NBA Advanced Stats in Action: Understanding a Game Through Data (2026 Finals Game 5 Walkthrough)Guides · Basketball · Stats explained

NBA Advanced Stats in Action: Understanding a Game Through Data (2026 Finals Game 5 Walkthrough)

Key points

  • Reading data demonstrated using 2026 Finals Game 5
  • Knicks beat Spurs 94:90, the key lay in free throws
  • Includes a 5-minute game-watching checklist

1Summary First

  • Judging the outcome of this game solely by the score or field goal percentage will mislead you: Knicks 94, Spurs 90; the Spurs had both a better field goal percentage (38.4%) and more assists (18), yet the Knicks won.
  • The answer is hidden in "free throws" of the Four Factors: Both teams had almost identical turnover rates and offensive rebound rates, and the Spurs' effective field goal percentage (eFG%) was even 2.8 percentage points higher; but the Knicks went 20/28 from the free throw line while the Spurs were 12/19, scoring 8 more points on free throws. Minus the 4 fewer points scored from field goals, they won by exactly 4 points.
  • Pace was only 92.8 possessions, much slower than the 2025-26 league average of 99.4; the offensive ratings of the two teams were 101.3 and 96.9 respectively, both well below the league average of 115.8: this was a slow-paced grind where scoring was difficult.
  • The Knicks' offense relied almost entirely on one person: Jalen Brunson scored 45 points by himself with a true shooting percentage of .670; the other Knicks players combined for 49 points with a true shooting percentage of only .373.
  • Read plus-minus with caution: The Spurs' Dylan Harper scored a team-high 25 points, but his plus-minus was -12. The most reliable order to read a game is: Pace → Efficiency → Four Factors → Players → By Quarter & Plus-Minus, taking about 5 minutes.

2Why This Game Was Chosen, and How to Read This Guide

This guide will not re-explain terminology from scratch. For what true shooting percentage (TS%), effective field goal percentage (eFG%), usage rate (USG%), offensive rating, and plus-minus are and how they are calculated, see "Introduction to NBA Advanced Stats"; here, we only demonstrate "when you get a box score, how to read the story step by step." Among the Four Factors, turnover percentage (TOV%), offensive rebound percentage (ORB%), and free throw rate (FT/FGA) were not covered in the introductory guide, so brief explanations will be added when they are used.

There are three reasons for choosing this game:

  1. The score was separated by only 4 points; there must be clear reasons for the outcome, and it cannot simply be brushed off as "who was better."
  2. Traditional stats are misleading: Field goals and assists both favored the Spurs.
  3. It was the final game of the 2025-26 season. The 2026-27 season already tipped off on Oct 21 Taiwan time (see "2026-27 NBA Season Viewing Guide"), so the demonstration uses last season's Finals; the method is identical, and can be applied to any game in the new season following these five steps.

Start with the "Traditional Box Score"

QuarterQ1Q2Q3Q4Total
Knicks1324282994
Spurs2319301890

The Knicks trailed by 10 points after Q1, trailed by 5 points at halftime at 37:42, still trailed by 7 points at 65:72 after Q3, and staged a 29:18 comeback in Q4.

Team StatsKnicksSpurs
Field Goals31/87 (.356)33/86 (.384)
3-Pointers12/3712/37
Free Throws20/2812/19
Rebounds4847
Assists1418
Turnovers1413

(BR box score, team totals.) Looking only at this table, the Spurs shot more accurately, had more assists, and committed only 1 fewer turnover; it is hard to tell why the Knicks won. The following five steps "translate" this table into a story.

3Step 1: Look at Pace (Possessions) First

Because teams play a different number of possessions in each game, the first task is to know how many possessions were in this game. BR's possession formula is:

Team Possessions ≈ FGA + 0.4 × FTA − 1.07 × [ORB ÷ (ORB + Opponent DRB)] × (FGA − FG) + TOV

Calculate it for each team and average them:

CalculationPossessions
Knicks87 + 0.4 × 28 − 1.07 × [13 ÷ (13 + 33)] × (87 − 31) + 14Approx. 95.3
Spurs86 + 0.4 × 19 − 1.07 × [14 ÷ (14 + 35)] × (86 − 33) + 13Approx. 90.4
Average(95.3 + 90.4) ÷ 292.8

This matches the Pace (92.8) listed on BR's Four Factors table. Compared to the 2025-26 regular-season league average of 99.4 possessions per game (BR Season Summary), this game had 6.6 fewer possessions, making it a slow pace.

The takeaway from this step: A low score does not necessarily mean the offense was poor; it could also mean there were fewer possessions. To determine whether the offense was good or not, move on to Step 2.

