MLB Model vs Market: Week in Review (Sep 14 to 20, 2026)

How we did

Graded MLB games, Sep 14 to 20, 2026 · reports, not picks

Games graded

77

2026-W38

Moneyline record

49-28

63.6% higher-prob side

our >50% side won

Moneyline Brier

0.237

better than 0.250 coin flip

lower is better

Totals avg miss

2.82

runs |proj − actual|

77 games

Totals vs market

-0.07

leads market (2.82 vs 2.89)

77 lined games

Where the week's results came from

A fixed set of splits, published every week whether it flatters us or not · slices under n=5 show their count and no rate

Moneyline split	Record	n
Model's side was the home team	32–21	53
Model's side was the road team	15–7	22
Model's side was the market favorite	41–21	62
Model's side was the market underdog	8–7	15
Coin-flip calls (50-55%)	33–21	54
Moderate calls (55-65%)	15–7	22
Strongest calls (65%+)	not scored: below the n=5 floor (n=1)
Games with a total under 8.5	32–13	45
Games with a total 8.5-9.5	11–14	25
Games with a total above 9.5	6–1	7

Totals split	Our avg miss	Market's	n
Projection sat above the line	2.9	2.98	67
Projection sat below the line	1.37	1.39	9
Lines under 8.5	2.87	3.12	45
Lines 8.5-9.5	2.9	2.66	25
Lines above 9.5	2.18	2.21	7

1 projection landed exactly on the line, so the above/below splits do not sum to the lined-game count.

Confidence bucket	We said	It happened	n
Coin-flip calls (50-55%)	52.3%	61.1%	54
Moderate calls (55-65%)	59.5%	68.2%	22
Strongest calls (65%+)	not scored: below the n=5 floor (n=1)

League base rates for the same games: home teams went 41–36 (n=77); market favorites went 49–28 (63.6%, n=77); overs went 38–35 against the line (n=77). Every model result above sits inside that backdrop.

The three largest totals misses account for 12.6% of the week's total absolute totals error (27.4 of 216.8 runs across 77 games).

Strongest call that landed

SF @ LAD (Sep 18): we had the home side at 66%; the home side won.

Final SF 2, LAD 8; opening price on that side -325; total 8; the archived probables started.

Strongest call that missed

LAD @ CIN (Sep 16): we had the away side at 62%; the home side won. It stays in the record.

Final LAD 2, CIN 6; opening price on that side -218; total 8.5; the archived probables started.

One week is a small sample. These are counts of where the results came from, not conclusions about skill; the season-long record is the number that matters.

The biggest misses

We show them on purpose · totals, Sep 14 to 20, 2026

The point of a public track record is that the misses stay in. Our three largest totals misses this week:

Game	Projected	Actual	Off by
DET @ CWS	8.5	19	10.5
OAK @ CLE	9.1	18	8.9
MIL @ BAL	8.9	1	7.9

Every graded game

Projected total vs. actual, and whether our higher-probability moneyline side won · 77 games

