MLB Prediction System

KPI Player Effects

How each KPI adds or removes a hit, a strikeout, a walk or a home run, per player, against the arm he actually faces

Slate 20261011Games 1Schema kpi_player_effects_v2Built 2026-10-12 09:15Z

What this document measures

Every counting stat below has a KPI on both sides of the matchup, measured on the same denominator: a plate appearance. That is what lets the hitter's reading and the starter's reading be combined instead of listed side by side. The combination is the standard Log5 odds ratio, which is exactly additive in log odds, so the hitter's push and the pitcher's push separate cleanly and the remainder is reported rather than absorbed.

Hits use a derived KPI. Hit rate is not in config/kpi_registry.json: registering it would move the registry hash recorded in the model manifest and fail the live layer closed until a retrain. It is computed with the same strict as-of cutoff and the same shrinkage, stored in its own namespace, and never read by the runtime.

These per-player numbers are measured, not published. The live KPI layer moves team runs and the win probability. The hitter lines in the artifact were then scaled by one club-wide ratio, identical for the leadoff hitter and the ninth bat. The per-hitter columns here are what that single ratio discards.

The whole staff, not only the starter. A hitter takes roughly four in ten of his plate appearances against relievers. Each club's bullpen here is the pregame bullpen: the seven busiest healthy relievers of the last 30 days, the same arms the KPI forecast weighted and the simulation pitched, each sharing the relief work by the batters faced the forecast gave him. The bullpen's workload tonight is the club's real mean staff batters faced minus the starter's projection. The published staff alone holds only the relievers one simulation draw used, so it is shown beside the bullpen for comparison, not used.

Reading a delta. Own skill is measured against the league. The opponent effect is measured against the hitter's own rate, which is the only one that may adjust a published line, because the published line already carries his own skill. Adding both to a published line would count the hitter twice.

Coverage. 18 hitter-versus-starter pairings; 17 have both sides above their sample floor on hits and 18 carry a non-zero hit adjustment. A side below its floor falls back to the league rate and is marked n<floor, which is a stated absence of evidence rather than a measured zero. All four outcome metrics have value documents. Handedness source: model bundle pitcher_hands.json.

Largest per-player movements on the slate

Hitter	Club	Metric	Starter faced (+ bullpen)	Published	KPI adjusted, whole staff	Δ	Hitter band	Pitcher band	Game
Teoscar Hernández	LAD	Strikeouts	Jacob Misiorowski	1.526	2.046	+0.520	Below average	Very good	20261011_LAD@MIL
Shohei Ohtani	LAD	Strikeouts	Jacob Misiorowski	1.402	1.904	+0.502	Average	Very good	20261011_LAD@MIL
Enrique Hernández	LAD	Strikeouts	Jacob Misiorowski	1.278	1.758	+0.479	averagen<floor	Very good	20261011_LAD@MIL
Max Muncy	LAD	Strikeouts	Jacob Misiorowski	1.318	1.785	+0.467	Below average	Very good	20261011_LAD@MIL
Freddie Freeman	LAD	Strikeouts	Jacob Misiorowski	1.122	1.580	+0.458	Good	Very good	20261011_LAD@MIL
Will Smith	LAD	Strikeouts	Jacob Misiorowski	1.063	1.502	+0.439	Good	Very good	20261011_LAD@MIL
Andy Pages	LAD	Strikeouts	Jacob Misiorowski	1.028	1.432	+0.404	Average	Very good	20261011_LAD@MIL
Garrett Mitchell	MIL	Strikeouts	Tarik Skubal	1.631	2.000	+0.368	Poor	Very good	20261011_LAD@MIL
Brice Turang	MIL	Strikeouts	Tarik Skubal	1.472	1.840	+0.368	Below average	Very good	20261011_LAD@MIL
Kyle Tucker	LAD	Strikeouts	Jacob Misiorowski	0.885	1.252	+0.366	Good	Very good	20261011_LAD@MIL
Mookie Betts	LAD	Strikeouts	Jacob Misiorowski	0.770	1.124	+0.354	Very good	Very good	20261011_LAD@MIL
Jake Bauers	MIL	Strikeouts	Tarik Skubal	1.454	1.806	+0.351	Below average	Very good	20261011_LAD@MIL
Jackson Chourio	MIL	Strikeouts	Tarik Skubal	1.338	1.679	+0.341	Average	Very good	20261011_LAD@MIL
Joey Ortiz	MIL	Strikeouts	Tarik Skubal	1.115	1.415	+0.301	Average	Very good	20261011_LAD@MIL
Luis Lara	MIL	Strikeouts	Tarik Skubal	1.070	1.349	+0.279	Average	Very good	20261011_LAD@MIL
Andrew Vaughn	MIL	Strikeouts	Tarik Skubal	0.880	1.130	+0.250	Good	Very good	20261011_LAD@MIL
Cooper Pratt	MIL	Strikeouts	Tarik Skubal	0.879	1.122	+0.243	Good	Very good	20261011_LAD@MIL
William Contreras	MIL	Strikeouts	Tarik Skubal	0.762	0.984	+0.222	Good	Very good	20261011_LAD@MIL
Freddie Freeman	LAD	Hits	Jacob Misiorowski	0.817	0.699	-0.118	Very good	Very good	20261011_LAD@MIL
Mookie Betts	LAD	Hits	Jacob Misiorowski	0.789	0.674	-0.115	Good	Very good	20261011_LAD@MIL

