICE DE CLARE + PENTONVILLE / ZIPWIRE

Caution Marker List
• LION O – market weakness versus AU
• CROWNRIGHT – class-drop volatility
• HATAMOTO – beaten favourite LTO + cold jockey
• HAPPY CHANDLER – beaten favourite LTO
• FILLY FODEN – market weakness versus AU
• PENTONVILLE – first-time headgear

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

VALIDATION & TRUST LAYER

AU integrity

• Race 1: AU integrity evidenced — VICTORY GOLD led uploaded points totals with 15pts.
• Race 2: AU integrity evidenced — ABRAHAMSEN did not lead uploaded points totals; BLUESTONE LADY led with 10pts; ABRAHAMSEN retained by Rated to Win support and market compression.
• Race 3: AU integrity evidenced — CROWNRIGHT did not lead uploaded points totals; CAIM led with 15pts; CROWNRIGHT retained by Rated to Win and R&S Tips support.
• Race 4: AU integrity evidenced — HATAMOTO led uploaded points totals with 12pts.
• Race 5: AU integrity evidenced — MYNA led uploaded points totals with 12pts.
• Race 6: AU integrity evidenced — FRANTIC did not lead uploaded points totals; FILLY FODEN led with 14pts; FRANTIC retained by R&S Tips support and market compression.
• Race 7: AU integrity evidenced — ALICE DE CLARE led uploaded points totals with 13pts. SEE THE BLOG!

▸ LEICESTER TUESDAY 26 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Leicester reviewed through the V15 audit lens.
Structure first, betting outcome separate, charter discipline retained.

• AU figs kept as the primary structural driver
• Smart Stats used only where directly evidenced
• Forecast zones reviewed against the original V15 build
• TOTE structure assessed through Win Pick anchor rules
• Caution markers retained for exposed risk and chaos control
• No post-race repair added to the original structure
• Betting result separated from model integrity
• Critique kept audit-grade, factual and structure-first

Read the full card and post-race critique: Leicester Tuesday 26 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

Cumulative Outcome Analysis

Structured Yankee:

• Stake: £3.30
• Return: £0.00
• Outcome: Lost
• Betting P/L: -£3.30

V15 Win Pick performance:

• Race 1: VICTORY GOLD — 4th
• Race 2: ABRAHAMSEN — 1st
• Race 3: CROWNRIGHT — 2nd
• Race 4: HATAMOTO — 3rd
• Race 5: MYNA — 4th
• Race 6: FRANTIC — 2nd
• Race 7: ALICE DE CLARE — 1st

V15 Exacta performance:

• Race 1: Failed
• Race 2: Landed
• Race 3: Failed
• Race 4: Failed
• Race 5: Failed
• Race 6: Failed
• Race 7: Landed

V15 Boxed Trifecta performance:

• Race 1: Failed
• Race 2: Failed
• Race 3: Failed
• Race 4: Failed
• Race 5: Failed
• Race 6: Failed
• Race 7: Failed

TOTE returns printed only where landed and officially listed:

• Race 2 Exacta: £3.50 (P/L: +£1.50)
• Race 7 Exacta: £2.00 (P/L: £0.00)

No failed TOTE bet receives a payout or P/L bracket.

Model Reflection & Refinement Notes

• The clearest structural success came when the Win Pick won and the first partner followed it home: Race 2 and Race 7.
• The main weakness was not total runner exclusion; it was Win Pick anchoring in races where a forecast partner or outside runner won.
• Race 1 and Race 3 show that partner strength was present, but the winner-first anchor was wrong.
• Race 4, Race 5 and Race 6 show framed or near-framed Win Picks without conversion, which protects some model integrity but fails the winner-first objective.
• Betting outcome and model integrity separate cleanly here: the Yankee lost entirely, while the V15 structure still produced two official Exacta landings from seven races.
• The refinement point is anchor discipline: where AU strength and market compression conflict, the critique should focus on whether the Win Pick carried too much unresolved caution.
• No post-race repair is applied.
• No unsupported race-shape explanation is added.
• No missing dividend is inferred.
• Charter discipline enforced.

