Fansided

The Most Unlikely Performances of the Year by Daily RAPM Estimate

April 3, 2015; Los Angeles, CA, USA; Los Angeles Lakers center Robert Sacre (50) attempts a shot against the Portland Trail Blazers during the second half at Staples Center. Mandatory Credit: Gary A. Vasquez-USA TODAY Sports
April 3, 2015; Los Angeles, CA, USA; Los Angeles Lakers center Robert Sacre (50) attempts a shot against the Portland Trail Blazers during the second half at Staples Center. Mandatory Credit: Gary A. Vasquez-USA TODAY Sports
April 3, 2015; Los Angeles, CA, USA; Los Angeles Lakers center Robert Sacre (50) attempts a shot against the Portland Trail Blazers during the second half at Staples Center. Mandatory Credit: Gary A. Vasquez-USA TODAY Sports
April 3, 2015; Los Angeles, CA, USA; Los Angeles Lakers center Robert Sacre (50) attempts a shot against the Portland Trail Blazers during the second half at Staples Center. Mandatory Credit: Gary A. Vasquez-USA TODAY Sports

Almost nightly, basketball fans can run through the eveningā€™s box scores and single out one player who seems to have just thenĀ played the game of hisĀ life. Likewise, fans are also quick to notice when a player plays much more poorly than expected. Naturally, I was curious which box score line was in fact the most out-of-character performance of the year. To do this, I turned to our one game metric, Daily RAPM Estimate (or if you have forgotten about it during the offseason,Ā DRE.)

I went ahead and found the averages and standard deviations for each playerā€™s DRE scores[1. The Daily DRE, if you will.], and then calculated the corresponding z-score for every game in order to look for the most unlikely performances of the year given a playerā€™s usual performance and consistency. [1. Assuming normal distribution of single game scores, which based on my tests proved accurate.] For the sake of sample size, I decided to limit this exercise to players who appeared is 58 or more games [2. The cutoff to be eligible for a scoring title] so that I felt confident that the yearā€™s sample captured a good idea of eachĀ playerā€™s general range of single-game DRE.

The idea is that the higher the absolute value of the z-score, the more unlikely of an event occurred. For example, Russell Westbrook has games with a DRE of 22.1 (2/6 vs New Orleans) and of -17.3 (1/7 vs Sacramento); however, his scores were so wildly variableĀ throughout the year Ā (i.e. had a high standard deviation) that his z-score for thoseĀ games arenā€™t in the overallĀ Ā top or bottom 5 for the season. Generally, high usage players are more prone to a wide range of performances, so the unlikely performances should belong to role players who took took on a rare high usage game rather than the usual stat-stuffing suspects.

Without further adieu, the most unlikely performances of the year:

The Good Unlikely:

Trevor Bookerā€“ 4/11/15 Jazz @Ā Portland ā€“ DRE: 19, Z-score: 4.64 (box)

TrevorBooker
TrevorBooker

Jared Dudley ā€“ 12/26/14 Milwaukee @Ā Atlanta ā€“ DRE: 18.9, Z-score: 4.20 (box)

JaredDudley
JaredDudley

Ersan Ilyasova ā€“ 3/26/15 Milwaukee vsĀ Indiana ā€“ DRE: 20, Z-score: 4.19 (box)

Ersan
Ersan

Cody Zeller ā€“ 1/31/15 Charlotte @Ā Denver ā€“ DRE: 15.4, Z-score: 4.13 (box)

CodyZeller
CodyZeller

Robert Sacre ā€“ 11/23/14 Ā LA Lakers vsĀ Denver ā€“ DRE: 9.8, Z-score: 4.07 (box)

SacreGood
SacreGood

The Bad Unlikely:

Robert Sacre ā€“ 12/5/14 LA Lakers @Ā Boston ā€“ DRE:-10.3 ,Z-score: -3.56 (box)

SacreBad
SacreBad

Amarā€™e Stoudemire ā€“ 11/26/14 New York @Ā Dallas ā€“ DRE: -10.5, Z-score: -3.34 (box)

Amare
Amare

Carlos Boozer ā€“ 10/29/14 LA Lakers @Ā Phoenix ā€“ DRE: -12.7, Z-score: -3.27 (box)

Boozer
Boozer

Hollis Thompson ā€“ 2/9/15 Philadelphia vsĀ Golden State ā€“ DRE: -15.3, Z-score: -3.27 (box)

HollisThompson
HollisThompson

Michael Carter-Williams ā€“ 12/27/14 Philadelphia @Ā Utah ā€“ DRE: -21, Z-score: -3.18 (box)

MCW
MCW

Some thoughts:

  • Itā€™s impressive that Sacre made both lists. This indicates how incredibly consistent his performances were outside of these two games. In fact, here is a graph of his games by DRE.
SacreGraph
SacreGraph
  • As expected, the good outliersĀ are mostly role players throwing up high usage games with incredible and unsustainable efficiency.
  • Turnovers are bad, while steals are in fact good. The top 5 performances had 1 turnover total between them, while the bottom 5 netted only one steal.
  • Ooph, MCWā€™s game was awful

Again, these results are not meant to be the best or worst game of the seasons but rather the most unlikely given the playerā€™s overall season. So if you want to argue Trevor Booker played the game of his life atĀ the Moda CenterĀ in early April 2015, you might not be far off.