Do NBA Players Play Better in Contract Years? What 11 Seasons of Data Show

We tracked 448 NBA contract-year seasons from 2015 to 2025. Players did not play better, and the 2027 free-agent class should not expect a bump either.
Do NBA Players Play Better in Contract Years? What 11 Seasons of Data Show
RotoWire
Win your week.
Every start/sit, every waiver, every schedule edge. We make the call.
Every start/sit, every waiver, every schedule edge. We make the call.

Do NBA Players Play Better in Contract Years?

No. We tracked 448 NBA contract-year seasons across 11 free-agent classes (2015 to 2025), and the average player posted a PER 0.25 lower than the year before. Against similar players who were not on expiring deals, the gap was 0.21 lower. Only 42% of contract-year players improved their PER, compared with 49% of similar players.

The popular idea is that a player with a payday coming pushes harder, takes more shots and produces more. The numbers say otherwise. Points, minutes, usage and assists all slipped slightly in contract years, and efficiency barely moved. Here is the full study, the biggest bumps and busts, and what it means for the 2027 NBA free agents who are in a contract year right now.

Data Study
Is the NBA Contract-Year Bump Real?
We tracked 448 players across 11 free-agent classes (2015-2025) and compared their contract-year seasons to the year before, matched against similar players who were not on expiring deals.
448
Contract-Year Seasons
−0.25
PER vs. Prior Year
42%
Players Who Improved
−0.30
PER Year After
No.
Players in the final year of their contract did not play better. Across 448 qualifying seasons they averaged a PER 0.25 lower than the year before, and 0.21 lower than similar players (same age band, same prior-year PER tier) who were not on expiring deals.
Contract-year players vs. similar non-expiring players
Share whose PER improved
Contract-year players42%
Similar non-expiring49%
Share with a PER jump of 3 or more
Contract-year players8.5%
Similar non-expiring12.0%
Share with a PER drop of 3 or more
Contract-year players10.7%
Similar non-expiring10.3%
Does the bump show up in any group? By age and by prior-year PER
Age <25−0.33n=74
Age 25-27−0.41n=117
Age 28-30−0.27n=129
Age 31++0.10n=128
PER <13−0.31n=144
PER 13-16+0.08n=137
PER 16-20−0.32n=108
PER 20+−0.44n=59
And after the payday?
Players who qualified in the year after signing (n=381) posted a PER 0.30 lower than their contract-year PER (0.22 lower vs. similar players). Because teams pay for strong seasons, some of that is ordinary regression, not a post-payday slump.
How we measured it. Contract year = the last season before a player signed as a free agent (Spotrac, classes 2015-2025). A season counts only with 40+ games and 20+ minutes per game, in both the contract year and the prior year. The comparison group is 1,345 non-free-agent player-seasons matched on age band and prior-year PER tier. Players who retired or went unsigned, and players who signed extensions before free agency, are not in the sample. Averages are descriptive and not proof of cause.
Average change from the prior season to the contract year (n=448)
StatChangeVs. similarRead
PER
−0.25
−0.21
Slightly down
Points
−0.32
−0.32
Slightly down
Minutes
−0.47
−0.30
Flat
Usage %
−0.46
−0.37
Slightly down
Assists
−0.06
−0.12
Slightly down
Rebounds
−0.09
−0.08
Flat
3-Pointers Made
+0.05
−0.01
Flat
Steals
−0.02
−0.02
Flat
Blocks
−0.02
−0.01
Flat
FG% (pts)
+0.08
+0.04
Flat
True Shooting % (pts)
+0.43
+0.16
Flat
Games Played
−0.9
+0.5
Flat