 

[AJ the Hobbyist]

▸ LINGFIELD TUESDAY 26 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Lingfield card built through the V15 audit lens.
AU figs, Smart Stats, market shape and caution markers kept separate.

• Six-race structure built from uploaded racecard, AU-style panels and Smart Stats layers
• AU hierarchy retained as the primary driver across the card
• Market prices used only for compression, support or caution — not override
• Caution stacks flagged where beaten-favourite, stable-switch, headgear or market weakness evidence appeared
• Forecast and TOTE structures bound to the same Win Pick anchor
• Smart Stats flags handled only where directly evidenced
• Big-field market weakness treated with harder discipline where relevant
• Tactical overlay system only — not a tipping service

Read the full card... Lingfield Tuesday 26 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: ACROSS EARTH
• Race 2: ASSET
• Race 3: MISS MAGIC DRAGON
• Race 4: ASCENDING STAR
• Race 5: THUNDEROUS LOVE
• Race 6: LUMINARE

Forecast Combos
• Race 1: ACROSS EARTH → SYNERGISM / RUSSIAN RUMOUR
• Race 2: ASSET → BINTKNIGHT / GREEN AREA
• Race 3: MISS MAGIC DRAGON → COUGAR FORCE / TICKETTOTHESTARS
• Race 4: ASCENDING STAR → WEFFAAG / STORMING POINT
• Race 5: THUNDEROUS LOVE → BELLA BISBEE / HILLTOP
• Race 6: LUMINARE → MIGHTY VEGA / VENETIAN ROMANCE

EW/Combo Value Inclusions
• RUSSIAN RUMOUR
• BINTKNIGHT
• TICKETTOTHESTARS
• STORMING POINT
• BELLA BISBEE
• VENETIAN ROMANCE

TOTE Combos Recap
• Race 1: ACROSS EARTH + SYNERGISM / RUSSIAN RUMOUR
• Race 2: ASSET + BINTKNIGHT / GREEN AREA
• Race 3: MISS MAGIC DRAGON + COUGAR FORCE / TICKETTOTHESTARS
• Race 4: ASCENDING STAR + WEFFAAG / STORMING POINT
• Race 5: THUNDEROUS LOVE + BELLA BISBEE / HILLTOP
• Race 6: LUMINARE + MIGHTY VEGA / VENETIAN ROMANCE

Caution Marker List
• ACROSS EARTH – beaten favourite LTO + stable switch + cold trainer evidenced from uploaded layers
• ASSET – cold jockey evidenced from uploaded Smart Stats layer
• MISS MAGIC DRAGON – beaten favourite LTO + market weakness versus AU in a big-field handicap evidenced from uploaded layers
• ASCENDING STAR – beaten favourite LTO + stable switch + market weakness versus AU evidenced from uploaded layers
• SUMMER HEAT – first-time tongue strap + cheek piece and cold trainer evidenced from uploaded layers
• LOHOOBB – beaten favourite LTO + market strength without leading AU points evidenced from uploaded layers

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

VALIDATION & TRUST LAYER

AU integrity

• Race 1: AU integrity evidenced — ACROSS EARTH led uploaded points totals with 12pts.
• Race 2: AU integrity evidenced — ASSET led uploaded points totals with 16pts.
• Race 3: AU integrity evidenced — MISS MAGIC DRAGON led uploaded points totals with 12pts.
• Race 4: AU integrity evidenced — ASCENDING STAR led uploaded points totals with 13pts.
• Race 5: AU integrity evidenced — THUNDEROUS LOVE led uploaded points totals with 10pts.
• Race 6: AU integrity evidenced — LUMINARE led uploaded points totals with 9pts. SEE THE BLOG!

