1. The number of WNBA games on the slate tonight. Las Vegas at Golden State, 3:30 PM ET tip from Chase Center. The bot's published picks file shows zero HIGH-confidence calls (n_picks=0, n_games_on_board=1). The card has the spread at PASS (edge 1.45 points, well under the 5-point threshold) and the total at STRONG LEAN UNDER 167.5 (model 158.9, edge minus 8.6).
That STRONG UNDER is the only meaningful signal in the morning data. By the discipline rule we re-codified last week after the Thursday losses, STRONG UNDER calls require structural confirmation — both teams bottom-half in pace, both teams thinned by injury — before they get sized to half stake. Tonight does not have that confirmation.
Las Vegas is healthy outside of Dana Evans (a depth piece, Out). Golden State is missing Iliana Rupert (Out, not injury-related) but is otherwise running normal rotations. Neither team is bottom-half in pace. The UNDER edge is real on the model math (the projected quarter totals from the agent add to 161 versus a 167.5 line, which is a clean 6.5-point projected gap), but the framework filter says do not size without structural confirmation.
PASS the spread. PASS the total. Zero exposure tonight.
Yesterday in honest accounting.
Three games on Saturday's slate. We passed all three per the post-Thursday discipline. The bot's STRONG LEAN cards went 2-2 on the night.
Portland covered Indiana by a lot. Final 100-84 against the model's STRONG LEAN POR plus 10.2 line. Spread win for the bot. The IND-POR total went 184, well over the 173.5 line that the model had as a STRONG UNDER (edge minus 9.3). Total loss for the bot.
Seattle at Toronto was the framework-conflicting card. Model had SEA minus 5.7 as a STRONG LEAN with edge plus 9.0, market had a much shorter line on the road favorite. Toronto won at home 93-72. Spread loss for the bot. Total at 165 was clean under the 170.3 line. Total win for the bot.
Las Vegas at Connecticut was a PASS-both card. Connecticut won 84-81 at home. No bet either way.
Bot's editorial ledger from the night: 1-1 on spreads, 1-1 on totals. Our actual sized exposure was zero units. We sat the slate.
That is the framework working as intended after a hard losing week. The Thursday minus-2-unit slate triggered the discipline to sit through Saturday's noisier read, and the bot's split confirms no clear edge was being left on the table. Two wins, two losses, net flat for the bot, net flat for us.
The architecture's standing.
After Dallas was removed from the conviction architecture calculation (per Friday's update), the non-Dallas applications stand at 5-for-5 since the season opened. PHO, ATL, MIN, TOR, SEA all covered as the architecture predicted. The two Dallas losses (WAS on May 18, LVA on May 28) are now out of the headline number.
Tonight does not present an architecture firing. Las Vegas is favored at Golden State by the market and by the model. That is a road favorite spot — the first architectural condition. But there is no late-quarter edge from the agent (zero quarter edges flagged, no HIGH calls). And Golden State does not have the structural Q4 weakness that the architecture rule requires from the home team. Two of the three conviction conditions are absent. The pattern does not fire.
The combined STRONG ledger through May 30 reads 21-20 (51 percent) on editorial-sized bets — functionally a coin flip on volume. The per-tier breakdown shows the conviction architecture carrying the line (5-0 on non-Dallas) while the half-stake and totals tiers are doing the dragging.
The plan for the week.
Tonight: zero exposure. PASS both markets on the lone game. Tomorrow: two games on the board already (Minnesota at Phoenix, Seattle at Dallas). The SEA at DAL card will be the first post-Dallas-exclusion test of whether Seattle hits the architecture firing without the Dallas Q4 read in the math. Worth previewing tomorrow morning when the picks file lands.
The discipline holds tonight. PASS the slate. Talk tomorrow.
[ End Report ]
Share This