’RixPicks

What the system learned1-1, -5.73u on Saturday's 2-pick CFB card. What won: Ohio Bobcats ML (-117, 5u) dominated Central Michigan 31-14 (+4.27u) behind explosive second-half drives. Losses named plainly: Pittsburgh Panthers ML (-170, 10u, -10.00u) fell 21-17 to North Carolina after fourth-quarter stalled drives failed to overcome the deficit.

What the system learned1-2, -4.80u on a 3-pick Friday slate across NHL and CFB. What won: Iowa State/BYU Under 45.5 (+104, 34 total points vs 45.5) cashed cleanly (+5.20u) as BYU's defense held Iowa State to 10 points and controlled tempo throughout. Losses named plainly: Rangers/Capitals Over 5.5 (-108, 3 goals vs 5.5) was choked off by elite goaltending in a 2-1 duel; San Jose State ML (65.6% Kalshi / -191, lost 16-13 in OT) saw a 13-6 4th-quarter lead slip away on a late Wyoming touchdown before falling in overtime.

What the system learned3-4, -2.94u on a 7-pick slate across NFL, CFB, and NHL. What won: Buccaneers/Cowboys Under 48.5 (-104, 40 total points) cashed cleanly as TNF short-week pace held; Mammoth/Bruins Over 5.5 (-122, 7 goals) hit with ease as Boston's offense dominated; Nashville Predators ML (+163, won 5-2 at Montreal) delivered a major plus-money cash (+8.15u). Losses named plainly: Wild/Lightning Over 5.5 (5 goals vs 5.5) fell half a goal short against elite goaltending; Missouri State ML (+108) dropped 34-13 on turnovers; Sabres ML (+108) was shut out 4-0 by Dallas; and Flyers ML (+163) lost a tight 2-1 defensive duel.

What the system learned2-1, +5.85u on the day across 3 carded picks. What won: Jax State/Kennesaw St Over 49.5 (53 points vs 49.5, model 49.4 vs Polymarket 47c ask +113) as 4th-quarter tempo surged to confirm full-game pace projections despite early drive stalls; Brewers/Padres Under 7.5 (4 runs vs 7.5, model 50.4 vs Polymarket 49c ask +104) in a 3-1 Petco Park elimination duel as manager urgency, short bullpen leashes, and heavy night marine air suppressed fly balls. Loss named plainly: Liberty/Dream Under 170.5 (+104, 199 points vs 170.5) died in a 101-98 overtime shootout, showing unpriced late-game foul and OT tail risk on tight WNBA playoff totals.

What the system learnedTwo losses, both owner positions posted under a one-day sub-bar suspension. Against the de-vigged book close, Penguins ML paid 62c vs about 61% (a 3-2 loss, mostly variance), and Flyers +1.5 paid 58c vs about 54%: a pricing miss before puck drop, because the NHL numbers ran high and the puck-line price ignored late empty-net goals. NHL model prices stay off the card until the rebuilt NHL model passes its market test.

What the system learnedTwo wins, both owner-directed exceptions: Dream ML's recorded net edge was about 1.35c, below the 2c bar; Brewers ML's 59.4% fair was below the 60c band and it was posted under a one-day suspension. The recorded results were NY 82 @ ATL 92 and SD 3 @ MIL 4.

What the system learnedSix owner-directed picks, all under the system's bars that morning, and four won. Rays ML (fair 55% against a 56c price), White Sox +1.5 (0.2c over the bar on a hand-adjusted fair), Padres +1.5 (corrected Oct 4 from the card's +2.5; net of the correction it posted at -122) and Lightning ML (about 1c edge, price moved 60c to 65c, goalie unconfirmed) all won. Both NHL Over 4.5 totals lost at 4 goals, after the red team had flagged the NHL totals model as running about 3c high. Results: Rays ML W, White Sox +1.5 W, Padres +1.5 W, Lightning ML W, Kraken-Oilers Over 4.5 L, Capitals-Lightning Over 4.5 L.

What the system learnedNo official picks cleared the bars this morning. The owner then put his own four positions on the record. Blues-Stars Under 6.5 went up before the delayed opening faceoff; the other three were added by owner override after their finals, results known, and are labeled so. Results: Under 6.5 W, Under 3.5 W, Over 43.5 W, Under 54.5 L.

What the system learned3-1, +3.93u. The Devils (3-2 in OT), Aces -4.5 (won by 11) and Mammoth (6-0) won; Under 38.5 lost, with Steelers-Browns going for 51. This card was built by hand outside the system's process: no fair values recorded, prices that were not the Kalshi ask, a Las Vegas team (the Aces) and a 6u stake that is not on the ladder. It ran on Instagram only, at 8:03 AM PT before the first game, while the site said no official picks; the row was filed a day late and kept by the owner's ruling. Its 'Yesterday 2-0, Overall 6-4' line was wrong too (it was 1-0 and 21-11). The Vegas rule and the ladder are now hard stops in the card builder.