▸ LINGFIELD TUESDAY 26 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE
Lingfield reviewed through the V15 audit lens.
Structure first, betting outcome separate, charter discipline retained.
• AU figs kept as the primary structural driver
• Smart Stats used only where directly evidenced
• Forecast zones reviewed against the original V15 build
• TOTE structure assessed through Win Pick anchor rules
• Caution markers retained for exposed risk and chaos control
• Non-runner disruption kept separate from model integrity
• Market position used only for compression, support or caution
• Critique kept audit-grade, factual and structure-first
Read the full card and post-race critique: Lingfield Tuesday 26 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

Cumulative Outcome Analysis

Structured Yankee:

• Stake: £3.30
• Return: £0.00
• Outcome: Lost
• Betting P/L: -£3.30

V15 Win Pick performance:

• Race 1: ACROSS EARTH — NonRunner
• Race 2: ASSET — 2nd
• Race 3: MISS MAGIC DRAGON — unplaced
• Race 4: ASCENDING STAR — 1st
• Race 5: THUNDEROUS LOVE — unplaced
• Race 6: LUMINARE — 3rd

V15 Exacta performance:

• Race 1: Failed
• Race 2: Failed
• Race 3: Failed
• Race 4: Landed
• Race 5: Failed
• Race 6: Failed

V15 Boxed Trifecta performance:

• Race 1: Failed
• Race 2: Landed
• Race 3: Failed
• Race 4: Failed
• Race 5: Failed
• Race 6: Failed

TOTE returns printed only where landed and officially listed:

• Race 2 Trifecta: £7.00 (P/L: +£1.00)
• Race 4 Exacta: £6.10 (P/L: +£4.10)

No failed TOTE bet receives a payout or P/L bracket.

Model Reflection & Refinement Notes

• The clearest structural hold was Race 4, where ASCENDING STAR anchored the race correctly and WEFFAAG completed the Exacta structure.
• Race 2 held as a boxed Trifecta but failed the winner-first anchor test because ASSET did not win.
• Race 6 showed forecast-zone presence through MIGHTY VEGA winning and LUMINARE placing, but the Win Pick anchor was wrong.
• Race 3 and Race 5 were full anchor failures against the uploaded placed results.
• Race 1 cannot be repaired post-race because the selected Win Pick is listed as a NonRunner in the uploaded result.
• Betting outcome and model integrity separate cleanly: the Yankee lost, while the V15 structure produced one official Exacta and one official boxed Trifecta from six races.
• The refinement point is anchor discipline under caution pressure and non-runner disruption: partner strength is not enough if the Win Pick fails the winner-first requirement.
• No post-race repair is applied.
• No unsupported race-shape explanation is added.
• No missing dividend is inferred.
• Charter discipline enforced.

 

[AJ the Hobbyist]

▸ BEVERLEY WEDNESDAY 27 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Beverley card built through uploaded AU figs, Smart Stats, racecard layers and market structure.
Audit-first overlay only — not a tipping service.

• AU hierarchy retained as the primary structure across all seven races
• Market prices used only as alignment, compression or caution evidence
• Penny Arcade, Poet's Dawn and Freddy Robinson held the cleanest AU-led structures
• Hibernate and Thornaby Pearl required caution-aware handling where points leaders carried risk
• Smart Stats flags included hot/cold jockey-trainer evidence, headgear, class drops and weighted-to-win runners
• Caution markers included beaten-favourite LTO, headgear stacks, market weakness versus AU and draw-related risk
• Forecast and TOTE structures were built outward from the named Win Pick only

Read the full card... Beverley Wednesday 27 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: PENNY ARCADE
• Race 2: DEI GRATIA REGINA
• Race 3: WOODSTOCK
• Race 4: HIBERNATE
• Race 5: POET'S DAWN
• Race 6: FREDDY ROBINSON
• Race 7: THORNABY PEARL