"Vs. similar" subtracts the average change for non-expiring players of the same age band and prior-year PER tier. "Slightly down" means the controlled change is at least two standard errors below zero. Minutes, points and usage all slipped, so there is no hidden role bump behind flat efficiency. Field goal and true shooting percentages ticked up by a few tenths of a point, but neither is statistically distinguishable from zero after matching.
Average PER change vs. similar players, by contract-year season
2014-15+0.49n=54
2015-16−0.24n=49
2016-17+0.36n=43
2017-18+0.22n=38
2018-19−0.55n=57
2019-20−0.16n=43
2020-21−0.40n=45
2021-22−0.68n=22
2022-23−0.87n=35
2023-24−0.71n=34
2024-25−0.23n=28
The bump has faded. Contract-year seasons from 2014-15 through 2017-18 averaged +0.21 PER vs. similar players (n=184). Seasons from 2018-19 through 2024-25 averaged −0.50 (n=264), with every class below zero. That split was chosen after seeing the data, so treat it as descriptive.
Biggest contract-year bumps (PER change vs. prior season)
1
Jimmy Butler
2014-15 · CHI · age 25 · 21.3 next season
PER 13.5 → 21.3
+7.8
2
Quentin Grimes
2024-25 · 2 teams · age 24 · 12.9 next season
PER 8.9 → 15.9
+7.0
3
George Hill
2016-17 · UTA · age 30 · 13.3 next season
PER 13.2 → 19.3
+6.1
T-4
Nikola Vučević
2018-19 · ORL · age 28 · 21.9 next season
PER 19.7 → 25.5
+5.8
T-4
P.J. Tucker
2021-22 · MIA · age 36 · 5.7 next season
PER 5.5 → 11.3
+5.8
6
Lou Williams
2014-15 · TOR · age 28 · 17.7 next season
PER 14.2 → 19.9
+5.7
7
Brandon Ingram
2019-20 · NOP · age 22 · 19.2 next season
PER 13.4 → 18.8
+5.4
8
Bobby Portis
2020-21 · MIL · age 25 · 17.7 next season
PER 14.6 → 19.9
+5.3
T-9
Marvin Williams
2015-16 · CHA · age 29 · 13.7 next season
PER 11.8 → 16.9
+5.1
T-9
Dion Waiters
2016-17 · MIA · age 25 · did not qualify next season
PER 9.4 → 14.5
+5.1
T-9
Mike Conley
2020-21 · UTA · age 33 · 17.2 next season
PER 14.1 → 19.2
+5.1
Biggest contract-year busts
1
Ty Lawson
2015-16 · 2 teams · age 28 · 15.4 next season
PER 18.5 → 9.7
−8.8
2
Kevin Love
2014-15 · CLE · age 26 · 19.0 next season
PER 26.9 → 18.8
−8.1
3
Stephen Curry
2016-17 · GSW · age 28 · 28.2 next season
PER 31.5 → 24.6
−6.9
4
Kevin Love
2022-23 · 2 teams · age 34 · did not qualify next season
PER 19.6 → 13.1
−6.5
T-5
Patrick Beverley
2022-23 · 2 teams · age 34 · 11.6 next season
PER 14.8 → 8.9
−5.9
T-5
Joe Ingles
2021-22 · UTA · age 34 · 10.7 next season
PER 15.9 → 10.0
−5.9
7
Arron Afflalo
2014-15 · 2 teams · age 29 · 10.9 next season
PER 16.0 → 10.7
−5.3
8
Bradley Beal
2021-22 · WAS · age 28 · 19.7 next season
PER 22.7 → 17.5
−5.2
T-9
Marc Gasol
2019-20 · TOR · age 35 · did not qualify next season
PER 16.7 → 11.8
−4.9
T-9
Austin Rivers
2018-19 · 2 teams · age 26 · 10.4 next season
PER 12.6 → 7.7
−4.9
Data: Basketball-Reference (PER and per-game stats), Spotrac (free-agent classes 2015-2025). Method: contract year vs. prior year, 40+ games and 20+ MPG in both seasons, matched to non-expiring players by age band and prior-year PER tier. Updated: October 2026.

NBA Contract Year Bump: The Key Findings

•        PER: −0.25 vs. the prior season (−0.21 vs. similar players), n=448.

•        Who improved: 42% of contract-year players raised their PER, vs. 49% of similar non-expiring players.

•        Big jumps are rarer: 8.5% gained 3+ PER, vs. 12.0% of similar players. About 10.7% lost 3+ PER, vs. 10.3%.

•        Role did not grow: Minutes (−0.47 per game), points (−0.32) and usage (−0.46 points) all slipped.

•        The year after signing: PER fell another 0.30 (n=381), though part of that is ordinary regression after a strong season.