Per game

Club Player

20261011_LAD@MIL — team layer withheld (values_fresh); runs moved away -0.6301 / home 1.0124

Tarik Skubal (LHP, 22.4 BF)
H 0.2076 Average · SO 0.2910 Very good · BB 0.0527 Very good · HR 0.0259 Good

Jacob Misiorowski (RHP, 18.9 BF)
H 0.1679 Very good · SO 0.3593 Very good · BB 0.0727 Good · HR 0.0235 Very good

Warnings: away published staff projects 26.3214 batters faced, outside the real p10-p90 band of 30-45; the exposure here uses the club's real 35.71 per game instead

LAD hitters vs Jacob Misiorowski and the bullpen

The pipeline applied no club-wide rescale to LAD hitter lines. The columns below are the measured alternative: each hitter moved by his own KPI pairing with the whole opposing staff, the starter and the full relief corps combined at the share of his plate appearances each takes. Click a name for the starter-only and bullpen decomposition and the contextual readings.

Player	Pos	Faces	PA	Hits		Strikeouts		Walks		Home runs
Player	Pos	Faces	PA	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ
Mookie Betts	SS	Jacob Misiorowski + bullpen	4.23	0.789 → 0.674	-0.115	0.770 → 1.124	+0.354	0.349 → 0.323	-0.026	0.149 → 0.122	-0.027

Freddie Freeman	1B	Jacob Misiorowski + bullpen	4.37	0.817 → 0.699	-0.118	1.122 → 1.580	+0.458	0.435 → 0.403	-0.032	0.110 → 0.090	-0.020

Will Smith	C	Jacob Misiorowski + bullpen	4.14	0.748 → 0.638	-0.110	1.063 → 1.502	+0.439	0.400 → 0.371	-0.029	0.128 → 0.104	-0.024

Shohei Ohtani	DH	Jacob Misiorowski + bullpen	4.28	0.779 → 0.665	-0.114	1.402 → 1.904	+0.502	0.523 → 0.485	-0.038	0.186 → 0.152	-0.034

Teoscar Hernández	LF	Jacob Misiorowski + bullpen	4.01	0.775 → 0.662	-0.113	1.526 → 2.046	+0.520	0.315 → 0.291	-0.023	0.126 → 0.103	-0.023

Max Muncy	3B	Jacob Misiorowski + bullpen	4.03	0.721 → 0.615	-0.106	1.318 → 1.785	+0.467	0.432 → 0.401	-0.032	0.183 → 0.150	-0.033

Andy Pages	CF	Jacob Misiorowski + bullpen	3.73	0.715 → 0.611	-0.104	1.028 → 1.432	+0.404	0.304 → 0.281	-0.023	0.125 → 0.102	-0.023

Kyle Tucker	RF	Jacob Misiorowski + bullpen	3.88	0.671 → 0.571	-0.100	0.885 → 1.252	+0.366	0.430 → 0.399	-0.031	0.106 → 0.086	-0.020

Enrique Hernández	2B	Jacob Misiorowski + bullpen	3.35	0.621 → 0.529	-0.092	1.278 → 1.758	+0.479	0.239 → 0.221	-0.018	0.116 → 0.095	-0.021

LAD starter against the MIL lineup

Player	Pos	Faces	BF	Hits		Strikeouts		Walks		Home runs
Player	Pos	Faces	BF	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ
Tarik Skubal	Starter	MIL	22.38	3.924 → 4.130	+0.206	6.300 → 6.300	+0.001	1.186 → 1.388	+0.202	0.537 → 0.504	-0.033

LAD bullpen: the full relief corps (7 arms)

Exposure. Starter 22.4 batters faced, bullpen 13.3 (37.3% of the game), from the club's real 35.7 staff batters faced per game over 20 complete games in the last 30 days. The published staff alone would have given the bullpen 15.0% across 1 arms, because it holds only the relievers the selected simulation draw used.