Forecast Combos
• Race 1: PENNY ARCADE → LAIRY MARY / SPIRIT OF MARY
• Race 2: DEI GRATIA REGINA → SOU'WESTER / OUT OF THIS WORLD
• Race 3: WOODSTOCK → HARSWELL RUBY / RAJAPOUR
• Race 4: HIBERNATE → TRIPLE FORCE / DAWN OF LIBERATION
• Race 5: POET'S DAWN → HOSTELRY / RING OF GOLD
• Race 6: FREDDY ROBINSON → SISTERANDBROTHER / BRUCE BANNER
• Race 7: THORNABY PEARL → NORDIC GLORY / SOUTH SHORE

EW/Combo Value Inclusions
• LAIRY MARY
• SPIRIT OF MARY
• SOU'WESTER
• OUT OF THIS WORLD
• HARSWELL RUBY
• RAJAPOUR
• TRIPLE FORCE
• DAWN OF LIBERATION
• HOSTELRY
• RING OF GOLD
• SISTERANDBROTHER
• BRUCE BANNER
• NORDIC GLORY
• SOUTH SHORE

TOTE Combos Recap
• Race 1: PENNY ARCADE + LAIRY MARY / SPIRIT OF MARY
• Race 2: DEI GRATIA REGINA + SOU'WESTER / OUT OF THIS WORLD
• Race 3: WOODSTOCK + HARSWELL RUBY / RAJAPOUR
• Race 4: HIBERNATE + TRIPLE FORCE / DAWN OF LIBERATION
• Race 5: POET'S DAWN + HOSTELRY / RING OF GOLD
• Race 6: FREDDY ROBINSON + SISTERANDBROTHER / BRUCE BANNER
• Race 7: THORNABY PEARL + NORDIC GLORY / SOUTH SHORE

Caution Marker List
• SPIRIT OF MARY – class-drop volatility, cold jockey and market weakness versus AU evidenced.
• HARSWELL RUBY – beaten favourite last time out.
• TRIPLE FORCE – first-time visor, cold trainer and market weakness versus AU evidenced.
• REGAL GLORY – beaten favourite last time out and headgear evidenced.
• NORDIC GLORY – market weakness versus AU and wide draw evidenced.

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

VALIDATION & TRUST LAYER

AU integrity

• Race 1: AU integrity evidenced — PENNY ARCADE led uploaded points totals with 18pts.
• Race 2: AU integrity evidenced — DEI GRATIA REGINA led uploaded points totals with 12pts.
• Race 3: AU integrity evidenced — WOODSTOCK led uploaded points totals with 7pts.
• Race 4: AU integrity evidenced — HIBERNATE held 10pts; TRIPLE FORCE led uploaded points totals with 13pts; HIBERNATE retained by R&S Tips support, market alignment and caution isolation on TRIPLE FORCE.
• Race 5: AU integrity evidenced — POET'S DAWN led uploaded points totals with 16pts.
• Race 6: AU integrity evidenced — FREDDY ROBINSON led uploaded points totals with 14pts.
• Race 7: AU integrity evidenced — THORNABY PEARL held 7pts; NORDIC GLORY led uploaded points totals with 11pts; THORNABY PEARL retained by R&S Tips support and caution isolation on NORDIC GLORY. SEE THE BLOG!

 

[AJ the Hobbyist]

▸ KEMPTON WEDNESDAY 27 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE
Kempton built through AU-first structure, smart stats checks and market alignment.
Forecast and TOTE shape stay audit-led, not outcome-led.
• 8-race Kempton AW card structured from uploaded layers only
• AU hierarchy retained as the primary build driver
• Market prices used for compression, not override
• Smart Stats flags checked across headgear, BF LTO, class drops and switchers
• Caution markers isolated where evidenced
• Forecast combos built outward from the named Win Pick
• TOTE anchors bound to the same Win Pick structure
• Not a tipping service; no simulation or hindsight logic
Read the full card... Kempton Wednesday 27 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: Kit Gabriel
• Race 2: Phalanx Nation
• Race 3: Erudition
• Race 4: Storming Point
• Race 5: Water To Wine
• Race 6: Ya Karim
• Race 7: Bulletin
• Race 8: Moon Chime

