CURRENT NBA ROSTER · PLAYER INTELLIGENCE

Keshad Johnson

Miami Heat current context with accepted historical evidence kept separate.

Miami Heat · current teamFArizona

AF-PLAYERPendingsample gate

NBA RANK—not ranked

PH$—pending · how PH$ works

STATUSUNSPECIFIEDsource status

EXPECTED MIN0.0governed role

NEXT GAMEvs Minnesota Timberwolves2026-10-21 WHAT IF IMPACTWithheldteam win probability if OUT · model counterfactual

REGULAR-SEASON GAMES42

Player-safe accepted corpus.

CAREER PPG · CORPUS3.8

2019–20 through 2025–26 window.

CAREER RPG · CORPUS2.0

Historical window, not full NBA career.

CAREER APG · CORPUS0.3

Historical window, not full NBA career.

LATEST COMPLETED SEASON

2025–26

29 games

MIN8.3

PTS4.1

REB2.0

AST0.3

STL0.5

BLK0.3

3PM0.5

FG%48.3%

3P%33.3%

FT%70.4%

+/-+1.9

NEXT GAME

Current context

2026–27 · Miami Heat

Assigned schedule

NBA Official roster

2026-10-21 · HOMEvs Minnesota Timberwolves56.4% winProjected 113–116

2026-10-23 · HOMEvs Detroit Pistons50.9% winProjected 113–114

2026-10-24 · AWAY@ Charlotte Hornets38.9% winProjected 113–117

2026-10-26 · HOMEvs Dallas Mavericks66.8% winProjected 113–117

2026-10-28 · AWAY@ Atlanta Hawks38.8% winProjected 114–116

E66 · PLAYER PROJECTION

Loading next-game projection…

PTS / REB / AST / 3PM use a separately validated player-stat model. Game Win, projected team score, Form and milestones keep their own authorities.

Current team Open Player What If

MINUTES + FORM

Role-adjusted performance

CURRENT FORM · 2026–27

Waiting for completed player boxscores

PRESEASON / NO SAMPLE

ProbHoops will activate Last 5 / Last 10 form only after observed 2026–27 player boxscores exist. No current form label is borrowed from 2025–26.

LATEST OBSERVED REFERENCE · 2025–26

Historical Minutes Intelligence

DESCRIPTIVE · NOT CURRENT

Expected minutes were frozen before each historical game. Role-Consistent means actual minutes stayed between 75% and 125% of that pregame reference. Low minutes are diagnostic only: ProbHoops does not infer rest, injury or a coaching decision from the box score.

LATEST ACTUAL25minutes

EXPECTED2.73pregame reference

DEVIATION+817.4%expanded minutes

LAST 5

ALL OBSERVED5 G5 PTS · 1.6 REB · 0.8 AST

ROLE-CONSISTENT2 G0.5 PTS · 0.5 REB · 0 AST

LAST 10

ALL OBSERVED10 G2.6 PTS · 0.8 REB · 0.4 AST

ROLE-CONSISTENT2 G0.5 PTS · 0.5 REB · 0 AST

creation—

efficiency—

minutes—

rebounding—

scoring—

Form label withheld: insufficient role-consistent sample. This is a descriptive historical reference, not a next-game projection, AF-Player input, availability claim or causal diagnosis.

SPLITS + EFFICIENCY

How the production is built

EFFICIENCY

Shooting, derived from the box score

CAREER · 42 GAMES

2PM and 2PA need no source: a two-pointer is a field goal that was not a three. eFG% credits the extra point a three is worth; TS% is the only one of these that accounts for free throws. A rate below its attempt floor is shown as — rather than as a flattering number.

FIELD GOALS53.0%61/115

2-POINT65.2%45/69

3-POINT34.8%16/46

FREE THROW63.9%23/36

EFFECTIVE FG%60.0%

TRUE SHOOTING%61.5%

AST / TO0.75

PTS PER 3617.25rate, not projection

HOME / AWAY

Where the games were played

21 HOME · 21 AWAY

Split	G	MIN	PTS	REB	AST	3PM	STL	BLK
Home	21	9	3.9	2.33	0.43	0.29	0.43	0.33
Away	21	7	3.76	1.67	0.14	0.48	0.43	0.24
Home − Away	—	+2.0	+0.1	+0.7	+0.3	-0.2	0.0	+0.1

Home minus away, computed only when both sides reach the sample floor. A difference is a measurement, not a home-court effect.

CONSISTENCY

What the average hides

42 GAMES

Two players averaging the same points can be 20-21-24 and 5-38-12. Floor, median and ceiling are the 20th / 50th / 80th percentile, linearly interpolated.

Category	Floor	Median	Ceiling	Mean	Volatility
assists	0	0	1	0.29	high
minutes	2	4	13.8	8	high
points	0	2	7	3.83	high
rebounds	0	1	3	2	high

Percentiles, not minimum and maximum: one exceptional night is not a ceiling.