Game	Proj total	Actual	Miss	Home win%	ML side
SF @ STL	9.3	3	6.3	0.53	Hit
SD @ COL	11.4	15	3.6	0.38	Hit
CWS @ CLE	8.2	10	1.8	0.50	Hit
NYY @ MIN	9.2	11	1.8	0.45	Hit
BAL @ NYM	9.6	3	6.6	0.59	Miss
ATL @ CHC	8.9	10	1.1	0.53	Hit
SEA @ LAA	8.2	10	1.8	0.53	Hit
DET @ TOR	9.1	11	1.9	0.54	Miss
MIA @ ARI	9.3	15	5.7	0.57	Hit
LAD @ CIN	9.3	5	4.3	0.38	Hit
KC @ HOU	8.3	6	2.3	0.54	Hit
SEA @ LAA	8.2	3	5.2	0.46	Miss
MIA @ ARI	9.8	6	3.8	0.58	Miss
BOS @ TEX	9.6	10	0.4	0.51	Hit
ATL @ CHC	8.9	12	3.0	0.57	Hit
SD @ COL	11.5	12	0.5	0.46	Hit
KC @ HOU	8.2	7	1.2	0.53	Miss
OAK @ TB	8.8	7	1.8	0.60	Hit
SF @ STL	9.8	11	1.2	0.55	Miss
MIL @ PIT	8.7	9	0.3	0.46	Hit
BAL @ NYM	8.8	8	0.8	0.53	Miss
MIA @ ARI	9.0	7	2.0	0.59	Miss
CWS @ CLE	8.7	9	0.3	0.53	Hit
LAD @ CIN	10.3	8	2.3	0.38	Miss
SEA @ LAA	8.3	9	0.7	0.51	Miss
NYY @ MIN	9.2	9	0.2	0.43	Miss
PHI @ WSH	9.2	3	6.2	0.38	Hit
DET @ TOR	9.3	6	3.3	0.54	Hit
BOS @ TEX	8.5	7	1.5	0.50	Hit
LAD @ CIN	10.1	10	0.1	0.38	Hit
SD @ COL	11.2	11	0.2	0.38	Hit
MIN @ LAA	8.4	9	0.6	0.50	Hit
MIL @ PIT	9.3	11	1.7	0.49	Miss
PHI @ NYM	8.8	3	5.8	0.54	Miss
OAK @ TB	8.8	11	2.2	0.60	Hit
KC @ HOU	9.7	8	1.7	0.50	Hit
DET @ CWS	8.8	4	4.8	0.50	Hit
MIN @ LAA	8.8	3	5.8	0.50	Miss
BOS @ TB	8.3	6	2.3	0.55	Miss
KC @ PIT	8.8	13	4.2	0.53	Hit
MIA @ SD	8.3	10	1.7	0.60	Hit
NYY @ ARI	9.4	11	1.6	0.50	Miss
SF @ LAD	8.7	10	1.3	0.66	Hit
DET @ CWS	8.5	19	10.5	0.50	Miss
ATL @ HOU	8.4	8	0.4	0.51	Miss
WSH @ STL	8.8	10	1.2	0.50	Hit
OAK @ CLE	8.4	8	0.4	0.53	Hit
NYY @ ARI	9.0	8	1.0	0.38	Miss
WSH @ STL	9.2	13	3.8	0.51	Miss
TOR @ TEX	9.1	8	1.1	0.53	Hit
SEA @ COL	10.5	8	2.5	0.47	Miss
ATL @ HOU	9.2	9	0.1	0.53	Miss
MIA @ SD	9.1	13	3.9	0.57	Hit
PHI @ NYM	8.5	13	4.5	0.53	Hit
OAK @ CLE	9.1	18	8.9	0.50	Hit
MIN @ LAA	8.2	11	2.8	0.54	Hit
DET @ CWS	8.6	4	4.6	0.52	Hit
CHC @ CIN	9.7	7	2.7	0.46	Hit
KC @ PIT	8.7	11	2.3	0.52	Hit
SF @ LAD	8.7	14	5.3	0.59	Hit
BOS @ TB	8.8	3	5.8	0.52	Hit
MIL @ BAL	8.9	1	7.9	0.52	Miss
NYY @ ARI	9.3	12	2.7	0.48	Miss
BOS @ TB	8.6	6	2.6	0.53	Hit
PHI @ NYM	8.3	9	0.7	0.53	Miss
ATL @ HOU	8.9	6	3.0	0.60	Miss
WSH @ STL	9.6	8	1.6	0.54	Hit
SEA @ COL	10.6	3	7.6	0.47	Hit
TOR @ TEX	8.7	9	0.3	0.51	Miss
DET @ CWS	8.6	9	0.4	0.51	Hit
MIA @ SD	9.4	10	0.6	0.55	Hit
SF @ LAD	8.8	4	4.8	0.60	Hit
MIL @ BAL	9.0	3	6.0	0.46	Hit
OAK @ CLE	8.3	1	7.3	0.58	Hit
KC @ PIT	8.9	7	1.9	0.53	Hit
MIN @ LAA	8.7	8	0.7	0.52	Miss
CHC @ CIN	9.2	10	0.8	0.45	Hit

What the market says

A market we deliberately don't model: the market's own record

MLB run line

42.8% of −1.5 favorites cover

Across 11,972 final regular-season games with a standard run line, the moneyline favorite laying −1.5 covered 42.8% of the time and won outright 58.3%. We publish no run-line projection, because the market prices this one efficiently, so this is the market's record, not ours.

Market benchmark, not a Nebula projection.

One number: Why we show the losses

A rotating look at how we grade ourselves

Every projection is logged before first pitch, immutably, and graded after the game. Nothing is added or removed later. The misses stay in. A record you can only read forward is the only kind worth trusting.

Keep reading

The full record and the live board

Full track record All insights MLB board How we grade

How we did

Where the week's results came from

The biggest misses

Every graded game

What the market says

One number: Why we show the losses

Keep reading

Get the weekly projections recap