§ Week 8 · HAWKISH

The week, scored.

For each question: how the group spread their bets, what actually happened, and the calibration lesson it taught.

#01 · sportsRESOLVED · NO

Will the winning score at the Rocket Classic (Detroit Golf Club, final round Sunday August 2) be 20-under-par or better? Resolves YES if the champion's 72-hole total is 20 strokes under par or lower (e.g. -20, -24), per the official PGA TOUR final leaderboard. If the tournament is shortened to fewer than 72 holes, resolves per the official winning score as declared by the PGA TOUR relative to the par for holes played.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 58.4% outcome bucket
§ lesson

NO. Michael Thorbjornsen won the Rocket Classic at 262, which is 18-under on a par-70 Detroit Golf Club, two clear of Xander Schauffele. The bar was 20-under or better. The trap: Detroit Golf Club played PAR 70 in 2026 for the first time (the 7th and 17th were converted from par 5s). Every pre-2026 winning score in the record book was made on a par 72, so -20 looked far more reachable than it actually was. The eight players who went above 65% here were all pricing history that no longer applies.

#02 · sportsRESOLVED · NO

Will the men's singles final at the DC Open (Washington, scheduled Sunday August 2) go to a deciding third set? Resolves YES if the final is decided in three sets. Resolves NO if it ends in straight sets, or if the match ends by retirement or walkover before a third set begins. If the final is not played before this week resolves, resolves NO.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 44.3% outcome bucket
§ lesson

NO -- by the question's own written rule, not by a scoreline. Daylong rain and thunderstorms in Washington forced the DC Open to call off the men's singles final between Taylor Fritz and Rafael Jodar at roughly 9:10pm ET on Sunday August 2, rescheduling it for Monday afternoon August 3 behind the women's final resuming at noon. This week resolves Monday August 3 at 08:00 UTC, which is 4:00am ET -- so at resolution time not a third set but not even a first set had been played. The question states: 'If the final is not played before this week resolves, resolves NO.' That clause was written for exactly this contingency, so unlike q89 in Week 7 this is NOT a void. The rule was on the page before anyone submitted.

#03 · sportsRESOLVED · YES

Will any MLB game go 12 or more innings between Wednesday July 29 and Sunday August 2, 2026 (inclusive)? Resolves YES if at least one regular-season MLB game in that window records a completed 12th inning or beyond, per official box scores.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 54.1% outcome bucket
§ lesson

YES, and it was over on day one. Two games reached a completed 12th inning on Wednesday July 29: Baltimore 10 at Detroit 9, and the NY Yankees 5 at the Chicago White Sox 6. Per official MLB Stats API box scores. Across a five-day, 71-game window a 12-inning game is close to a coin-flip-free proposition; the field averaged 52% on something that has fired in the vast majority of comparable windows. Daniel at 20% and Luna at 30% were pricing a single game, not a five-day window.

#04 · sportsRESOLVED · YES

Will home teams win MORE THAN HALF of the MLB games completed on Saturday August 1, 2026? Resolves per official box scores for games completed on that calendar date. An exact 50/50 split resolves NO ('more than half' is strict). Suspended games count on the date they conclude. Neutral-site games, if any, are excluded from both the numerator and the denominator.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 57.0% outcome bucket
§ lesson

YES, by one game. Of the 15 MLB games completed Saturday August 1, home teams won 8 and away teams won 7 -- 53.3%, strictly more than half. Per official box scores. This was the tightest question of the week: flip any one result and it is 7-8 and resolves NO. The field clustered 52-67% around a true base rate near 53%, which is honest pricing -- nobody should feel clever or robbed here.

#05 · marketsRESOLVED · NO

Will Bitcoin close ABOVE $66,000 on Friday July 31, 2026 (Coinbase BTC-USD, 5:00pm ET)? Resolves per the Coinbase BTC-USD price at 5:00pm ET (21:00 UTC) on July 31; $66,000.00 or below resolves NO.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 35.8% outcome bucket
§ lesson

NO, and not close. Coinbase BTC-USD printed 62,853.05 US dollars at 21:00 UTC (5:00pm ET) on Friday July 31, roughly 3,150 dollars below the 66,000 line. Bitcoin was at 65,158 when this question was authored on July 27 and traded down all week. Fisherman at 66% and Brasus at 57% were the only players on the wrong side of a move that was already underway.