What the system learned1-0, +6.90u on a one-pick day. White Sox ML at Houston (+138) went up by owner directive with the sub-bar disclosure on the card: the model had Chicago at 46.7%, below the 60% band, though the price edge cleared the 2c floor. It was posted on the site at 7:34 AM PT and on Instagram before the 2:00 PM PT first pitch, and it won 7-3. One underdog win is one result; it does not show the band is set wrong.

What the system learned2-3, -7.19u. The morning read found no qualifying edge, and at 7:24 AM PT the site said no card. Later that morning five picks went up by owner directive, before the first game, with a sub-bar disclosure on the card. The card showed the model on each: Braves 64.3%, Maple Leafs 51.7%, Bruins 51.1%, Yordan Alvarez two-plus hits 31.3%, Abushaar 26.5%. Only the Braves sat inside the 60% band, and the Braves won; of the four below it, only the Bruins won. Five picks on one day are too few to judge owner-directed picks, so the day is graded on results and no read conclusion is drawn.

What the system learned5-2, +1.16u on the day across 7 carded picks. Closing-line value on the 7 picks with a closing line on file: avg +2.3pp (+5.0, +2.9, +0.1, +5.4, -0.5, +1.4, +1.9pp). What won: Lions ML vs Jets (Jahmyr Gibbs' 3rd touchdown with 2:25 left lifts Lions to 31-24 win over Jets); Phillies ML vs Rays (Phillies clinch NL wild card behind Wheeler for 5th straight playoff appearance with win over Rays); Brewers ML vs Cardinals (Misiorowski continues Cy Young Award push and Turang gets 100th RBI as Brewers beat Cardinals 6-4); Dodgers ML @ Giants (Call's RBI double, Pages' 3-run HR in 10th lift Dodgers to win No. 100, 5-1 over Giants); Dream ML vs Mystics (Howard, Reese help Dream rally from 11-point deficit, beat Mystics 92-77 in Game 1 of playoff series). Losses named plainly: Bengals ML @ Steelers (L 27-30, win probability peaked at 71% before the collapse); Lynx ML vs Liberty (L 75-91, win probability peaked at 77% before the collapse). All 7 results verified twice against the sheet; every fact above is pulled from the sheet, ESPN summary feeds, and closing lines on file - engine output, not a hand summary. (corrected Oct 6: the Bengals and Dodgers figures now read from their own side; the first version used the home team's, for an average of -6.8pp and a Bengals peak of 99%+)

What the system learnedThe market's margin was the misprice, not the side. MSST was a thin-edge home favorite the model liked but couldn't card (1.2pp below bar) while books hung 66c against a Mizzou team traveling without its starting EDGE and lead RB - a 2.6c book-vs-exchange divergence confirmed the gap. It sweated (55/56 tied in Q4) and won; DraftKings closed MSST at -175 against the -205 card - CLV -3.6pp. (corrected Oct 6: the first version said it closed 99/100 for +33pp CLV; 99/100 is not a pre-game price, so the DraftKings close from ESPN is shown)

What the system learned2-1, -0.87u. Baltimore and Cleveland were clean reads that won; Houston was the right price at the wrong outcome - led 1-0, gave up six in the 6th, rallied from 6-1 down to 6-5, and the last two threats died on a strikeout with the tying run on third and a game-ending double play. The +437 parlay died with the Houston leg. Pricing discipline held against the DraftKings close: all three picks beat it (+4.3, +3.3 and +3.8pp at ESPN). (corrected Oct 6: the first version said all three beat or matched the Kalshi close; the system's grade ledger has Baltimore 0.8pp behind that close)

What the system learned2-1 with the loss named plainly: Boston was a process failure, not bad luck - an assumption shipped where a check belonged, and it cost 5u and the +339 parlay with it. The rule that came out of it: probable pitchers and splits get verified, never assumed. Arizona beat the close, the Yankees won at a worse-than-close price, and the day finished +0.80u.

What the system learnedPerfect day, 2-0, +7.88u. Honest note even in a sweep: the ARI win came through a different door than the thesis predicted (unearned runs, not the rookie getting hit), so the system now requires a direct confirmation path when a pick leans on one pitcher imploding. Wins get audited too, not just losses.

What the system learnedSplit the day 4-4 and still lost 14.44u. The three MLB losses were two shutouts and a blowout - read failures the gate has to catch earlier. The WNBA lesson was pure math: massive minus prices risk a lot to win pennies, and one overtime loss at -1011 cost more than six times what both WNBA wins paid. Chalk sizing rules tightened from this day.