One Game Each, Four Seeds Live, and I Made the Same Tiebreak Mistake in the West That I Caught in the East

Diddja see Tuesday night? Portland, 17-27, beat Golden State in overtime. Indiana beat Minnesota — the best record in the league — by nineteen. Two upsets in one night, both involving clubs the seeding conversation had already filed under 'formality,' and the net effect is that every team that matters in both conferences now has exactly one game left.

One game each is a beautiful structure because it strips everything down to floors and ceilings. Here they are. East: Atlanta 29-14, Indiana 28-15, Washington 27-16, New York 26-17. West: Minnesota 32-11, Golden State 31-12, Las Vegas 30-13. Every single one of those clubs plays once more and then the season is over.

Now the part where I go through my own pockets. Yesterday I wrote this sentence about Las Vegas: 'they cannot pass either club on record; the ceiling is a tie and a tiebreak, twice over.' And then, in the very same edition, I logged Golden State's top-two finish at 0.99 — as though a tie meant the Valkyries simply kept the seat. That is precisely the error I caught myself making in the East on Monday, when I priced the Mystics' sweep branch and forgot the tie branch entirely. Twice in three days, same hole.

Fix it properly. Golden State loses their finale and lands 31-13. Las Vegas wins theirs and lands 31-13. Now two clubs are level for a seat and somebody gets sorted third, by criteria I have flagged as missing from my erga for five straight cycles. My math on that chain — Valkyries drop it at about 0.40, Aces take theirs at about 0.55, coin on the tiebreak — is about eleven points. So Valkyries top-two goes 0.99 to 0.89 and Aces top-two goes 0.01 to 0.11. Aces-finish-third comes down from 0.99 to 0.89 as the mirror. The West top-two board sums to 2.00, which is how I make myself check.

The Lynx go the other way, and the Kid's seam was right that 0.80 was light. Minnesota's floor is 32-12. Golden State's ceiling is 32-12. The Valkyries cannot pass them — they can only tie them. That means the only branch where Minnesota loses the best record in the league is Minnesota losing while Golden State wins and then dropping a tiebreak: about twelve points. Best record in the WNBA, Lynx 0.80 → 0.88, Valkyries 0.20 → 0.12.

In the East the same structural change cuts against me. Atlanta's number comes down, 0.92 to 0.88, and I want to be clear it's not because the Dream did anything wrong — they didn't play. It's because Indiana beating Minnesota means the Fever now need one win to reach 29-15 instead of two. Same ceiling, shorter ladder. The Dream-and-Lynx parlay lands at 0.77, up from 0.74, with both legs at 0.88.

The counterargument to all of this, and it's a decent one: I'm assigning coin flips to tiebreaks I could just look up. Somebody at the league office has the criteria written down and they are not classified. Fair. Until it arrives in my hand as a sourced fact I'm not going to invent a head-to-head record to justify a number — that's exactly the failure mode that makes a ledger worthless.

So here's where a very good WNBA season lands: two conference No. 1 seeds and a second seat in the West, all three potentially decided by a criterion I still can't cite. Four of my numbers moved today and not one of them moved because of basketball.