Reliever	Role	In the corps because	Relief BF, 30d	Apps	Share	BF tonight
Edgardo Henriquez	Reliever	pregame bullpen	22	4	19.4%	2.59
Tanner Scott	Reliever	pregame bullpen	34	9	19.0%	2.53
Alex Vesia	Middle Reliever (High Leverage)	pregame bullpen, published staff	28	9	15.5%	2.07
Edwin Díaz	Reliever	pregame bullpen	27	8	15.3%	2.05
Evan Phillips	Reliever	pregame bullpen	23	8	10.6%	1.41
Roki Sasaki	Reliever	pregame bullpen	19	4	10.1%	1.35
Brock Stewart	Reliever	pregame bullpen	21	7	10.0%	1.34

LAD relievers against the MIL lineup

Player	Pos	Faces	BF	Hits		Strikeouts		Walks		Home runs
Player	Pos	Faces	BF	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ
Edgardo Henriquez	Reliever	MIL (19% of relief)	2.59	0.487 → 0.513	+0.026	0.714 → 0.714	0	0.253 → 0.294	+0.041	0.055 → 0.051	-0.003

Tanner Scott	Reliever	MIL (19% of relief)	2.53	0.519 → 0.546	+0.027	0.762 → 0.762	0	0.180 → 0.210	+0.030	0.064 → 0.060	-0.004

Alex Vesia	Middle Reliever (High Leverage)	MIL (16% of relief)	2.07	0.458 → 0.483	+0.025	1.244 → 1.244	0	0.482 → 0.558	+0.076	0.053 → 0.050	-0.003

Edwin Díaz	Reliever	MIL (15% of relief)	2.05	0.444 → 0.467	+0.023	0.453 → 0.453	0	0.182 → 0.212	+0.030	0.062 → 0.058	-0.004

Evan Phillips	Reliever	MIL (11% of relief)	1.41	0.305 → 0.321	+0.016	0.311 → 0.311	0	0.125 → 0.146	+0.020	0.043 → 0.040	-0.003

Roki Sasaki	Reliever	MIL (10% of relief)	1.35	0.292 → 0.307	+0.015	0.315 → 0.315	0	0.123 → 0.143	+0.020	0.050 → 0.046	-0.003

Brock Stewart	Reliever	MIL (10% of relief)	1.34	0.290 → 0.305	+0.015	0.296 → 0.296	0	0.119 → 0.138	+0.019	0.041 → 0.038	-0.003

MIL hitters vs Tarik Skubal and the bullpen

The pipeline applied no club-wide rescale to MIL hitter lines. The columns below are the measured alternative: each hitter moved by his own KPI pairing with the whole opposing staff, the starter and the full relief corps combined at the share of his plate appearances each takes. Click a name for the starter-only and bullpen decomposition and the contextual readings.

Player	Pos	Faces	PA	Hits		Strikeouts		Walks		Home runs
Player	Pos	Faces	PA	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ
Jackson Chourio	LF	Tarik Skubal + bullpen	4.44	1.070 → 1.017	-0.053	1.338 → 1.679	+0.341	0.357 → 0.270	-0.087	0.163 → 0.142	-0.021

Brice Turang	2B	Tarik Skubal + bullpen	4.26	0.955 → 0.907	-0.048	1.472 → 1.840	+0.368	0.424 → 0.323	-0.101	0.113 → 0.099	-0.015

William Contreras	C	Tarik Skubal + bullpen	4.23	0.964 → 0.915	-0.048	0.762 → 0.984	+0.222	0.381 → 0.289	-0.092	0.117 → 0.102	-0.015

Jake Bauers	1B	Tarik Skubal + bullpen	4.09	0.861 → 0.817	-0.044	1.454 → 1.806	+0.351	0.477 → 0.365	-0.111	0.142 → 0.124	-0.018

