The full ledger from last night's 11-game NBA slate: the winner picks went 7-for-11, and the misses are worth naming plainly. The model called Kings at Clippers for Sacramento, and the Clippers won by 25. It called the Suns at Trail Blazers coin flip for Phoenix, and Portland won by 41. It had Timberwolves at Heat for Miami, and Minnesota won by 25, and it had the Bucks at Wizards for Milwaukee, and Washington won by 12. Four real misses, no way around them.
And the other side of the ledger: the line was three calls, and two of them landed exactly. The Lakers beat Golden State 127-105, winning going away just like the line said, with Los Angeles projected at 121.9 and landing at 127. The Raptors and Grizzlies were both supposed to cover their 17-point sims, and they did it with room to spare: Toronto beat Chicago 135-92, Memphis beat Utah 123-103, right near the projected 124.3. The Hornets at Nets was the model's quietest masterpiece, Charlotte 106.7 projected and 110 actual, Brooklyn 102.8 projected and 103 actual. The Pacers at Pelicans, Hawks at Magic, and Mavericks at Rockets all came through for the projected winners too.
So: 7-for-11 on winners against a line that said 8-for-11, exact or near-exact team scores in five games, and the two biggest blowout calls of the night nailed on the margin. That is a winning night with four honest misses, and that is the whole point of running hundreds of sims instead of picking from the gut.
The headliner: 1,000+ sims and the Bears lean at home over the Patriots
8:15 PM ET in Chicago, Thursday Night Football, and the biggest sim sample on the board belongs to Patriots at Bears: Chicago 26.8, New England 24.5, across 1,080 simulations. A 2.2-point lean. That is not the model planting a flag, that is the model saying these teams are close and the home field tips it.
My take: this game is won in the turnover column, and the sims know it. The projected stats have Caleb Williams at 231.3 passing yards with 1.5 touchdowns and just 0.4 interceptions, plus another 47.7 yards on the ground, while Drake Maye sits at 237.6 yards, 1.65 touchdowns, and 1.35 picks. One quarterback protecting the ball, one giving it away, in a game the sims call by two points. That split is the story of the night. The other quiet edge: D'Andre Swift projected at 108.0 rushing yards and 1.05 touchdowns on 23 carries. If the Bears feed him, the sims think this stays a Chicago game.
The Patriots are live, obviously. 1,080 sims producing a two-point margin means New England wins this thing roughly four times in ten. Maye slinging it for 237 yards with two scores is not a projection anyone should dismiss. But the sims were run on the full week of data, and they kept landing on Chicago. Lean Bears, and lean under total plays given how tight every run looks.
The undercard: a Thursday night Sun Belt triple-header
The college slate stacks three FBS games back to back, and all three are worth a look. At 7:00 PM, James Madison at Appalachian State: the sims have App State 33.8, JMU 28.5, a 5.4-point home call across 660 runs. Boone has been one of the toughest road trips in the Sun Belt for a decade, and the model is buying the home-field math again. Points expected here, on both sides.
At the same hour, Liberty at Kennesaw State: Liberty 30.4, Kennesaw State 26.9, a 3.5-point road call. The sims like the Flames' offense on the road, but a field-goal margin in a Thursday night road spot is a lean, not a lock. This is the one where the projected stats decide it for you before kickoff.
And the coin flip of the night tips at 7:30: East Carolina at Memphis, Memphis 29.8, ECU 29.3, a half-point margin across 660 sims. That is as close to a true coin flip as the model produces. My take: in a game this tight, the edge lives in the totals and the projected player stats, not in forcing a winner pick. Run the sims yourself, check the total, and let the numbers talk you into or out of it.
Perry line: Bears edge the Patriots in Chicago behind Swift and a cleaner Caleb Williams, App State and Liberty both hold serve at home, and the ECU-Memphis coin flip lands within a point of the sim projection.
All scores above are MyGameSim projections from hundreds of simulations, and none of these games have been played yet. Last night the model went 7-for-11 with the two big blowout calls nailed. Run your own sims, check the totals and projected stats, and tell me where you land tonight.
