Knicks 2026 Historic Run — Findings Outline

Author

Justin Pietsch

Published

July 4, 2026

Draft

Internal analyst reference. Cross-referenced to knicks_2026_historic_results.md. Update when findings prose or RESULTS change.

§1. The Verdict (Up Front)

  • Record: 16-3, win rate 0.842, 88th pct among 43 champions (1983-84 through 2025-26).
  • Raw margin: +14.9 pts/game, 100th pct (best in dataset).
  • Adj margin (reg-SRS): +11.2 pts/game, 100th pct (best in dataset). Opp SRS weighted by games played per series.
  • Pace caveat: 2025-26 is highest-scoring era (115.6 pts/team/game); but possessions only ~4% above average — efficiency, not pace. Per-100-poss margin stays #1.
  • East strength: West-East SRS gap +0.39 pts/game = 37th pct of West dominance (below historical mean +0.78). East was not weak.
  • Health: No opponent missed rotation players to injury (avg availability 98%). Can’t rule out playing hurt (Fox, Spurs).
  • Spurs: Final-round opponent elevated from reg-SRS +8.28 to playoff SRS +14.48 — the biggest improvement of any Knicks opponent.
  • knicks_2026_historic_results.md refs: §1 THE RAW CLAIM, §10 VERDICT

§2. The Raw Numbers: Two Sweeps and a Tight Finals

  • 16-3 record; win rate 0.842 vs. champion mean 0.752; best (2016-17 Warriors) 0.941.
  • +14.9 pts/game margin, 100th pct; mean among champions +7.0 pts/game.
  • 95% CI on margin (19 games): [+7.4, +22.4]. Even lower bound ranks 63rd pct.
  • Round breakdown: Hawks 4-2 (+17.5), 76ers 4-0 (+22.2), Cavaliers 4-0 (+19.2), Spurs 4-1 (+2.4).
  • Finals: 4 of 5 games decided by 4 pts or fewer.
  • knicks_2026_historic_results.md refs: §1 THE RAW CLAIM

§3. Was the East Weak?

  • West-East SRS gap +0.39 pts/game = 37th pct. Historical mean +0.78, std 1.81. Well within normal variation (z = -0.21).
  • East inter-conference win rate 0.487 (near parity). 37th pct (100th = worst for East).
  • Top 3 most West-dominant: 2013-14 (+4.08), 2003-04 (+3.73), 2000-01 (+3.11). 2025-26 well below those.
  • Knicks games-weighted opp SRS +3.67 = 49th pct among champions (mean +3.68). Schedule at historical median.
  • knicks_2026_historic_results.md refs: §2 WAS THE EAST WEAK, §3 HOW WEAK IS THE EAST HISTORICALLY, §4 WHO DID THE KNICKS BEAT

§4. Opponent-Adjusted Dominance

  • Adj margin +11.23 pts/game (raw +14.89 minus games-wtd opp SRS +3.67) = 100th pct.
  • Top 5: 2025-26 +11.23, 2016-17 Warriors +10.23, 1986-87 Lakers +9.52, 1990-91 Bulls +9.20, 1985-86 Celtics +7.72.
  • Playoff overperformance vs. reg-SRS prediction: +12.52 pts/game, 97.7th pct (2nd all-time behind 2000-01 Lakers +14.49).
  • Playoff SRS +17.53 vs. reg-season SRS +6.05 = elevation +11.48, 97.7th pct (2nd all-time, behind 2000-01 +12.58).
  • knicks_2026_historic_results.md refs: §5 OPPONENT-ADJUSTED DOMINANCE

§5. Was the East Weak in the Playoffs?

