Skip to content
Joshua Kelley

#27Joshua Kelley

RB·UCLA·201920182.0 pts line valueDay 3 (Rds 4–7)

Joshua Kelley is a 2-year Featured Back for UCLA. Joshua's 2019 season ranks in the 13th percentile nationally by opponent-adjusted EPA per play across 227 plays — a developing rate for the RB.

What projects, and what doesn't · RBs · held out 2019-2025

How well one season predicts the next, measured on seasons the model never trained on. 1.00 would be perfectly predictable; 0.00 means last year told us nothing.

  • Usage share (volume)0.52
    Beats guessing the RB average by 18%. n=4,219 RB seasons
  • EPA per play (efficiency)0.10
    Not projectable — we do not forecast this. n=3,575 RB seasons
  • Total EPA (value)0.41
    Beats guessing the RB average by 9%. n=3,575 RB seasons

Across all positions, usage carries year to year at 0.47 and efficiency at 0.07. So Joshua Kelley's projection is a projection of opportunity — how much of the offense he runs through. How well he converts it is something this model does not claim to know a year in advance, and the number above is why.

2019 Production

Rushing
1060 Rush yards12 Rush TD229 Carries4.6 Yards/carry
Receiving
11 Receptions71 Rec yards1 Rec TD6.5 Yards/rec

Performance Analysis · 2019 · vs RB peers

  • Efficiency13
  • Volume82
  • Explosiveness32
  • Consistency65
  • Receiving22
Player type
Featured Back

The centerpiece of the run game — high carry volume, used in early downs and goal-line, true workhorse role.

Primary ball carrierHigh volumeGoal-line threat
Peer percentiles
Opponent-adjusted EPA (WEPA/play)13th %ile · below avg
Game-to-game consistency65th %ile · average
Key findings
  • Below-average efficiency vs RB peers — value comes through volume, not per-play impact.
  • High-volume role — one of the team's most-used RBs by play share.
  • Particularly dangerous on passing downs — efficiency spikes in obvious pass situations.

Historical Comparables · RB · efficiency + volume + value

Players from 2013–2025 matched on EPA efficiency, play volume, and adjusted value tier — not just one metric.

PlayerTeamWEPA/playLine valTotal EPA
Saquon BarkleyPenn State0.3403.997.9
Dalvin CookFlorida State0.3404.399.3
Breece HallIowa State0.3404.5102.3
Hassan HaskinsMichigan0.3804.8106.4
James FlandersTulsa0.3904.9103.7

Comps are statistical — efficiency, volume, and value tier all factor in. Style and conference context differ.

Game Log · box score + EPA, week by week

+0.2900.29Wk 2 vs San Diego State: +0.10 EPA/play2Wk 3 vs Oklahoma: -0.22 EPA/play3Wk 4 vs Washington State: +0.13 EPA/play4Wk 5 vs Arizona: +0.23 EPA/play5Wk 6 vs Oregon State: +0.21 EPA/play6Wk 8 vs Stanford: +0.29 EPA/play8Wk 9 vs Arizona State: +0.25 EPA/play9Wk 10 vs Colorado: +0.15 EPA/play10Wk 12 vs Utah: +0.02 EPA/play12Wk 13 vs USC: +0.03 EPA/play13Wk 14 vs California: +0.10 EPA/play14
EPA per play · x-axis: weekabove 0 = added points · below = lost
WkOpponentResultOpp SP+CarRush YdsAvgRush TDRecRec YdsRec TDEPA/play
2vsSan Diego StateL14-232.015533.512200.10
3vsOklahomaL14-4822.918512.80-0.22
4@Washington StateW67-6311.120904.5021910.13
5@ArizonaL17-20-8.0271274.711900.23
6vsOregon StateL31-48-0.721743.511400.21
8@StanfordW34-16-4.7181769.8121300.29
9vsArizona StateW42-326.3341644.840.25
10vsColoradoW31-14-4.1231265.520.15
12@UtahL3-4919.519784.1011200.02
13@USCL35-5213.115453.011900.03
14vsCaliforniaL18-28-2.619764.011300.10

Usage & Situational · Pro

Snap-share proxy
Overall
28.6%
Passing plays
3.5%
Rushing plays
54.2%
Standard downs
32.9%
Passing downs
19.6%
EPA by down type
Standard downs
0.13
Passing downs
0.28
Pass / Rush EPA
0.13 / 0.16

Usage = share of team plays (CFBD has no true snap counts).

Career · rolling EPA, game by game

Per-game EPA5-game avg
+0.780−0.312019
EPA per play · 2018 — 2019 · 22 games
SeasonTeamLine valueTotal EPA
2018UCLA
4.4
85.7
2019UCLA
2.0
58.8

Chart shows per-game EPA (bars) and rolling 5-game average (line). Season breaks marked with dashed lines. Line value = est. points over replacement per game.

EPA = expected points added (opponent-adjusted). NIL estimates are model-based ranges, not reported deals. Data: CollegeFootballData. Not betting advice.