#06 · marketsRESOLVED · NO

Will the S&P 500 (^GSPC) close HIGHER on Wednesday July 29 (FOMC decision day) than its Tuesday July 28 close? Resolves per official ^GSPC closing values; equal or lower resolves NO.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 53.2% outcome bucket
§ lesson

NO. The S&P 500 closed 7316.15 on Wednesday July 29 against 7428.78 on Tuesday July 28 -- down 1.52%, the worst session since April 2025, with the Dow off about 1,100 points. An afternoon rebound failed and selling accelerated into the bell as the 10-year yield hit 4.66% and the 30-year topped 5.20% for the first time since 2007. The whole field sat between 42% and 62% on this one; it was correctly seen as a coin flip and it behaved like one until the last hour.

#07 · newsRESOLVED · NO

Will the FOMC RAISE the target range for the federal funds rate at its July 29, 2026 meeting? Resolves YES if the statement released at 2:00pm ET on July 29 sets a target range above the current 3.50%-3.75%. A hold at 3.50%-3.75% resolves NO. A cut resolves NO.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 34.1% outcome bucket
§ lesson

NO -- but the hawks showed up. The FOMC voted 9-3 to MAINTAIN the target range at 3-1/2 to 3-3/4 percent. Beth Hammack, Neel Kashkari and Lorie Logan all dissented in favor of a quarter-point hike, the most dissents on the committee in years. This is the question that named the week: a hawkish hold is still a hold, and the question asked only whether the range was RAISED. Reading the room correctly and answering a different question than the one on the page is the single most expensive habit in this tournament.

#08 · newsRESOLVED · NO

Will the ADVANCE estimate of Q2 2026 real GDP growth (released Thursday July 30, 8:30am ET) come in ABOVE 2.0% (seasonally adjusted annual rate)? Resolves per the BEA advance estimate headline figure as first released. Exactly 2.0% resolves NO. Later revisions do not change the resolution.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 41.9% outcome bucket
§ lesson

NO. The BEA advance estimate put Q2 2026 real GDP growth at +1.5% annualized, below the 2.0% threshold; Q1 was +2.1%. Consumer spending, investment and exports contributed while government spending fell. Of the public forecasts available at authoring, Atlanta Fed GDPNow at 1.7% was much closer than the Philadelphia Fed SPF at 2.1%. The nowcast beat the survey, which is the usual result and worth remembering.

#09 · wildcardRESOLVED · YES

Will 'Spider-Man: Brand New Day' gross MORE THAN $250 million domestically in its opening weekend (Friday July 31 - Sunday August 2)? Resolves per the Sunday estimates reported Sunday August 2 (Comscore / studio). Monday actuals are NOT used -- they publish after this week resolves. Exactly $250.0M resolves NO. This is a question about gross, not rank.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 53.9% outcome bucket
§ lesson

YES, emphatically. 'Spider-Man: Brand New Day' opened to an estimated 355 million dollars domestic on the Sunday August 2 estimates -- the second-biggest domestic opening ever, about 2 million short of Avengers: Endgame. It did 169.3 million Friday including previews, 101.5 million Saturday, with Sunday projected at 84 million. The 250 million bar was cleared by more than 40%. Tracking had climbed from 180-190 million to 260-300 million during authoring week, and even that was too low. Nobody in the field was above 80%.

#10 · wildcardRESOLVED · NO

Will a named tropical storm or hurricane form in the Atlantic basin between Wednesday July 29 and Sunday August 2, 2026 (inclusive)? Resolves YES if the National Hurricane Center assigns a name to a tropical or subtropical storm in the Atlantic basin (including the Caribbean Sea and the Gulf) during that window. Tropical depressions that are never named do not count.

1
11
21
31
41
51
61
71
81
91
N = 11mean = 20.5% outcome bucket
§ lesson

NO. The National Hurricane Center named no Atlantic system between July 29 and August 2. The 2026 Atlantic season has produced exactly two named storms all year -- Arthur, named June 17, and Bertha, named July 21 and dissipated July 24 -- and the basin was quiet through the entire window. Per NHC ATCF best-track data: only bal012026 and bal022026 exist, there is no bal03. The NHC's July 27 outlook had already said formation was not expected in the next 7 days, and the field mostly respected that; Obi at 1% and tstroup at 1% were the sharpest reads on the board this week.

est. 2026 · ten weeks · ten questions

“The number you submit is yours. The math is just keeping score.”

Calibrate or perish.
✦ ◊ §
© 2026 The Practice · Brier-graded since day onev0.1.0
✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ §✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ § ✦ ◊ §