4Step 2: Offensive Rating—Converting the Score to "Per 100 Possessions"

Offensive Rating (ORtg) = Points ÷ Possessions × 100:

  • Knicks: 94 ÷ 92.8 × 100 ≈ 101.3
  • Spurs: 90 ÷ 92.8 × 100 ≈ 96.9 (BR calculates using unrounded possession counts)

The Knicks were 4.4 points higher. Checking this in reverse: 4.4 × 0.928 ≈ 4.1, which is the 4-point scoring margin. In other words, the offensive rating gap merely converts the score into per-100-possession units; it does not tell you the reason. For the reason, look at the next step.

Additionally, both teams' ORtg values were far below the 2025-26 league average of 115.8. Be careful: that average is from the regular season, and playoff defensive intensity is inherently different; comparing a single game to it can only tell you that "scoring was very difficult in this game," not that both teams' offenses deteriorated.

5Step 3: Four Factors—Finding Out "Where It Was Won"

Basketball analyst Dean Oliver proposed the "Four Factors," positing that teams win games based on four things and assigning approximate weights: Shooting (40%), Turnovers (25%), Rebounding (20%), and Free Throws (15%) (BR: Four Factors). BR's box score includes this table directly; the results for this game were:

FactorMetricKnicksSpursAdvantage2025-26 League Avg
ShootingeFG%.425.453Spurs.546
TurnoversTOV% (lower is better)12.412.1Virtual tie12.7
ReboundingORB%28.328.6Virtual tie26.0
Free ThrowsFT/FGA.230.140Knicks.206

(BR Four Factors; league averages from BR 2025-26 regular season.) Examining them item by item:

  • Shooting: Both teams were poor. The Knicks made only 4 of 22 shots in Q1 (BR quarter-by-quarter table), finishing with a game eFG% of .425; the Spurs' .453 was slightly better, but also below the league average of .546. On this factor, the Spurs won.
  • Turnover Rate (TOV%): The formula is Turnovers ÷ (FGA + 0.44 × FTA + Turnovers), representing how many times possession is lost per 100 plays. At 12.4 vs. 12.1, there was no meaningful difference. One small detail: the sum of individual Knicks players' turnovers was only 10, whereas the team column listed 14, which is 4 more than the individual total (for the Spurs, it was 12 vs. 13); the extras are team turnovers not charged to an individual player. NBA.com's team turnover column listed 10 and 12, whereas BR listed 14 and 13; this is a difference between the two websites, not a calculation error by either. At the individual level, the most noticeable was the Knicks' Karl-Anthony Towns: in 22 minutes and 44 seconds of playing time, he had 5 turnovers and fouled out with 6 fouls.
  • Rebound Rate (ORB%): The formula is Offensive Rebounds ÷ (Offensive Rebounds + Opponent Defensive Rebounds), representing the proportion of missed shots a team recovers. At 28.3% vs. 28.6%, there was likewise no difference, and both teams were above the league average of 26.0%. The Knicks' Mitchell Robinson played only 19 minutes and 59 seconds yet grabbed 6 offensive rebounds; the Spurs' Victor Wembanyama also grabbed 6. The two teams canceled each other out.
  • Free Throws (FT/FGA): The formula is Free Throws Made ÷ Field Goal Attempts, reflecting both "getting to the line" and "making free throws." The Knicks were .230 and the Spurs were .140; on this factor, the Knicks won decisively.

Converting the Four Factors Back into Points

Scoring SourceKnicksSpursDiff
2-Pointers38 (19 made)42 (21 made)-4
3-Pointers36 (12 made)36 (12 made)0
Free Throws20 (20/28)12 (12/19)+8
Total9490+4

(Converted from made shots in the BR box score.) In field goals, the Knicks scored 4 fewer points; in free throws, they scored 8 more points, resulting in the final +4. This game was not won because the Knicks shot better from the floor, but because the Knicks earned 9 more free throw attempts (28 to 19).

Wrapping Up with True Shooting Percentage (TS%)

TS% accounts for both field goals and free throws, serving as the final checkpoint for "overall efficiency": Knicks .473, Spurs .477. Both teams had virtually identical overall efficiency. Why then were the Knicks able to score 4 more points? Because they had more "scoring attempts": FGA + 0.44 × FTA was 87 + 0.44 × 28 ≈ 99.3 for the Knicks and 86 + 0.44 × 19 ≈ 94.4 for the Spurs, giving the Knicks roughly 5 more attempts. With identical efficiency and 5 more opportunities, they scored 4 more points. The surplus opportunities came almost entirely from free throws.

