What is UP, Duke Sizemore here, and today we are putting the MODEL under the microscope! Here’s what we’ve got: Detroit gets the early edge over Boston, the Celtics are carrying a serious Elo bruise into preseason, and Sacramento’s Ben Simmons move is the lone fresh transaction on the board.
Okay okay okay, Detroit-Boston! For that October 20 matchup, Detroit owns a 62-point Elo differential once COURT VISION folds in its 50-point home-court constant. That starts the Pistons at a 58.76 percent home-win probability. Jayson Tatum’s day-to-day knee soreness adds only about six-hundredths of a point in the weighted injury adjustment, there’s no rest adjustment, and the final number lands at 58.78 percent Detroit. That is a REAL specific forecast!
But Boston’s trajectory is the part that has me fired up. The Celtics peaked at a 1706 Elo in the displayed sequence, then Philadelphia hit ’em three straight: 113-97, 106-93, 109-100. Boston’s recent swing: MINUS 43.1 Elo points. And remember, we are talking preseason positioning here, not pretending those are new games. The modelled games run through June 14, and the next confirmed game on the calendar is Miami-Toronto on October 3.
Boston does have another roster note: Jordan Walsh has a three-year, $15 million extension running through 2029-30. Tatum remains day-to-day with that knee soreness.
Now, Sacramento: Ben Simmons has a one-year contract with the Kings. That’s today’s fresh league-news item, but since September 5, nobody has crossed COURT VISION’s qualifying power-mover threshold. No market fireworks to manufacture!
And finally, grade the grader! Across 1,246 games, COURT VISION sits at 68.3 percent accuracy with a .206 Brier score. Its average predicted home-win rate is 56.44 percent versus an actual 55.06 — a 1.38-percentage-point home bias. THAT is the kind of number we track as the season approaches.
Keep your eyes on the data and your ears right here. Duke Sizemore, out!