Forecast Combos
• Race 1: Kit Gabriel → Tex / Valentine Boy
• Race 2: Phalanx Nation → Esha'a / Moonrise
• Race 3: Erudition → Makerstoun / Khaleejy
• Race 4: Storming Point → Masked Warrior / Tyrant Gg
• Race 5: Water To Wine → Tripoli Flyer / Barrister
• Race 6: Ya Karim → Caviar Cowboy / Exotic Baby
• Race 7: Bulletin → Ernst Blofeld / Son Of Man
• Race 8: Moon Chime → Senor Cortez / Filibustering

EW/Combo Value Inclusions
• Tex
• Valentine Boy
• Esha'a
• Moonrise
• Makerstoun
• Khaleejy
• Masked Warrior
• Tyrant Gg
• Tripoli Flyer
• Barrister
• Caviar Cowboy
• Exotic Baby
• Ernst Blofeld
• Son Of Man
• Senor Cortez
• Filibustering

TOTE Combos Recap
• Race 1: Kit Gabriel + Tex / Valentine Boy
• Race 2: Phalanx Nation + Esha'a / Moonrise
• Race 3: Erudition + Makerstoun / Khaleejy
• Race 4: Storming Point + Masked Warrior / Tyrant Gg
• Race 5: Water To Wine + Tripoli Flyer / Barrister
• Race 6: Ya Karim + Caviar Cowboy / Exotic Baby
• Race 7: Bulletin + Ernst Blofeld / Son Of Man
• Race 8: Moon Chime + Senor Cortez / Filibustering

Caution Marker List
• Kit Gabriel – cold jockey Mason Paetel and cold trainer S Pearce evidenced in Smart Stats.
• Phalanx Nation – beaten favourite last time out evidenced in Smart Stats.
• Masked Warrior – first-time cheekpieces evidenced in uploaded Smart Stats.
• Exotic Baby – market weakness versus AU evidenced in a big-field handicap.
• Son Of Man – market weakness versus AU evidenced in a big-field handicap.
• Filibustering – class drop and stable switch evidenced in uploaded Smart Stats.

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

VALIDATION & TRUST LAYER

AU integrity

• Race 1: AU integrity evidenced — Tex led uploaded points totals with 8pts; Kit Gabriel retained by Rated to Win and R&S Tips tie-break support.
• Race 2: AU integrity evidenced — Phalanx Nation led uploaded points totals with 15pts.
• Race 3: AU integrity evidenced — Erudition and Khaleejy tied on 14pts; Erudition retained by Rated to Win and R&S Tips support.
• Race 4: AU integrity evidenced — Storming Point led uploaded points totals with 13pts.
• Race 5: AU integrity evidenced — Water To Wine led uploaded points totals with 16pts.
• Race 6: AU integrity evidenced — Ya Karim led uploaded points totals with 8pts.
• Race 7: AU integrity evidenced — Bulletin led uploaded points totals with 11pts.
• Race 8: AU integrity evidenced — Moon Chime led uploaded points totals with 15pts.

 

[AJ the Hobbyist]

▸ BEVERLEY WEDNESDAY 27 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Beverley card built through uploaded AU figs, Smart Stats, racecard layers and market structure.
Audit-first overlay only — not a tipping service.