  • Each opponent’s playoff SRS computed excluding Knicks series (independent read). Hawks excluded (no other playoff games).
  • 76ers: reg SRS -0.27, playoff SRS -1.43 (gap -1.16); Cavaliers: reg +3.77, playoff +2.32 (gap -1.45). Both slight dips but based on few games.
  • Spurs: reg SRS +8.28, playoff SRS +14.48 (gap +6.20) — based on full West bracket, most independent. Biggest improvement of any Knicks opponent.
  • Full-run playoff-SRS-adjusted margin: +9.05 pts/game, 100th pct. Effectively tied with 1986-87 Lakers +8.99.
  • knicks_2026_historic_results.md refs: §6 ROUND-BY-ROUND (playoff SRS table)

§6. The Rise: Biggest in the Field, 2nd-Biggest Ever

  • Knicks: reg SRS +6.05 → playoff SRS +17.53, elevation +11.48 — largest of any 2026 playoff team.
  • Spurs: reg SRS +8.28 → playoff SRS +15.13, elevation +6.85 — second largest.
  • OKC led regular season (+11.04) but barely changed (+0.38 elevation).
  • Finals paired the two biggest risers in the field, not the two best regular-season teams.
  • Note: Spurs playoff SRS here (+15.13) includes the Knicks series; the §5 independent measure (+14.48) excludes it.
  • How historic, champion-wide: elevation +11.48 ranks 2nd of 43 (behind 00–01 Lakers +12.58); overperformance +12.52 ranks 2nd of 43 (behind 00–01 Lakers +14.49). Chart: knicks_2026_rise_ranking (both metrics, Knicks + 2000-01 Lakers highlighted).
  • What it’s made of: round-by-round (§5) — East demolitions (+17.5, +22.2, +19.2), Finals near-even (+2.4) and opponent-adjusted negative; rise rests on the East rounds, not the Finals.
  • What’s behind the jump (observable): roster continuity. Core 5 together 56% of reg season vs 89% of playoffs; margin full-core +8.09 vs short-handed +4.08 (full-season +6.33). Health lifts the baseline a couple points but +8.09 still +14.9 short of the playoff margin — health is a piece, not the whole. Chart: knicks_2026_core_continuity. No box-score/lineup data to pin the mechanism.
  • knicks_2026_historic_results.md refs: §5 OVERPERFORMANCE, §6 ROUND-BY-ROUND, §8 PLAYOFF SRS AND ELEVATION, §8b ROSTER CONTINUITY

§7. Ordinary at Home, Dominant on the Road

  • Clutch (decided by <=5 pts): 31.6% of games, 83.7th pct (mean 0.256).
  • Home: 9 games, 77.8% win rate, 23rd pct vs. champions (below average at home).
  • Away: 10 games, 90.0% win rate, 97.7th pct (better than all but one champion).
  • The road dominance is the unusual piece; the home number is ordinary.
  • knicks_2026_historic_results.md refs: §7 OTHER DEFLATORS

§8. What the Betting Market Said

  • 14-5 ATS; avg beat spread by +12.9 pts/game across 19 games.
  • ATS vs. East rounds: 11-3 (R1 4/6 +13.5, R2 3/4 +18.2, CF 4/4 +16.5).
  • Finals ATS: 3/5, covered as slight underdogs (+2.4 actual vs. +2.5 spread, +4.9 ATS).
  • Note: ATS margin is not independent of raw margin, it’s the same scoreline plus a near-zero spread. Series clustering barely changes the effective sample here (ICC ≈ 0.00, effective N ≈ 19 of 19).
  • knicks_2026_historic_results.md refs: §12 BETTING-MARKET EXPECTATIONS

§9. Opponent Availability (Not the Same as Full Health)

  • Availability: Hawks 100%, 76ers 96%, Cavaliers 97%, Spurs 100%.
  • Avg 98%; most depleted was 76ers at 96%. Spurs missed no rotation players.
  • Definition: rotation players averaging >=15 min/game across all their 2025-26 playoff appearances; only catches players missing games, not players suiting up hurt (Fox played every Finals game for the Spurs while carrying a known injury).
  • knicks_2026_historic_results.md refs: §11 OPPONENT PLAYER AVAILABILITY

§10. How Solid Is the #1 Ranking?