Fourth Quarter: How Free Throws Decided the Outcome

The final 12 minutes made this especially apparent:

Q4KnicksSpurs
Points from Field Goals1816
Free Throws11/16 (11 pts)2/6 (2 pts)
Total2918

(BR Q4 quarter-by-quarter table.) In Q4, field goals differed by only 2 points, while free throws had a 9-point margin; 80% of the 11-point differential came from the free throw line. NBA.com's post-game report noted that with less than 8 minutes remaining in Q4, the Knicks trailed 73:83, after which Brunson scored 10 consecutive points to tie the game, finishing with 29 points in the second half and 15 points in Q4 (NBA.com Postgame 4 Takeaways); BR's quarter-by-quarter table shows that of those 15 points, 7 came from free throws (7 of 8 from the line).

6Step 4: Player Level—Who Boosted and Who Dragged Down Efficiency

Having learned at the team level that the game was "won at the free throw line," the next question is: who created these free throws and points? Looking at scoring and advanced metrics side by side (TS% and USG% from BR advanced box score):

PlayerPTSFGFTTS%USG%
Jalen Brunson (Knicks)4514/2713/15.67037.7%
Other Knicks combined4917/607/13.373—
Dylan Harper (Spurs)2510/193/6.57831.1%
Victor Wembanyama (Spurs)197/194/5.44827.4%
Fox (Spurs)73/150/0.23319.5%
Castle (Spurs)61/104/6.23721.8%
Vassell (Spurs)125/80/0.7509.1%

(BR advanced table; "Other Knicks combined" was calculated in this article by subtracting Brunson from BR team totals, with TS% calculated using 49 ÷ (2 × (60 + 0.44 × 13)).)

How to read this table:

  • Brunson = the engine of this game. With a usage rate of 37.7% and TS% of .670, he carried the heaviest workload while maintaining the highest efficiency; moreover, he shot 15 free throws alone, compared to 19 for the entire Spurs team and only 13 combined for the rest of the Knicks. The rest of the Knicks had a combined TS% of just .373, meaning the offense was almost single-handedly kept afloat by him.
  • The Spurs' problem was in the backcourt. Fox and Castle combined shot 4 for 25, scoring only 13 points (Fox 7, Castle 6), both with TS% below .24, yet with usage rates near 20% and 22%. Possessions spent on highly inefficient players were a primary reason why the Spurs' shooting efficiency was only marginally better than the Knicks'.
  • Wembanyama is a blind spot for advanced stats. Scoring 19 points on 19 shots with a .448 TS% looks unimpressive, but his 14 rebounds and 5 blocks do not appear in TS%. TS% only measures "scoring efficiency"; looking at it alone will undervalue defensive stars.
  • Vassell had the highest efficiency and the least usage: TS% of .750 but a usage rate of only 9.1%, going 5 for 8. Whether this means "he should have been given more touches" or "he just happened to get lucky" cannot be distinguished from a single game.

Single-game player TS% fluctuates dramatically; the reliability of the number is very different for someone with 8 attempts versus someone with 27 attempts. It is suitable for explaining what happened in this specific game, but cannot serve as an overall evaluation of a player.

7Step 5: By Quarter and Plus-Minus—Finding Turning Points, But Don't Over-rely on It

Finally, look at plus-minus. Above BR's box score, there are toggle buttons for Q1, Q2, H1, Q3, Q4, and H2, allowing you to view player stats for a specific quarter or half.

First look at full-game plus-minus (BR box score):

PlayerMPPTS+/-
Johnson (Spurs)16:097+17
Hart (Knicks)39:1013+15
Brunson (Knicks)41:0745+10
Vassell (Spurs)39:1212-11
Harper (Spurs)31:0725-12
Shamet (Knicks)13:135-14

Three pitfalls:

  1. Scoring the most points does not guarantee a good plus-minus. Harper was the Spurs' leading scorer, yet his plus-minus was -12. The Spurs lost by 4 points over the full game; calculated over the 16 minutes and 53 seconds he sat on the bench, the Spurs actually outscored their opponent by 8 points (-4 − (-12) = +8). This does not mean the Spurs are better without Harper: plus-minus is influenced by teammates and opponents on the court at the same time, and over a single game of just a few dozen minutes, it contains significant noise.
  2. Players with short playing time can also produce extreme numbers. Johnson played only 16 minutes, shooting 2 for 5 for 7 points, yet had a game-high +17, which was 29 points higher than top-scorer Harper (-12).
  3. Examine it alongside "On/Off" court splits. The Knicks won by 4 points overall, and Hart was +15 on the court; by implication, during the 8 minutes and 50 seconds he rested, the Knicks were outscored by 11 points (4 − 15 = -11). This signal of "collapsing as soon as he sits" is more informative than looking at a standalone +15, though it remains a single-game sample.