DOUBLE-DOUBLES

Observed frequency across the corpus

42 GAMES

DOUBLE-DOUBLES12.4% of games

TRIPLE-DOUBLES00.0% of games

10+ ASSISTS00.0%

10+ BLOCKS00.0%

Observed frequency across the measured corpus, not a forecast. Category probabilities are not multiplied together: production categories are dependent through minutes and role.

GAME LOG · 2025–26

One log, the window you choose

10 games

Date	Opp	MIN	EXP MIN	MIN Δ	PTS	FG	2P	3P	FT	OR	DR	REB	AST	STL	BLK	TOV	PF	+/-	TS%
2026-04-12	vs ATL	25	2.73	+817.4%	12	6/8	6/6	0/2	0/0	4	0	4	3	2	0	0	4	+19.0	75%
2026-04-09	@ TOR	11	2.36	+365.7%	6	3/3	3/3	0/0	0/0	0	2	2	1	0	0	0	1	+11.0	100%
2026-04-07	@ TOR	3	1.75	+71.9%	7	2/2	1/1	1/1	2/2	0	1	1	0	1	1	0	0	-1.0	122%
2026-04-04	vs WAS	1	1.91	-47.6%	0	0/0	0/0	0/0	0/0	1	0	1	0	0	0	0	0	+1.0	—
2026-03-27	@ CLE	5	4.84	+3.4%	1	0/0	0/0	0/0	1/2	0	0	0	0	0	0	0	0	+6.0	57%
2026-03-23	vs SAS	3	4.65	-35.5%	4	1/1	1/1	0/0	2/2	1	0	1	1	0	0	0	0	+5.0	106%
2026-03-17	@ CHA	20	3.61	+453.5%	15	6/12	3/3	3/9	0/0	1	4	5	1	3	1	2	2	-13.0	63%
2026-03-10	vs WAS	19	—	—	8	3/6	2/2	1/4	1/4	1	0	1	0	2	1	0	1	+7.0	52%
2026-02-21	vs MEM	2	1.13	+76.7%	2	1/3	1/2	0/1	0/0	1	2	3	0	0	1	1	1	-8.0	33%
2026-02-20	@ ATL	3	0.66	+351.6%	5	2/3	1/2	1/1	0/0	0	1	1	0	0	0	0	1	+4.0	83%

All E57 box-score columns remain available. E65 adds the frozen expected-minutes reference and deviation without removing any observed box-score field. The table scrolls inside its own region on narrow screens; highlighted rows are double-doubles and the strongest highlight is a triple-double.

HISTORY

Season and milestone evidence

DESCRIPTIVE MILESTONES · 2025–26

How often did the threshold occur?

historical frequency

These are observed historical frequencies, not current model probabilities and not betting lines.

20+ POINTS0.0%0/29 games

10+ REBOUNDS3.5%1/29 games

5+ ASSISTS0.0%0/29 games

2+ THREES6.9%2/29 games

SEASON HISTORY

Regular-season production

accepted player-safe games

Season	Team(s)	G	MIN	PTS	REB	AST	FG%	3P%	FT%
2025–26	MIA	29	8.3	4.1	2.0	0.3	48.3%	33.3%	70.4%
2024–25	MIA	13	7.3	3.3	2.1	0.3	69.2%	42.9%	44.4%

CURRENT / HISTORICAL BOUNDARY

Current context is active. Historical evidence stays historical.

governed identity

Current roster, literal availability, expected minutes and future-game context come from the governed current product. Historical production and with/without evidence are not rewritten as current-season claims.

Open team profile

IMPACT

Rating and role context

AF-PLAYER

Measured production, 2025–26

PENDING SAMPLE

ProbHoops has observed 29 games of this player in 2025–26 — below the sample floor required to rank. He carries no AF-Player and no PH$. Pending means not enough observation, never a low score.

E67 · ROLE & ROSTER INTELLIGENCE

Distinct signals, side by side

descriptive · current role + historical rating

Value Over Role, AF-Player Movement, Role Change and Star Dependency are separate diagnostics. They do not alter AF-Player, Form, Availability, Game Win or the governed What If scenario.

VALUE OVER ROLE—requires eligible AF-Player + governed role

AF MOVEMENT—two eligible completed seasons required

ROLE CHANGEUNAVAILABLEcomparison unavailable

STAR DEPENDENCY—governed What If sample unavailable

Player movement Opportunities Compare players

Boundary: Star Dependency is model dependency, not causal truth. Opportunity labels are mechanical screeners, not predictions or personnel recommendations. PH$ remains fictional game currency.

PLAYER IMPACT · 2025–26

Pregame role and with/without context

descriptive · not causal

Role measures are based on information frozen before games. With/without appears only when the historical sample passes the research gate; it does not prove that an absence caused the observed team change.

MIA

Miami Heat

EXPECTED MIN3.2

SCORING SHARE2.7%

CREATION SHARE0.3%

PARTICIPATION PRIOR39.4%

With/without sample did not qualify for publication.