Stat (change vs. prior year)

Contract-year change

Vs. similar players

Read

PER

−0.25

−0.21

Slightly down

Points per game

−0.32

−0.32

Slightly down

Minutes per game

−0.47

−0.30

Flat

Usage %

−0.46

−0.37

Slightly down

Assists per game

−0.06

−0.12

Slightly down

Rebounds per game

−0.09

−0.08

Flat

3-pointers made

+0.05

−0.01

Flat

Steals per game

−0.02

−0.02

Flat

Blocks per game

−0.02

−0.01

Flat

FG% (points)

+0.08

+0.04

Flat

True shooting % (points)

+0.43

+0.16

Flat

Games played

−0.90

+0.48

Flat

"Similar players" are non-expiring players of the same age band and prior-year PER tier. "Slightly down" means the controlled change is at least two standard errors below zero.

How We Measured the NBA Contract Year Effect

Contract-year studies tend to go wrong in two places: who counts as a contract-year player, and what they are compared against. Here is how we handled both.

•        Contract year: the final season before a player signed as a free agent, from Spotrac's free-agent classes for 2015 through 2025 (1,468 signings, covering contract years from 2014-15 through 2024-25).

•        Qualifying seasons: at least 40 games and 20 minutes per game in both the contract year and the prior year. That left 448 contract-year seasons with a usable prior year (586 qualified in the contract year alone, and 381 also qualified the following year).

•        Metrics: PER and usage from Basketball-Reference's advanced tables, plus per-game points, rebounds, assists, steals, blocks, three-pointers made, field-goal percentage, true shooting and games played.

•        Comparison group: 1,345 non-free-agent player-seasons, matched on age band (24 and under, 25-27, 28-30, 31+) and prior-year PER tier (under 13, 13-16, 16-20, 20+). This matters because players naturally rise and fall with age and ability, and high performers regress toward the average.

Limits. The sample only includes players who signed as free agents. Players who retired or went unsigned, and players who signed extensions before free agency, are not in it. 63 of the 1,468 signings could not be matched to a Basketball-Reference season, mostly players who had not yet played in the NBA. These are averages, and they describe what happened without proving why.

More hoops data from RotoWire: Every NBA team's October record since 2016-17: win rates, ATS results, opening-night marks and the toughest 2026-27 October schedules.

The Contract-Year Bump Has Faded Since 2018-19

Early in the sample, contract-year players looked slightly better than similar players. The more recent classes tell a different story.

Contract-year season

Free-agent class

PER change vs. similar

Players (n)

2014-15

2015

+0.49

54

2015-16

2016

−0.24

49

2016-17

2017

+0.36

43

2017-18

2018

+0.22

38

2018-19

2019

−0.55

57

2019-20

2020

−0.16

43

2020-21

2021

−0.40

45

2021-22

2022

−0.68

22

2022-23

2023

−0.87

35

2023-24

2024

−0.71

34

2024-25

2025

−0.23

28

Contract years from 2014-15 through 2017-18 averaged +0.21 PER vs. similar players (n=184). Seasons from 2018-19 through 2024-25 averaged −0.50 (n=264), with every class below zero. We picked that split after looking at the data, so treat it as descriptive rather than a tested finding.

Does the bump show up in any age or ability group?

Group

PER change vs. similar

Players (n)

Read

Age 24 and under

−0.33

74

Flat

Age 25-27

−0.41

117

Flat

Age 28-30

−0.27

129

Flat

Age 31 and older

+0.10

128

Flat

Prior-year PER under 13

−0.31

144

Flat

Prior-year PER 13-16

+0.08

137

Flat

Prior-year PER 16-20

−0.32

108

Flat

Prior-year PER 20+

−0.44

59

Flat

 No group shows a bump, and none of these group-level changes is statistically distinguishable from zero. The closest thing to a gain is players 31 and older at +0.10 PER.

Biggest NBA Contract-Year Bumps and Busts

Averages hide individual stories. These are the largest PER swings from the prior season to the contract year among qualifying players. PER values are Basketball-Reference figures, and ties share a rank.