How to connect plus-minus back to earlier conclusions: The turning point for the Knicks in this game occurred in Q4. Looking at the Q4 quarter table, Brunson was +14 on the court in Q4, Anunoby was +14, and Hart and Bridges were both +11, while the Spurs scored only 18 points in that quarter. This corroborates Step 3's finding that "in Q4, the Knicks scored 11 points from free throws while the Spurs had 2": Plus-minus tells you "when," the Four Factors tell you "why," and putting the two together yields the complete story.

85-Minute Game-Watching Checklist

Open any box score on Basketball-Reference or NBA.com and follow the order below. Each step takes about 1 minute.

StepWhere to LookAsk YourselfAnswer for This Game
1. PacePace on the Four Factors tableFaster or slower than league average (99.4 in 2025-26)?92.8, slow
2. EfficiencyORtg on the Four Factors table, or calculate Points ÷ Possessions × 100Who is higher? By how much? Does it match the score margin?101.3 vs. 96.9, diff of 4.4, matches
3. Four FactorsFour Factors table: eFG%, TOV%, ORB%, FT/FGAWho holds the advantage in each of the four? Which has the largest gap?Shooting to Spurs, free throws to Knicks, rest tied
4. PlayersAdvanced table: TS%, USG%How efficient is the highest-usage player? Who boosted and who dragged down efficiency?Brunson carried at .670; Fox and Castle dragged it down
5. Turning PointsQuarter toggles (Q1–Q4) + plus-minusIn which quarter was the gap opened? Does plus-minus align with earlier findings?Q4 29:18, decided by free throws

Wrap It Up in One Sentence After Reading

If you can write it out following this template, it means you truly understood it:

Example: "This game had a slow pace; the Knicks won on free throws, primarily propelled by Brunson and the 4th quarter."

Three Reminders

  • Single-game numbers have significant noise, especially player TS% and plus-minus. A conclusion only holds weight when supported by two or three mutually corroborating indicators.
  • Compare against the league average of the relevant season, rather than looking only at absolute values. The averages in this article are from the 2025-26 regular season; if looking at the playoffs, switch to the playoff averages.
  • Advanced stats do not capture the entirety of defense: Blocks, rebounds, and contested shots are not necessarily reflected in TS% or ORtg.

9Frequently Asked Questions

Q: Is it reliable to look at the data from only a single game?

The application in this guide is to "explain what happened in this specific game," not to "evaluate how good or bad a player is." Single-game TS% and plus-minus have tiny sample sizes; evaluating a player requires at least a full season. In contrast, team-level Four Factors make it clear within a single game where the victory was won.

Q: Why is the possession formula here different from the one in "Introduction to NBA Advanced Stats"?

The introductory guide used a simplified version (FGA + 0.44 × FTA − ORB + TOV) to facilitate mental math and grasp the concept. The formula BR actually uses is more detailed: the free throw coefficient is 0.4, and offensive rebounds are adjusted by the proportion of opponent defensive rebounds; only with this formula can you calculate the exact Pace and ORtg matching BR. Both are estimates, not the official tracked counts of actual possessions.

Q: Do numbers differ between BR and NBA.com?

For this game, field goals, free throws, rebounds, and points were completely identical across both; only the team turnover column differed (NBA.com listed 10 and 12, BR listed 14 and 13) because BR's team totals include turnovers not credited to an individual player. The Four Factors table belongs to BR, and BR also allows toggling across quarters and halves.

Q: What should I do for games without a Four Factors table?

Just calculate them yourself; all you need are team totals for field goal attempts, makes, 3-pointers, free throws, rebounds, and turnovers. The formulas for eFG%, TOV%, and FT/FGA are all written above, while ORB% requires the opponent's defensive rebounds.

Q: Are the weights of the Four Factors (40/25/20/15) fixed?

Those were Dean Oliver's original estimates, and BR also labels them as "approximate" weights. Their practical purpose is to remind us that "shooting is most important"; they do not need to be used for precise weighting. This game is a prime example: free throws carry only a 15% weight, yet they happened to decide the outcome.

To review how each metric is calculated, see "Introduction to NBA Advanced Stats"; to understand where the Finals fit into the overall postseason, see "NBA Play-In Tournament and Playoffs"; to learn why there can be a disparity between net rating and win-loss record, see "Pythagorean Expectation".

10Sources

Last updated: 2026-10-24

Illustrated portrait of Xiao Qiao

About the author

Xiao Qiao

NBA columnist

Editorial pen name

Xiao Qiao covers NBA games, season formats and advanced statistics. She looks beyond the score to pace, efficiency, usage and lineup context, separating short-term shooting swings from longer trends and explaining each number with examples that new data-minded fans can follow.

View profile and articles ›

‹ Back to guides · Glossary