McBride Went Off and the Model Saw It Coming

Two games. Eighteen players graded. And the kind of injury-driven role expansion that turns a Sunday slate into a genuine puzzle — if you're paying attention to the right signals.

The big story out of Minnesota was simple and loud: Kayla McBride dropped 37 points, grabbed 6 rebounds, added 4 steals and 2 blocks in 35 minutes, and finished with 60 DraftKings points. That's not a good game. That's one of those lines you read twice. The box score on the PHX side told its own story — Alyssa Thomas stuffed the sheet in her own quiet way, racking up 12 assists, 8 rebounds, and 19 points across 37 minutes. Olivia Miles was efficient and dangerous at 33 points and 8 assists. DeWanna Bonner and Kahleah Copper both cleared value. The thread connecting nearly every notable performance? Injuries had reshuffled the minutes picture before tip-off, and the players who absorbed that extra run delivered.

The honest one-number receipt first: across 18 graded players, our mean absolute error came in at 9 DraftKings points. On a slate where the top scorer finished at 60, that's a reasonable spread — not a clean sheet, but a defensible one. Here's how the names broke down.

Top Scorers — Projection vs. Reality

• Kayla McBride (MIN, $9,400) — We had her at 30.9 DK points. She finished at 60. Line: 35 min, 37 pts, 6 reb, 1 ast, 4 stl, 2 blk. An injury had bumped her minutes, and while the model picked that up and adjusted her projection upward, McBride then played a tier above even the elevated ceiling. That 29-point beat is the kind of outlier that reminds you projections are probability distributions, not promises.
• Olivia Miles (MIN, $10,200) — Projected at 40.3, finished at 48.8. Line: 31 min, 33 pts, 3 reb, 8 ast. An injury had bumped her minutes too. Clean beat, right direction.
• Alyssa Thomas (PHX, $10,800) — Projected at 38.3, finished at 48.5. Line: 37 min, 19 pts, 8 reb, 12 ast, 1 stl. Minutes were bumped by injury. Thomas found her production through the assist column the way she usually does — quietly and relentlessly.
• DeWanna Bonner (PHX, $7,400) — Projected at 22.7, finished at 35.3. Line: 33 min, 17 pts, 9 reb, 3 ast, 1 stl. Injury bumped her minutes. At $7,400 salary, that's the kind of value outcome that matters most in lineup construction.
• Kahleah Copper (PHX, $8,500) — Projected at 30.7, finished at 33.8. Line: 32 min, 26 pts, 3 reb, 1 ast, 1 stl. Minutes bumped by injury. Solid, unspectacular beat — the model and reality shook hands.
• Natasha Howard (MIN, $10,000) — Projected at 38.4, finished at 29.3. Line: 35 min, 13 pts, 3 reb, 7 ast, 2 stl. Minutes were there — the scoring just wasn't.

Best Calls

Three stood out on the right side of the ledger. Alyssa Thomas at $10,800 finishing 10 points above projection is a strong hit at a salary that requires it. DeWanna Bonner beating her number by 12.6 at $7,400 is the cleaner value story — that's 4.77 DK points per $1K at a mid-tier price, and the model's injury-adjusted bump got you there. Kayla McBride was directionally correct and the best call on the slate in terms of identifying the right player for the right game — 60 actual DK points just lives outside the model's lane. You can't fully project an all-timer night. You can identify the setup. We did.

Worst Call

Own it: Nia Coffey (MIN, $7,800) was a miss. We had her at 25.4 DK points. She finished at 15.3. Line: 30 min, 5 pts, 3 reb, 0 ast, 1 stl, 2 blk. An injury had bumped her minutes too — the model read the opportunity correctly, but Coffey didn't convert it offensively. The minutes were real. The production wasn't. That's a 10-point miss and it belongs in the record.

The clearest takeaway from July 13 is also the simplest: the injury-adjusted minutes signal is real and it's worth following. Four of the six notable performers — McBride, Miles, Thomas, Bonner — all absorbed extra run because of lineup changes before tip-off, and all of them beat their projections. The model's walk-forward validation is built on exactly this kind of signal: identifying role expansions before they're priced in and letting the math work out over a large sample. One night isn't proof of anything. But a consistent pattern of upside beats among min-bumped players is something to watch going forward.

The Coffey miss is the honest counterweight. Minutes opportunity and scoring production are not the same thing, and the model doesn't always separate them cleanly when a player's role is newly expanded. That's a known variance source, and it'll show up again. The right response isn't to ignore the signal — it's to weight it alongside scoring role and usage context, not just raw minutes.

And McBride scoring 60 DraftKings points while we had her at 30.9 is not a failure of the model. It's a reminder that the best projections in the world still can't fully price in transcendent individual performances. What the model can do — and did — is put you in the right seat for the right game. What happens next is basketball.