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 20261010Games 1Schema kpi_player_effects_v2Built 2026-10-11 09:12Z

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; 16 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
Kyle Teel	CHW	Strikeouts	Gavin Williams	1.813	2.316	+0.503	averagen<floor	Very good	20261010_CHW@CLE
Munetaka Murakami	CHW	Strikeouts	Gavin Williams	1.885	2.300	+0.415	Poor	Very good	20261010_CHW@CLE
Colson Montgomery	CHW	Strikeouts	Gavin Williams	1.709	2.083	+0.375	Poor	Very good	20261010_CHW@CLE
Brenton Doyle	CHW	Strikeouts	Gavin Williams	1.466	1.815	+0.350	Poor	Very good	20261010_CHW@CLE
Chase Meidroth	CHW	Strikeouts	Gavin Williams	1.106	1.415	+0.309	Average	Very good	20261010_CHW@CLE
Miguel Vargas	CHW	Strikeouts	Gavin Williams	0.965	1.252	+0.286	Good	Very good	20261010_CHW@CLE
Tristan Peters	CHW	Strikeouts	Gavin Williams	0.962	1.248	+0.285	Good	Very good	20261010_CHW@CLE
Andrew Benintendi	CHW	Strikeouts	Gavin Williams	1.034	1.314	+0.280	Average	Very good	20261010_CHW@CLE
Sam Antonacci	CHW	Strikeouts	Gavin Williams	0.834	1.095	+0.261	Good	Very good	20261010_CHW@CLE
Jo Adell	CLE	Strikeouts	Sean Burke	1.209	1.340	+0.131	Average	Good	20261010_CHW@CLE
Patrick Bailey	CLE	Strikeouts	Sean Burke	1.143	1.261	+0.118	Below average	Good	20261010_CHW@CLE
Nathaniel Lowe	CLE	Strikeouts	Sean Burke	0.999	1.108	+0.109	Average	Good	20261010_CHW@CLE
Travis Bazzana	CLE	Strikeouts	Sean Burke	0.909	1.012	+0.103	Average	Good	20261010_CHW@CLE
Petey Halpin	CLE	Strikeouts	Sean Burke	0.829	0.921	+0.092	averagen<floor	Good	20261010_CHW@CLE
Chase DeLauter	CLE	Strikeouts	Sean Burke	0.733	0.822	+0.089	Good	Good	20261010_CHW@CLE
José Ramírez	CLE	Strikeouts	Sean Burke	0.700	0.787	+0.087	Very good	Good	20261010_CHW@CLE
Brayan Rocchio	CLE	Strikeouts	Sean Burke	0.612	0.686	+0.075	Good	Good	20261010_CHW@CLE
Steven Kwan	CLE	Strikeouts	Sean Burke	0.504	0.570	+0.066	Very good	Good	20261010_CHW@CLE
Sam Antonacci	CHW	Hits	Gavin Williams	0.854	0.799	-0.055	Good	Good	20261010_CHW@CLE
Miguel Vargas	CHW	Hits	Gavin Williams	0.837	0.782	-0.055	Average	Good	20261010_CHW@CLE

Per game

Club Player

20261010_CHW@CLE — team layer withheld (values_fresh); runs moved away -0.6051 / home -0.4448

Sean Burke (RHP, 19.8 BF)
H 0.2062 Good · SO 0.2528 Good · BB 0.0893 Average · HR 0.0286 Average

Gavin Williams (RHP, 20.5 BF)
H 0.1978 Good · SO 0.3179 Very good · BB 0.0854 Average · HR 0.0330 Below average

CHW hitters vs Gavin Williams and the bullpen

The pipeline applied no club-wide rescale to CHW 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
Sam Antonacci	LF	Gavin Williams + bullpen	4.16	0.854 → 0.799	-0.055	0.834 → 1.095	+0.261	0.357 → 0.372	+0.015	0.067 → 0.069	+0.002

Kyle Teel	C	Gavin Williams + bullpen	4.10	0.742 → 0.693	-0.049	1.813 → 2.316	+0.503	0.440 → 0.458	+0.018	0.104 → 0.107	+0.003

Miguel Vargas	3B	Gavin Williams + bullpen	4.17	0.837 → 0.782	-0.055	0.965 → 1.252	+0.286	0.555 → 0.577	+0.021	0.157 → 0.161	+0.004

Munetaka Murakami	1B	Gavin Williams + bullpen	4.40	0.747 → 0.696	-0.051	1.885 → 2.300	+0.415	0.662 → 0.687	+0.025	0.214 → 0.219	+0.005

