
MIL
18-16

STL
21-14
Line movement
200 snapshots
STL spread
+1.5
open +1.5
Total
O/U 8.0
open O/U 8.0
STL no-vig %
50.4%
open 51.7% · -1.3
Stepped lines reflect captured market snapshots from odds_snapshots. Spread sign convention: negative = STL favored. Pure CLV joins once moneyline closing-line ingest catches up across leagues.
Postgame · final
Line score, top performers, model verdict against Vegas, and how the closing line shaped up vs the actual outcome.
Final
MIL wins
MIL 0 · STL 0 (tied)
Model verdict
✗ Missed
Picked STL +0pp
Against the spread
No spread
Line score
What's next
See every model edge for tonight's remaining games and tomorrow's slate side-by-side, or jump straight to DraftKings & FanDuel for the full board.
21+ · we may earn a referral fee · your odds unchanged.
Tonight's box score
MIL · STL

MIL
18-16
| Player | H-AB | R | RBI | HR |
|---|
| Player | IP | H | ER | K |
|---|

STL
21-14
| Player | H-AB | R | RBI | HR |
|---|
| Player | IP | H | ER | K |
|---|
Current series
STL leads series 1-0
MIL3@
STL6
MIL0@
STL0today
MIL@
STLupcomingSeason series
STL leads series 1-0
MIL3@
STL6
MIL0@
STL0today
MIL@
STLupcoming
STL@
MILupcoming
STL@
MILupcoming
STL@
MILupcoming
MIL@
STLupcoming
MIL@
STLupcoming
MIL@
STLupcoming
MIL@
STLupcoming
STL@
MILupcoming
STL@
MILupcoming
STL@
MILupcomingVegas line center
DraftKings via ESPN · 21+
Spread
MIL -1.5
Total
7.5
Defensive · -1.0 vs avg
Moneyline
MIL-11052%
STL-11052%Implied probabilities back-computed from American odds — break-even win % a moneyline bet needs to be +EV.
Line movement · 200 snapshots
ESPN-tracked · 21+
Spread
1.5
0.0 since open
Total
7.5
0.0 since open
Betting line
MIL -1.5·O/U 7.5·MIL -110/STL -110
Player projections
Per-player stat projections built from a recency-weighted blend of the last ten games, season average, and matchup context. Confidence reflects sample size and stability — the top of each list is who to watch.
117
projections · 57 high confidence
Strikeouts
Hits
Total bases
RBIs
Earned runs
Projections recompute every 30 minutes · prop lines plug in once sportsbook ingest lands
Model ensemble · how the prediction is built
Each sub-model uses a different rating substrate. Bayesian model averaging weights them by rolling Brier score so the ensemble inherits each model's strengths. Disagreement flags games where the sub-models don't see eye-to-eye — lower confidence, wider band.
51.9%
ensemble · STL favored
Elo Static
50.4%
P(STL win)
33%
weight
Elo Pitching
54.0%
P(STL win)
33%
weight
Bullpen Park
51.2%
P(STL win)
33%
weight
Disagreement
1.54 pp
weighted σ across sub-models
Confidence
90% · high
maps from disagreement
Substrate count
3 / 3 active
ones with full inputs tonight
Weights are recalibrated nightly on a 90-day rolling window with strict point-in-time correctness — no model gets credit for a game it hasn't seen