  • Bootstrap (20,000 re-draws): #1 in 59.6% of re-runs; top 3 in 69.9%; top 5 in 82.4%. 90% interval on adj margin: [+5.08, +17.71].
  • Empirical-Bayes shrinkage: 19 games pin down only 41% of estimate; shrunken margin +6.50 pts/game (95% CI [+1.53, +11.48]), still beats 83% of champions.
  • Adding opponent-SRS uncertainty barely moves result: #1 chance stays ~59.4%. Game variance dominates.
  • Hierarchical (partial-pooling) test — every champion treated as uncertain: Knicks posterior mean +4.70 pts/game (90% CI [+1.80, +7.60]). P(true #1) = 9%. P(top 3) = 22.8%. Median rank 9.
  • Streakiness (huge sweeps + tight Finals) makes the run weaker evidence than a steady one — hence the hard pull from +11.2 to ~+4.7 in the hierarchical model.
  • Honest bottom line: clearly elite, very likely top-handful all-time, but data cannot crown them the single best. Chance of being true #1 runs from ~60% (holding history fixed) to ~9% (fairest test).
  • knicks_2026_historic_results.md refs: §13 ROBUSTNESS, §14 HIERARCHICAL

§11. A Second Opinion: Other Rating Systems

  • Bradley-Terry (wins-only, no margins): Knicks adj margin +11.37, rank #1 of 43. Correlation with SRS-adjusted: +0.990. Blowouts are not an artifact — dominance is real.
  • Capped-margin SRS (clip each game at ±15, robust to blowouts): adj margin +11.93, rank #1 of 43, corr with SRS +0.996. Schedule strength isn’t a blowout artifact either.
  • Elo (recency-weighted): opp rating +5.53 (tougher than SRS’s +3.67 — opponents played well down the stretch). Elo-adjusted margin +9.36, rank #3 of 43 (behind 2016-17 Warriors and 1990-91 Bulls).
  • All four systems put Knicks in top three (#1 on three; #3 only on Elo). What shifts them off #1 is crediting opponents’ recent form (Elo), not stripping out margins (BT) or capping blowouts.
  • knicks_2026_historic_results.md refs: §15 ELO CROSS-CHECK, §16 WINS-ONLY (BRADLEY-TERRY) CROSS-CHECK, §21 CAPPED-MARGIN (ROBUST) SRS CROSS-CHECK

§12. How Unlikely Was a 16-3 Run?

  • Forward model (reg-season SRS only): Knicks given only 14.7% title probability (Spurs out-rated them in regular season).
  • Expected losses in title run: 6.5 (actual: 3). P(championship run losing <=3 games): 6.7%. P(title AND <=3 losses): 1.0%.
  • Path-only caveat: model follows only the Knicks’ realized bracket, not the full field. Strongest reg-season team was Oklahoma City Thunder (+11.04 SRS), ahead of the Spurs (+8.28); OKC exited before the Finals so only the Spurs are scored.
  • Echoes the margin overperformance in §4: the Knicks played far above their regular-season level.
  • knicks_2026_historic_results.md refs: §18 SERIES-LEVEL WIN-PROBABILITY MODEL

§13. What Does NOT Diminish the Run

  • Synthesis section (prose only, no new RESULTS analysis): scorecard of the six deflationary objections, each with a verdict and a pointer to the section that tests it.
  • East weak → no (§3, gap +0.39); bracket soft → no (§3–§4, opp SRS +3.67, 49th pct); early-round teams fading → a little at the edges (§5, 76ers/Cavs ~1.2–1.5 below on few games, Spurs +6.2); margins padded → no (§11, wins-only #1); scoring-era inflation → mostly no (§14, per-possession still #1); opponent injured → no games missed (§9, 98% availability; can’t rule out playing hurt).
  • Two genuine caveats are not on the list: the 19-game small sample → single best only ~9% under the fairest test (§10); and the yardstick → the relative lenses (era spread §14, Elo recency §11) read the run as top-five, not a clear first.
  • Full per-objection treatment lives in the companion doc knicks_2026_historic_investigation.md.
  • knicks_2026_historic_results.md refs: none new (points to §3, §4, §5, §9, §11, §13, §14 of RESULTS via the cited findings sections).

§14. Comparing Across Eras: How Much of #1 Survives?

