a *ej@sdZddlZddlZejddddlZddlZhdZ hdZ hdZ dZ dd Z d d Zdd dZddZddZddZedkredS)a Is Pinnacle structurally mis-calibrated in SOUTH AMERICA? (read-only, trades nothing) Built 2026-07-25 from the user's hypothesis: "there's something structurally wrong with South American leagues." Pooling regions is the point -- no single SA league has the sample to say anything (BSA n=32, PE1 n=12, CL1 n=8), but pooled they do. Same engine as pin_calibration_by_league.py: EVERY fixture with Pinnacle odds and a result, not just the ones we bet. THE THREE CONTROLS THAT MAKE THIS MORE THAN A SCAN: 1. DROP THE MOTIVATING LEAGUE. The hypothesis came from BSA's 0-for-12, and BSA is 44% of the SA sample. If the pattern is really just BSA, removing it kills the effect. (It doesn't -- the effect gets STRONGER.) 2. SIGN CONSISTENCY across independent leagues. pin_calibration_by_league's own standing rule: randomly-signed deviations at small n = noise; a real structural bias is consistent across comparable leagues. Reported as a sign test. 3. WHICH SIDE WE ACTUALLY BET. A calibration error only costs us money if it lands on the side we take. This is the step that stops a real finding from being mis-read as an explanation for a bad record. RESULT ON FIRST RUN (2026-07-25): SA n=73: homes implied 48.2% -> actual 60.3% (z+2.06); draws 27.4% -> 16.4% (z-2.10); aways 24.4% -> 23.3% (z-0.21). Excluding BSA (n=41) the home/draw effect PERSISTS (z+1.87 / -1.55). Home z>0 in 5/5 SA leagues (sign test p=.062). Europe/N.America/Asia all flat (|z| <= 1.3). BUT: 25 of our 28 SA bets are AWAY, and away is the one side calibrated fine. So the finding does NOT explain the -35.3% SA record (which is z=-0.46 on win count = noise). If anything the home under-pricing is an OPPORTUNITY. Usage: python research\region_calibration_test.py NzE:\soccer\research>ZPE1ZAR1BSAZBR2ZCO1ZCL1>ZUSLZNWSLZMX1ZCANZMLS>ZCN1ZKO1ZJP1rcCs(|tvr dS|tvrdS|tvr$dSdS)N S.America N.AmericaAsiaEurope)SANAASIAlgr -E:\soccer\research\region_calibration_test.pyregion/srcCs6t||d|}|dkr.||||StdS)Nrnan)mathsqrtfloat)npZactualZsdr r r z_for9sr?cCsdt|d||dS)zCSmallest calibration error (pp) detectable at 95% conf / 80% power.gffffff@rd)rr)rbaser r r detectable>src st|}|dkr.td|dd|dddSd|dd|dg}dD]t\tfdd |D|}tfd d |D}|d |d d d||d d dt|||ddqH|dt|ddtd|dS)N Z26s n=4dz -- too few))pfhH)pfdD)ZpfaAc3s|]}|VqdS)Nr .0xkeyr r Jzline..c3s|]}|dkVqdS)ftrNr r$)letterr r r)Kr* r5.1fz->z (z+5.2f)z | detects z.0fZpp)lenprintsumappendrrjoin)sZlabelroutimpZcntr )r(r,r lineCs & r9cs6t}tddD]tfdd|Dqtdtdtdd|Ddtd d|Dd ttd d|Dtd td g}ttD]҉fdd|D}t|}|dkrqtdd|D|}tdd|D}tdd|D|}tdd|D}| |t |||t |||ftddd|dd|dddd|dddqt|}tdd|D} td d|D} td!| d"|d#| d"|td$|d%dd&|d'd(td)g} t D]>} t | } | r| d*t j kr| d+tv| d,<| | qd-d.d/fd0d1d/ffD]\}fd2d| D}d3D]fd4d|D}|sqht|}td5d|D}td6d|D|}td|d7d8d9d|d:d;|d<d=||d>d?d@tdAd|D|d>d?dB|d>dCdDqhtqJtdEdS)FNz?A) PINNACLE CALIBRATION BY REGION (all fixtures with a result) )rrrrcs g|]}t|dkr|qSr )rr%f)rgr r Wr*zmain..z( B) CONTROL -- does SA survive dropping z( (the league that raised the question)? cSsg|]}|dtvr|qSr )rr:r r r r=[r*zSA allcSs(g|] }|dtvr|dtkr|qSr )r MOTIVATINGr:r r r r=\r*z SA excluding cSsg|]}|dtkr|qSr )r>r:r r r r=^r*z alonezB C) SIGN CONSISTENCY across SA leagues (noise is randomly signed) csg|]}|dkr|qSr r r:r r r r=cr*rcss|]}|dVqdS)rNr r$r r r r)gr*zmain..css|]}|ddkVqdS)r+r Nr r$r r r r)hr*css|]}|dVqdS)r!Nr r$r r r r)ir*css|]}|ddkVqdS)r+r"Nr r$r r r r)jr*rZ6srrz home zr/z draw zcss|]}|ddkrdVqdS)r@rrNr r%rr r r r)nr*css|]}|ddkrdVqdS)rArrNr rBr r r r)or*z home z>0 in /z leagues, draw z<0 in z P(all z one way by chance) = g?z.3fz (two-tailed sign test)zA D) DOES IT LAND ON THE SIDE WE BET? (forward bets, post-freeze) gdZleaguesaz South AmericacSs|dSNrFr br r r zr*zmain..z rest of worldcSs |d SrGr rHr r r rJ{r*csg|]}|r|qSr r r%rI)selr r r=|r*)homeawaycsg|]}|dkr|qSsider rKrOr r r=~r*css|]}|dVqdS)wonNr rKr r r r)r*css*|]"}|drd|ddndVqdS)rQrpxr?Nr rKr r r r)r*Z14sr-Z5sZ3dz W=Z2dz WR=rr.z % implied=css|]}|dVqdS)rRNr rKr r r r)r*z% ROI=z+7.1f%zD A calibration error only costs money on the side we actually take.)PZ load_fixturesr2r9r>sortedrr1r3r4rFfire_with_sideZ FREEZE_DATE)fxZrowsr6rZiHZcHZiDZcDkphpd_Zbetsr;rIZlblZsswZroir )r r<rLrPr mainRsj ":     4 r]c Cs,|dD]}t|dt|dt|d}}}d|||fvs|dkr|dkr|dksjqt|||f}g}tjD]\}}t|d|\} } t|d|\} } | dus| dusd | krdkrnqd | krdksqq| | kr| | | } }}n| | | } }}|dusd |krBdksFqq||| ||}|||||}| |krtjd |krtjdksqqtjd |krtjdkrqnq|d d kr|d kp|d d ko|dk}|t ||||dq|rt |dddSqdS)zHfire_xlow, but keeping which side qualified (the base version drops it).ZpreZpin_homeZpin_drawZpin_awayNrZkalshiZpolyrr+r rMr#rN)clvrRrQrPcSs|dS)Nr^r )dr r r rJr*z fire_with_side..r') rVZ_fgetZ fair_multZSIDXitemsZ book_quoteZBANDr4dictmax)rXrCrZr[ZpaZpfZcandsrPZsiZkmZkaZpmZpa_ZmidrRotherr^ZocrQr r r rWs> &> ".(rW__main__)r)__doc__sysrpathinsertZpin_calibration_by_leaguerTZforward_test_unique_bookrVrrr r>rrrr9r]rW__name__r r r r s"!  8"