• AU hierarchy retained as the primary structure across all seven races
• Market prices used only as alignment, compression or caution evidence
• Penny Arcade, Poet's Dawn and Freddy Robinson held the cleanest AU-led structures
• Hibernate and Thornaby Pearl required caution-aware handling where points leaders carried risk
• Smart Stats flags included hot/cold jockey-trainer evidence, headgear, class drops and weighted-to-win runners
• Caution markers included beaten-favourite LTO, headgear stacks, market weakness versus AU and draw-related risk
• Forecast and TOTE structures were built outward from the named Win Pick only

Read the full card... Beverley Wednesday 27 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

FINAL SUMMARY SECTION

Top Win Picks
• Race 1: PENNY ARCADE
• Race 2: DEI GRATIA REGINA
• Race 3: WOODSTOCK
• Race 4: HIBERNATE
• Race 5: POET'S DAWN
• Race 6: FREDDY ROBINSON
• Race 7: THORNABY PEARL

Forecast Combos
• Race 1: PENNY ARCADE → LAIRY MARY / SPIRIT OF MARY
• Race 2: DEI GRATIA REGINA → SOU'WESTER / OUT OF THIS WORLD
• Race 3: WOODSTOCK → HARSWELL RUBY / RAJAPOUR
• Race 4: HIBERNATE → TRIPLE FORCE / DAWN OF LIBERATION
• Race 5: POET'S DAWN → HOSTELRY / RING OF GOLD
• Race 6: FREDDY ROBINSON → SISTERANDBROTHER / BRUCE BANNER
• Race 7: THORNABY PEARL → NORDIC GLORY / SOUTH SHORE

EW/Combo Value Inclusions
• LAIRY MARY
• SPIRIT OF MARY
• SOU'WESTER
• OUT OF THIS WORLD
• HARSWELL RUBY
• RAJAPOUR
• TRIPLE FORCE
• DAWN OF LIBERATION
• HOSTELRY
• RING OF GOLD
• SISTERANDBROTHER
• BRUCE BANNER
• NORDIC GLORY
• SOUTH SHORE

TOTE Combos Recap
• Race 1: PENNY ARCADE + LAIRY MARY / SPIRIT OF MARY
• Race 2: DEI GRATIA REGINA + SOU'WESTER / OUT OF THIS WORLD
• Race 3: WOODSTOCK + HARSWELL RUBY / RAJAPOUR
• Race 4: HIBERNATE + TRIPLE FORCE / DAWN OF LIBERATION
• Race 5: POET'S DAWN + HOSTELRY / RING OF GOLD
• Race 6: FREDDY ROBINSON + SISTERANDBROTHER / BRUCE BANNER
• Race 7: THORNABY PEARL + NORDIC GLORY / SOUTH SHORE

Caution Marker List
• SPIRIT OF MARY – class-drop volatility, cold jockey and market weakness versus AU evidenced.
• HARSWELL RUBY – beaten favourite last time out.
• TRIPLE FORCE – first-time visor, cold trainer and market weakness versus AU evidenced.
• REGAL GLORY – beaten favourite last time out and headgear evidenced.
• NORDIC GLORY – market weakness versus AU and wide draw evidenced.

Charter Reminder:

This is a tactical overlay system, not a tipping blog.
No simulation.

VALIDATION & TRUST LAYER

AU integrity

• Race 1: AU integrity evidenced — PENNY ARCADE led uploaded points totals with 18pts.
• Race 2: AU integrity evidenced — DEI GRATIA REGINA led uploaded points totals with 12pts.
• Race 3: AU integrity evidenced — WOODSTOCK led uploaded points totals with 7pts.
• Race 4: AU integrity evidenced — HIBERNATE held 10pts; TRIPLE FORCE led uploaded points totals with 13pts; HIBERNATE retained by R&S Tips support, market alignment and caution isolation on TRIPLE FORCE.
• Race 5: AU integrity evidenced — POET'S DAWN led uploaded points totals with 16pts.
• Race 6: AU integrity evidenced — FREDDY ROBINSON led uploaded points totals with 14pts.
• Race 7: AU integrity evidenced — THORNABY PEARL held 7pts; NORDIC GLORY led uploaded points totals with 11pts; THORNABY PEARL retained by R&S Tips support and caution isolation on NORDIC GLORY. SEE THE BLOG!