  • The cross-era goal made first-class: “most dominant ever” depends on the yardstick. Lays the techniques side by side.
  • Level adjustments leave the Knicks #1: scoring-share drops the raw margin to 3rd but the opp-adjusted margin stays #1; per-100 possessions (pace only) also #1 (2025-26 pace just ~4% above average, the scoring is efficiency).
  • Dispersion adjustment is the one that changes it: SRS spread grew from 3.1 (1984) to 5.9 (2026); graded on the era’s spread (z = adj_margin / season SRS SD), Knicks are +1.90 SD, rank #5 of 43; 16–17 Warriors lead (z +2.48).
  • Ladder of ranks: raw 1st, opp-adj 1st, + scoring-share 1st, + per-100 1st, + spread-standardized 5th, Elo (recency) 3rd. Absolute measures → #1; relative measures (spread, recency) → top-five.
  • knicks_2026_historic_results.md refs: §22 SPREAD-STANDARDIZED DOMINANCE (Z-SCORE); §9 ERA/SCORING-LEVEL, §17 POSSESSIONS

§15. Limitations

  • 19 games: wide CI. Comparison set only 43 champions.
  • Pre-1997: PLUS_MINUS null; derived from PTS (algebraically exact for game margins).
  • Era comparisons rest on §14 choices; spread-standardized z mixes playoff margin with regular-season spread, so the exact rank (~#5) is not settled though the direction is robust.
  • Playoff-SRS adjustment for §5 (opponent independent form) excludes Hawks (no independent data). Hawks series reads as raw only.

§16. Methodology

  • Python (pandas, numpy). Data via nba_api (LeagueGameFinder, LeagueStandingsV3).
  • SRS: least-squares system (I − A) @ srs = mean_margin, constrained to sum = 0.
  • Full details in knicks_2026_historic_results.md (linked as a companion doc, not inlined) and knicks_2026_analysis.py.

Appendix A: Companion Documents

  • Findings now end with “Appendix A: Companion Documents” linking the standalone Regression Results (RESULTS), The Investigation, the one-page Summary, the Stats Explainer, and this Outline.
  • The report is built with include_appendix=False: RESULTS is no longer inlined as a report appendix; it is a linked companion doc.

Appendix B: Every Champion, Opponent-and-Scoring-Adjusted

  • Full ranked table of all 43 champions by opp+scoring-adjusted margin (rank, season, champion, Raw/Opp-adj/Opp+Scoring, plus Era-spread score and Spread Rank for the era-spread normalization). Knicks +10.05 #1, 2016-17 Warriors a hair behind (effective tie); on Spread Rank the Knicks fall to 5th and the Warriors lead.
  • Reading the shape: scores are bell-shaped, not a power law. Mean +3.29, 65% within 1 SD; Knicks sit +2.0 SD above the mean (the expected spot for the best of 43 draws). Statistical echo of §10: real #1, but the leading edge of an ordinary spread.
  • knicks_2026_historic_results.md refs: §20 APPENDIX: FULL OPP+SCORING-ADJUSTED CHAMPION RANKING

Appendix C: The Full-Field Title Odds

  • Forward bracket sim of all 16 playoff teams (generalizes §12’s realized-path model): seeds into the fixed bracket, plays every round from reg-season SRS. Favorite = Oklahoma City Thunder at 54.5%, more than triple any rival.
  • Knicks only 3.5% as the East 3-seed, far below the realized-path 14.7% (§12); gap is the bracket draw (East 1/2 seeds upset before NY met them), not play. Neither knows they would elevate.
  • knicks_2026_historic_results.md refs: §19 FULL-FIELD TITLE ODDS (FORWARD BRACKET SIM)

Key Numbers Registry

Cross-reference table: each row maps a prose claim to its authoritative value in knicks_2026_historic_results.md. update_registry_lines.py (PostToolUse hook) auto-updates the first line number in “Findings location” whenever knicks_2026_historic_findings.md is edited.