Biggest contract-year bumps

Rank

Player

Season

Team

Age

Prior PER

Contract PER

Change

1

Jimmy Butler

2014-15

CHI

25

13.5

21.3

+7.8

2

Quentin Grimes

2024-25

2 teams

24

8.9

15.9

+7.0

3

George Hill

2016-17

UTA

30

13.2

19.3

+6.1

T-4

Nikola Vučević

2018-19

ORL

28

19.7

25.5

+5.8

T-4

P.J. Tucker

2021-22

MIA

36

5.5

11.3

+5.8

6

Lou Williams

2014-15

TOR

28

14.2

19.9

+5.7

7

Brandon Ingram

2019-20

NOP

22

13.4

18.8

+5.4

8

Bobby Portis

2020-21

MIL

25

14.6

19.9

+5.3

T-9

Marvin Williams

2015-16

CHA

29

11.8

16.9

+5.1

T-9

Dion Waiters

2016-17

MIA

25

9.4

14.5

+5.1

T-9

Mike Conley

2020-21

UTA

33

14.1

19.2

+5.1

Biggest contract-year busts

Rank

Player

Season

Team

Age

Prior PER

Contract PER

Change

1

Ty Lawson

2015-16

2 teams

28

18.5

9.7

−8.8

2

Kevin Love

2014-15

CLE

26

26.9

18.8

−8.1

3

Stephen Curry

2016-17

GSW

28

31.5

24.6

−6.9

4

Kevin Love

2022-23

2 teams

34

19.6

13.1

−6.5

T-5

Patrick Beverley

2022-23

2 teams

34

14.8

8.9

−5.9

T-5

Joe Ingles

2021-22

UTA

34

15.9

10.0

−5.9

7

Arron Afflalo

2014-15

2 teams

29

16.0

10.7

−5.3

8

Bradley Beal

2021-22

WAS

28

22.7

17.5

−5.2

T-9

Marc Gasol

2019-20

TOR

35

16.7

11.8

−4.9

T-9

Austin Rivers

2018-19

2 teams

26

12.6

7.7

−4.9

Both lists are full of regression to the mean. Players coming off unusually good or bad seasons tend to move back toward their normal level, which is why a contract year can look like a leap or a collapse without any change in effort. Jimmy Butler's 2014-15 season is the largest jump in the sample.

Also from RotoWire: A family of four pays $343 on average to attend an NBA game in 2026-27, from $168 in New Orleans to $1,277 at Knicks games. See all 30 teams.

What Happens After NBA Players Sign Their Contracts?

Among the 381 players who also qualified in the first season after signing, PER fell another 0.30 on average (0.22 vs. similar players), and 42% improved. Minutes and scoring were basically unchanged, while usage slipped slightly (−0.27 points) and shooting efficiency held flat. Because teams pay for strong seasons, some of that is regression, not a post-payday slump.

2027 NBA Free Agents: Who Is in a Contract Year in 2026-27?

Spotrac lists 254 players in the 2027 NBA free-agent class, and 238 of them have 2025-26 stats on Basketball-Reference. After the same 40-game, 20-minute filter used in the study, 116 remain: 45 on true expiring deals (32 unrestricted, 13 restricted), 22 with player options, 48 with club options and 1 with a mutual option. The league has changed a lot this offseason, so check each player's current team and contract status before the season starts.

Expiring deals worth watching

These players are on the final year of their contracts and were productive last season. The table shows their 2025-26 line, and teams are the 2025-26 team listed on Basketball-Reference.

Player

Age

2025-26 team

Type

PER

PPG

Usage %

Marvin Bagley III

28

2 teams

UFA

20.2

10.5

18.6

Michael Porter Jr.

29

BRK

RFA

19.9

24.2

30.5

Deandre Ayton

29

LAL

UFA

18.2

12.5

16.7

DeMar DeRozan

38

SAC

UFA

17.8

18.4

22.8

Nikola Vučević

36

2 teams

UFA

17.7

15.1

21.4

Keldon Johnson

28

SAS

UFA

17.3

13.2

21.0

R.J. Barrett

27

TOR

UFA

17.0

19.3

25.8

Brandon Miller

24

CHO

RFA

16.8

20.2

28.1

Miles Bridges

29

CHO

UFA

16.0

17.1

22.3

C.J. McCollum

36

2 teams

UFA

15.9

18.7

25.7

 Stars with player options

A player option means the player decides in the summer of 2027 whether to opt in or become a free agent, so the 2026-27 season is a contract year only in that sense. Last season's qualifying leaders in this group are Nikola Jokić (PER 32.3), Karl-Anthony Towns (22.0), Kevin Durant (21.1), LeBron James (20.8), Julius Randle (18.4), Rudy Gobert (17.5) and Jrue Holiday (16.4). Their decisions will depend on a lot more than one season of production.