Andrew Vaughn	DH	Tarik Skubal + bullpen	3.70	0.845 → 0.803	-0.042	0.880 → 1.130	+0.250	0.338 → 0.256	-0.081	0.076 → 0.066	-0.010

Garrett Mitchell	CF	Tarik Skubal + bullpen	3.87	0.862 → 0.818	-0.044	1.631 → 2.000	+0.368	0.326 → 0.248	-0.078	0.081 → 0.070	-0.011

Cooper Pratt	SS	Tarik Skubal + bullpen	3.72	0.821 → 0.779	-0.041	0.879 → 1.122	+0.243	0.318 → 0.241	-0.077	0.073 → 0.064	-0.009

Luis Lara	RF	Tarik Skubal + bullpen	3.81	0.814 → 0.772	-0.042	1.070 → 1.349	+0.279	0.307 → 0.232	-0.075	0.069 → 0.060	-0.009

Joey Ortiz	3B	Tarik Skubal + bullpen	3.64	0.756 → 0.718	-0.038	1.115 → 1.415	+0.301	0.295 → 0.223	-0.072	0.083 → 0.072	-0.011

MIL starter against the LAD lineup

Player	Pos	Faces	BF	Hits		Strikeouts		Walks		Home runs
Player	Pos	Faces	BF	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ
Jacob Misiorowski	Starter	LAD	18.85	3.028 → 3.134	+0.106	6.481 → 5.994	-0.488	1.242 → 1.431	+0.189	0.165 → 0.191	+0.026

MIL bullpen: the full relief corps (7 arms)

Exposure. Starter 18.9 batters faced, bullpen 18.1 (48.9% of the game), from the club's real 36.9 staff batters faced per game over 18 complete games in the last 30 days. The published staff alone would have given the bullpen 46.6% across 3 arms, because it holds only the relievers the selected simulation draw used.

Reliever	Role	In the corps because	Relief BF, 30d	Apps	Share	BF tonight
JoJo Romero	Long Reliever	pregame bullpen, published staff	37	8	18.9%	3.42
Aaron Ashby	Middle Reliever (High Leverage)	pregame bullpen, published staff	53	9	18.1%	3.26
DL Hall	Reliever	pregame bullpen	52	9	17.1%	3.09
Antonio Senzatela	Middle Reliever	pregame bullpen, published staff	32	9	15.7%	2.83
Abner Uribe	Reliever	pregame bullpen	46	11	13.8%	2.49
Trevor Megill	Reliever	pregame bullpen	34	9	13.4%	2.41
Chad Patrick	Reliever	pregame bullpen	28	6	3.0%	0.55

MIL relievers against the LAD lineup

Player	Pos	Faces	BF	Hits		Strikeouts		Walks		Home runs
Player	Pos	Faces	BF	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ	published → KPI	Δ
JoJo Romero	Long Reliever	LAD (19% of relief)	3.42	1.008 → 1.041	+0.033	1.385 → 1.261	-0.124	0.369 → 0.425	+0.056	0.184 → 0.213	+0.029

Aaron Ashby	Middle Reliever (High Leverage)	LAD (18% of relief)	3.26	0.995 → 1.028	+0.033	1.853 → 1.699	-0.154	0.633 → 0.726	+0.093	0.094 → 0.109	+0.015

DL Hall	Reliever	LAD (17% of relief)	3.09	0.537 → 0.556	+0.019	0.785 → 0.717	-0.068	0.379 → 0.433	+0.054	0.084 → 0.098	+0.013

Antonio Senzatela	Middle Reliever	LAD (16% of relief)	2.83	1.033 → 1.066	+0.033	0.768 → 0.697	-0.071	0.412 → 0.474	+0.062	0.059 → 0.068	+0.009

Abner Uribe	Reliever	LAD (14% of relief)	2.49	0.502 → 0.519	+0.017	0.631 → 0.577	-0.055	0.205 → 0.236	+0.031	0.061 → 0.070	+0.010

Trevor Megill	Reliever	LAD (13% of relief)	2.41	0.467 → 0.483	+0.016	0.650 → 0.595	-0.055	0.175 → 0.202	+0.027	0.051 → 0.059	+0.008

Chad Patrick	Reliever	LAD (3% of relief)	0.55	0.109 → 0.113	+0.004	0.117 → 0.107	-0.011	0.057 → 0.066	+0.008	0.015 → 0.018	+0.002

Measured from published artifacts and stored KPI values. Nothing recomputed. Per-player adjustments are a measurement of the catalog, not the published line.