Andrew Benintendi	DH	Gavin Williams + bullpen	3.43	0.722 → 0.674	-0.048	1.034 → 1.314	+0.280	0.318 → 0.331	+0.013	0.090 → 0.092	+0.002

Tristan Peters	RF	Gavin Williams + bullpen	3.76	0.787 → 0.736	-0.051	0.962 → 1.248	+0.285	0.309 → 0.322	+0.013	0.085 → 0.087	+0.002

Chase Meidroth	2B	Gavin Williams + bullpen	3.86	0.848 → 0.793	-0.055	1.106 → 1.415	+0.309	0.355 → 0.370	+0.014	0.089 → 0.091	+0.002

Colson Montgomery	SS	Gavin Williams + bullpen	3.55	0.614 → 0.572	-0.042	1.709 → 2.083	+0.375	0.324 → 0.337	+0.013	0.132 → 0.135	+0.003

Brenton Doyle	CF	Gavin Williams + bullpen	3.52	0.741 → 0.692	-0.049	1.466 → 1.815	+0.350	0.233 → 0.243	+0.010	0.092 → 0.095	+0.002

CHW starter against the CLE 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
Sean Burke	Starter	CLE	19.75	4.216 → 4.266	+0.050	5.087 → 4.321	-0.767	1.868 → 1.969	+0.101	0.706 → 0.644	-0.062

CHW bullpen: the full relief corps (7 arms)

Exposure. Starter 19.8 batters faced, bullpen 15.7 (44.3% of the game), from the club's real 35.5 staff batters faced per game over 2 complete games in the last 30 days. The published staff alone would have given the bullpen 47.5% 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
Sean Newcomb	Long Reliever	pregame bullpen, published staff	38	7	21.0%	3.31
David Sandlin	Reliever	pregame bullpen	5	1	18.5%	2.91
Grant Taylor	Reliever	pregame bullpen	0	0	17.1%	2.70
Anthony Kay	Reliever	pregame bullpen	9	2	13.5%	2.13
Jordan Hicks	Middle Reliever	pregame bullpen, published staff	0	0	11.3%	1.78
Noah Schultz	Long Reliever	pregame bullpen, published staff	0	0	9.8%	1.54
Bryan Hudson	Reliever	pregame bullpen	16	5	8.7%	1.36

CHW relievers against the CLE 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
Sean Newcomb	Long Reliever	CLE (21% of relief)	3.31	1.030 → 1.042	+0.012	1.246 → 1.052	-0.193	0.435 → 0.459	+0.024	0.097 → 0.088	-0.009

David Sandlin	Reliever	CLE (18% of relief)	2.91	0.630 → 0.637	+0.007	0.680 → 0.575	-0.105	0.242 → 0.255	+0.013	0.098 → 0.089	-0.009

Grant Taylor	Reliever	CLE (17% of relief)	2.70	0.532 → 0.539	+0.006	0.801 → 0.686	-0.115	0.242 → 0.255	+0.013	0.071 → 0.065	-0.006

Anthony Kay	Reliever	CLE (14% of relief)	2.13	0.478 → 0.483	+0.005	0.391 → 0.328	-0.064	0.174 → 0.184	+0.009	0.065 → 0.059	-0.006

Jordan Hicks	Middle Reliever	CLE (11% of relief)	1.78	0.912 → 0.923	+0.011	1.223 → 1.044	-0.179	0.602 → 0.633	+0.031	0.098 → 0.089	-0.009

Noah Schultz	Long Reliever	CLE (10% of relief)	1.54	1.538 → 1.556	+0.018	1.623 → 1.367	-0.257	0.350 → 0.369	+0.019	0.134 → 0.122	-0.012

Bryan Hudson	Reliever	CLE (9% of relief)	1.36	0.312 → 0.315	+0.004	0.289 → 0.243	-0.045	0.099 → 0.105	+0.005	0.035 → 0.032	-0.003

CLE hitters vs Sean Burke and the bullpen

The pipeline applied no club-wide rescale to CLE 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
Steven Kwan	LF	Sean Burke + bullpen	4.49	0.934 → 0.898	-0.036	0.504 → 0.570	+0.066	0.586 → 0.593	+0.006	0.047 → 0.044	-0.004

José Ramírez	3B	Sean Burke + bullpen	4.53	0.882 → 0.848	-0.035	0.700 → 0.787	+0.087	0.515 → 0.521	+0.006	0.103 → 0.096	-0.007