▸ BEVERLEY WEDNESDAY 27 MAY 2026 — V15 EARLY DOORS | LIVE STRUCTURE

Beverley handled through the V15 audit layer: AU figs first, Smart Stats second, market structure controlled.
Structure-first review only, with the post-race critique kept separate from tipping language.

• AU figs remained the primary driver across the card
• Smart Stats flags were logged for hot/cold riders, trainers, headgear and weighted-to-win runners
• Forecast zones were built outward from the named Win Pick only
• Market prices were treated as compression, alignment or caution evidence
• Caution markers were used to control exposed profiles rather than force upgrades
• TOTE structure was reviewed through exacta and boxed-trifecta logic only
• Post-race critique separates betting outcome from model integrity

Read the full card and post-race critique: Beverley Wednesday 27 May 2026 – V15 Overlay Blog | Tactical Forecasts + Smart Stats Integration | Horse Racing Hobbyist

Structure first. Noise second.
Cumulative Outcome Analysis

Structured Yankee: LOST
Stake: £3.30
Returns: £0.00

V15 Exacta outcomes:
Race 1 — FAILED
Race 2 — FAILED
Race 3 — FAILED
Race 4 — FAILED
Race 5 — FAILED
Race 6 — FAILED
Race 7 — FAILED

V15 Boxed Trifecta outcomes:
Race 1 — FAILED
Race 2 — FAILED
Race 3 — FAILED
Race 4 — FAILED
Race 5 — FAILED
Race 6 — LANDED
Race 7 — FAILED

Only Race 6 produced a valid boxed Trifecta under the uploaded result and hard-rule conditions.

The official Race 6 Tote Trifecta dividend was £15.10, giving a £9.10 profit against the £6 boxed Trifecta stake.

No Exacta landed because no V15 Win Pick both won and had a forecast partner finish 2nd.

No other TOTE P/L brackets are printed because the qualifying conditions were not met.

Model Reflection & Refinement Notes

What held structurally:

Race 6 held as a boxed three-runner structure.

Race 1 partially held through Lairy Mary winning and Penny Arcade placing, but the full structure did not land.

Race 2 partially held through Out Of This World and Dei Gratia Regina filling the first two positions, but the order and third position failed.

Race 6 showed that partner strength can preserve TOTE structure even when the named Win Pick anchor fails.

What failed structurally:

The Win Pick layer underperformed across the card.

The Exacta structure failed in every race because the V15 Win Pick did not win.

Race 3 exposed a structural miss where Coolree’s course-linked evidence was present but not promoted into the forecast trio.

Race 4 exposed that avoiding a caution-loaded points leader did not automatically create a cleaner winning anchor.

Race 5 failed at both anchor and partner level.

Race 7 failed despite caution-aware handling, with Thornaby Pearl finishing 4th and South Shore the only placed forecast runner.

Betting outcome versus model integrity:

The Yankee loss is a betting-structure failure.

The V15 model did not fully collapse because Race 6 boxed Trifecta landed and several races contained partial structural reads.

However, the primary Win Pick objective failed across the card.

The day should not be over-credited from partial places or partner survival.

Refinement notes:

Yankee-style exposure remains unsuitable when confidence is uneven across legs.

Win-only multiples should not be built from mixed-confidence anchors where the V15 structure itself shows caution pressure.

Course-linked H4C evidence needs stricter review when it appears outside the selected trio, especially where it later proves structurally relevant.

Partner survival must not be treated as anchor success.

Boxed-trifecta survival is useful, but it does not repair failed Win Pick performance.

Caution-aware downgrades remain valid only when the replacement anchor has sufficiently clean AU and market support.

Final classification:

Win Pick layer: FAILED
Exacta layer: FAILED
Boxed Trifecta layer: PARTIAL HOLD
Yankee betting structure: FAILED
Model integrity: DAMAGED BUT NOT BROKEN
Carry-forward lesson: tighten Win Pick approval where AU points, market position and caution evidence conflict.