Injury-limited wild cards

Several of the biggest names in the class did not reach 40 games last season, so they fall outside the study's filter and their baselines are noisier.

Player

Type

Games (2025-26)

Minutes per game

PER

Age

Giannis Antetokounmpo

PO

36

28.9

32.6

32

Jimmy Butler

UFA

38

31.1

23.5

38

Anthony Davis

PO

20

31.3

21.5

34

Kristaps Porziņģis

PO

32

24.0

19.6

32

Tyler Herro

UFA

33

31.3

17.1

27

Zach LaVine

UFA

39

31.4

14.7

32

 Kyrie Irving (player option), Fred VanVleet and Damian Lillard (player option) did not appear in Basketball-Reference's 2025-26 tables at all, so there is no recent baseline to compare.

What the contract-year study says about this class

The data gives no reason to expect a bump from any of these players. Contract-year players slipped slightly against similar players in every age band except 31 and older, where the change was a statistically flat +0.10. Players who came in with a PER of 20 or higher averaged −0.44 vs. similar players, though that group is small (n=59) and the gap is not statistically significant.

The typical year-to-year spread is wide, with a standard deviation of about 2.3 PER. About 8.5% of contract-year players jumped 3 or more PER, and about 10.7% fell by that much. The reasonable expectation for any player on this list is a season close to last year's, with real uncertainty in both directions. Any individual "breakout" or "decline" call is a guess the data does not support.

NBA Contract Year FAQ

What is a contract year in the NBA?

A contract year is the final season of a player's current contract, after which he becomes a free agent (or decides whether to opt out). In this study it means the last season before a player signed as a free agent.

Do NBA players play better in contract years?

Not on average. From 2014-15 through 2024-25, contract-year players posted a PER 0.25 lower than the year before, and 0.21 lower than similar players on non-expiring deals. Only 42% improved.

Does performance drop after an NBA player signs a big contract?

Slightly. PER fell another 0.30 in the first season after signing (n=381), though regression after a strong season explains part of it. Minutes and scoring were nearly unchanged.

Who are the best 2027 NBA free agents?

Last season's most productive players in the class include Nikola Jokić, Karl-Anthony Towns, Kevin Durant and LeBron James (all with player options), plus expiring-contract names such as Michael Porter Jr., Deandre Ayton, DeMar DeRozan and R.J. Barrett. Giannis Antetokounmpo and Jimmy Butler are also in the class but missed significant time last year.

Which NBA stars have player options in 2027?

Per Spotrac, they include Nikola Jokić, Karl-Anthony Towns, Kevin Durant, LeBron James, Giannis Antetokounmpo, Anthony Davis, Julius Randle, Rudy Gobert and Jrue Holiday. A player option lets the player choose whether to stay or hit free agency in 2027.

Methodology and Data Sources

Free-agent classes for 2015 through 2025 and the 2027 class come from Spotrac. PER, usage, true shooting and per-game statistics come from Basketball-Reference season tables for 2013-14 through 2025-26. Players were matched by name, with manual fixes where the two sources spell a name differently, and matched to Basketball-Reference's player IDs for every comparison season. Players traded mid-season use their combined season totals. The comparison group excludes every player who signed as a free agent that offseason. Figures are averages and describe what happened without establishing cause. Data as of October 15, 2026.

ABOUT THE AUTHOR
Thomas Leary is a senior editor and sports betting expert for RotoWire, covering all aspects of the gambling industry with an emphasis on sports data. He previously spent over six years at Sports Business Journal, where he helped identify emerging sectors across the industry, such as legalized gambling, and helped launch a digital newsletter division. Outside of work, Thomas can often be found at the golf course, and he still laments the many Louis Oosthuizen major heartbreaks. Thomas is a native of Charlotte, North Carolina and graduated from the University of North Carolina at Wilmington.
RotoWire Logo

Continue the Conversation

Join the RotoWire Discord group to hear from our experts and other NBA fans.

Top News

Related Stories