ID Findings location Prose claim knicks_2026_historic_results.md section Authoritative value
K01 §1 L15 16-3, 88th pct win rate §1 THE RAW CLAIM win rate 0.842; 88.4th pct; mean 0.752
K02 §1 L16 +14.9 pts/game raw margin, highest ever §1 THE RAW CLAIM +14.89 pts/game, 100th pct
K03 §1 L18 CI [+7.4, +22.4] §1 THE RAW CLAIM 95% CI [+7.4, +22.4] (t-interval, 19 games)
K04 §1 L24 adj margin +11.2 pts/game, first all-time §5 OPPONENT-ADJUSTED adj margin +11.23, 100th pct
K05 §3 L84 West-East SRS gap +0.39, 37th pct §3 HOW WEAK IS THE EAST gap +0.39, 37.2th pct; mean +0.78
K06 §1 L29 63% of seasons West more dominant §3 HOW WEAK IS THE EAST 37.2th pct West dominance → 62.8% more dominant
K07 §1 L38 Spurs SRS +8.28 reg → +15.13 playoff §6 ROUND-BY-ROUND (field elevation table) Spurs reg +8.28, playoff SRS +15.13 (incl. Knicks series), elev +6.85
K08 §1 L44 avg opponent availability 98% §11 OPPONENT PLAYER AVAILABILITY avg health 98%
K09 §2 L56 win rate 0.842 vs. mean 0.752 §1 THE RAW CLAIM mean 0.752; best 0.941
K10 §2 L71 4 of 5 Finals games decided by 4 pts or fewer §7 OTHER DEFLATORS not directly; 31.6% games <=5 pts
K11 §3 L89 most West-dominant: 2013-14 +4.08 §3 HOW WEAK IS THE EAST top 3: 13-14 +4.08, 03-04 +3.73, 00-01 +3.11
K12 §3 L94 opp SRS +3.67, 49th pct §4 WHO DID THE KNICKS BEAT games-wtd opp SRS +3.67, 48.8th pct
K13 §4 L120 Warriors adj +10.2 (2nd) §5 OPPONENT-ADJUSTED (top 5) 16-17: adj +10.23
K14 §4 L130 overperformance +12.5 pts/game, 2nd all-time §5 OPPONENT-ADJUSTED (overperf) +12.52 pts/game, 97.7th pct
K15 §4 L132 elevation +11.48, 2nd all-time §8 PLAYOFF SRS AND ELEVATION elev +11.48, 97.7th pct; best 2000-01 +12.58
K16 §5 L172 playoff-SRS-adj margin +9.1 pts/game, best §6 ROUND-BY-ROUND (full-run po-SRS adj) +9.05 pts/game, 100th pct; 86-87 +8.99
K17 §5 L166 Spurs reg +8.28, playoff +14.5, +6.2 jump §6 ROUND-BY-ROUND (opp SRS table) Spurs PO SRS +14.48, gap +6.20
K18 §6 L186 Knicks elevation +11.48, largest in 2026 field §6 ROUND-BY-ROUND (field elevation) Knicks +11.48, Spurs +6.85, Blazers +3.01
K19 §6 L191 Spurs climbed to playoff SRS +15.13, rise +6.85 §6 ROUND-BY-ROUND (field elevation) Spurs reg +8.28, PO +15.13, elev +6.85
K20 §6 L205 OKC led regular season (+11.04), barely changed (+0.38) §6 ROUND-BY-ROUND (field elevation) OKC reg +11.04, PO +11.42, elev +0.38
K21 §7 L220 31.6% games decided by <=5 pts, 83.7th pct §7 OTHER DEFLATORS 0.316; 83.7th pct; mean 0.256
K22 §7 L224 home win rate 77.8%, 23rd pct §7 OTHER DEFLATORS home 9G, 0.778, 23.3th pct
K23 §7 L226 away win rate 90.0%, 98th pct §7 OTHER DEFLATORS away 10G, 0.900, 97.7th pct
K24 §8 L239 beat spread by avg +12.9 pts/game §12 BETTING-MARKET avg ATS margin +12.9; ATS record 14-5
K25 §8 L248 11-3 ATS vs. East opponents §12 BETTING-MARKET (round-by-round ATS) R1 4/6, R2 3/4, CF 4/4 = 11-3