Chase DeLauter	RF	Sean Burke + bullpen	4.28	0.937 → 0.901	-0.035	0.733 → 0.822	+0.089	0.462 → 0.467	+0.005	0.111 → 0.103	-0.008

Jo Adell	DH	Sean Burke + bullpen	4.09	0.876 → 0.842	-0.034	1.209 → 1.340	+0.131	0.253 → 0.256	+0.003	0.125 → 0.116	-0.009

Nathaniel Lowe	1B	Sean Burke + bullpen	3.84	0.791 → 0.761	-0.031	0.999 → 1.108	+0.109	0.377 → 0.381	+0.004	0.120 → 0.111	-0.009

Travis Bazzana	2B	Sean Burke + bullpen	3.90	0.769 → 0.739	-0.030	0.909 → 1.012	+0.103	0.432 → 0.437	+0.005	0.102 → 0.094	-0.007

Patrick Bailey	C	Sean Burke + bullpen	3.75	0.756 → 0.727	-0.030	1.143 → 1.261	+0.118	0.324 → 0.328	+0.004	0.108 → 0.101	-0.008

Petey Halpin	CF	Sean Burke + bullpen	3.48	0.729 → 0.700	-0.028	0.829 → 0.921	+0.092	0.222 → 0.225	+0.003	0.087 → 0.081	-0.006

Brayan Rocchio	SS	Sean Burke + bullpen	3.46	0.741 → 0.713	-0.029	0.612 → 0.686	+0.075	0.265 → 0.268	+0.003	0.079 → 0.073	-0.006

CLE starter against the CHW 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
Gavin Williams	Starter	CHW	20.51	3.730 → 3.704	-0.025	7.658 → 8.272	+0.614	2.239 → 2.462	+0.223	0.403 → 0.451	+0.048

CLE bullpen: the full relief corps (7 arms)

Exposure. Starter 20.5 batters faced, bullpen 16.8 (45.1% of the game), from the club's real 37.4 staff batters faced per game over 19 complete games in the last 30 days. The published staff alone would have given the bullpen 49.3% across 5 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
Joey Cantillo	Reliever	pregame bullpen	60	4	22.0%	3.70
Tanner Bibee	Setup Man	pregame bullpen, published staff	3	1	15.6%	2.62
Erik Sabrowski	Reliever	pregame bullpen	39	10	15.0%	2.52
Daniel Espino	Middle Reliever (High Leverage)	pregame bullpen, published staff	8	3	14.7%	2.47
Shawn Armstrong	Middle Reliever	pregame bullpen, published staff	19	6	11.3%	1.90
Tim Herrin	Middle Reliever	pregame bullpen, published staff	38	10	11.2%	1.88
Hunter Gaddis	Middle Reliever (High Leverage)	pregame bullpen, published staff	52	13	10.4%	1.75

CLE relievers against the CHW 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
Joey Cantillo	Reliever	CHW (22% of relief)	3.70	0.783 → 0.778	-0.005	0.910 → 0.991	+0.081	0.416 → 0.456	+0.040	0.100 → 0.112	+0.012

Tanner Bibee	Setup Man	CHW (16% of relief)	2.62	0.951 → 0.945	-0.006	1.055 → 1.157	+0.102	0.596 → 0.656	+0.060	0.142 → 0.159	+0.017

Erik Sabrowski	Reliever	CHW (15% of relief)	2.52	0.459 → 0.455	-0.003	0.843 → 0.909	+0.066	0.323 → 0.353	+0.030	0.065 → 0.073	+0.008

Daniel Espino	Middle Reliever (High Leverage)	CHW (15% of relief)	2.47	0.978 → 0.972	-0.006	1.665 → 1.819	+0.154	0.394 → 0.433	+0.039	0.182 → 0.204	+0.022

Shawn Armstrong	Middle Reliever	CHW (11% of relief)	1.90	0.570 → 0.566	-0.004	0.996 → 1.089	+0.093	0.338 → 0.371	+0.033	0.059 → 0.066	+0.007

Tim Herrin	Middle Reliever	CHW (11% of relief)	1.88	0.587 → 0.583	-0.004	0.783 → 0.855	+0.072	0.303 → 0.333	+0.029	0.046 → 0.051	+0.005

Hunter Gaddis	Middle Reliever (High Leverage)	CHW (10% of relief)	1.75	0.997 → 0.990	-0.007	0.893 → 0.976	+0.083	0.263 → 0.289	+0.026	0.136 → 0.152	+0.016

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