K26 §8 L249 Finals covered as underdogs: 3/5 ATS §12 BETTING-MARKET (round-by-round ATS) Finals 3/5; avg margin +2.4 vs. spread +2.5
K27 §9 L272 avg opponent availability 98% §11 OPPONENT PLAYER AVAILABILITY avg 98%; Spurs 100%
K28 §9 L274 close Finals margin +2.4 reflects real competition (games-missed basis only, not injury-free) §11 OPPONENT PLAYER AVAILABILITY Spurs 100% availability confirmed
K29 §10 L306 #1 in 60% of bootstrap re-runs §13 ROBUSTNESS (bootstrap rank) P(rank #1) = 59.6%
K30 §10 L306 top 3 in 70% §13 ROBUSTNESS P(top 3) = 69.9%
K31 §10 L307 top 5 in 82% §13 ROBUSTNESS P(top 5) = 82.4%
K32 §10 L309 adj margin range [+5.1, +17.7] §13 ROBUSTNESS 90% interval [+5.08, +17.71]
K33 §10 L316 shrunken margin +6.5/game §13 ROBUSTNESS (EB shrinkage) posterior +6.50; 95% CI [+1.53, +11.48]
K34 §10 L316 shrinkage range [+1.5, +11.5] §13 ROBUSTNESS 95% credible interval [+1.53, +11.48]
K35 §10 L337 hierarchical: Knicks true #1 only ~9% §14 HIERARCHICAL P(true #1) = 9.0%; posterior mean +4.70
K36 §10 L345 hierarchical pulls from +11.2 down to ~+4.7 §14 HIERARCHICAL Knicks posterior +4.70; 90% CI [+1.80, +7.60]
K37 §11 L371 wins-only (BT): Knicks #1 of 43 §16 WINS-ONLY (BRADLEY-TERRY) BT adj +11.37, rank #1; corr with SRS +0.990
K38 §11 L381 Elo-adjusted margin +9.4/game, 3rd §15 ELO CROSS-CHECK Elo adj +9.36, rank #3; opp rating +5.53
K39 §12 L406 model gave Knicks only 15% title probability §18 SERIES-LEVEL WIN-PROBABILITY P(title) = 14.7%
K40 §12 L408 only 7% of model championship runs that clean §18 SERIES-LEVEL WIN-PROBABILITY P(title run <=3 losses) = 6.7%
K41 §12 L409 barely 1% produced title AND <=3 losses §18 SERIES-LEVEL WIN-PROBABILITY P(title AND <=3 losses) = 1.0%
K42 §1 L21 possessions only ~4% above average §17 POSSESSIONS-BASED PACE 2025-26 pace 101.8 vs. mean 98.0 = +3.8/game (~4%)
K43 §12 field’s strongest reg-season team was Oklahoma City Thunder (+11.04 SRS) §18 SERIES-LEVEL WIN-PROBABILITY (field note) strongest field SRS +11.04 vs. Spurs +8.28
K44 App.B champion scores bell-shaped, mean +3.29, Knicks +2.0 SD out §20 APPENDIX RANKING (shape) mean +3.29, SD 3.38, 65% within 1 SD
K45 App.C full-field favorite Oklahoma City Thunder 54.5%, Knicks 3.5% §19 FULL-FIELD TITLE ODDS favorite 54.5%, Knicks 3.5% (seed 3), Spurs 14.2%
K46 §11 capped-margin SRS: Knicks adj +11.93, #1 of 43 §21 CAPPED-MARGIN (ROBUST) SRS cap ±15, adj +11.93, rank #1, corr SRS +0.996
K47 §14 spread-standardized: Knicks +1.90 SD, #5 of 43 §22 SPREAD-STANDARDIZED DOMINANCE z +1.90, rank #5; leader 16–17 z +2.48; SD 1984 3.1 → 2026 5.9
K48 §14 era-adjustment ladder: opp-adj #1, scoring-share #1, spread #5 §22 SPREAD-STANDARDIZED + §9 + §17 ladder ranks oppadj 1, scoring-share 1, z 5