Draftanomics
  • 2027 board
  • Consensus board
  • 2028 way-too-early
  • Draft news
  • Simulator
  • The house mock
  • Draft order
  • Trade calculator
  • The Scouting Slate
  • Sleepers & reaches
  • Movers & fallers
  • Player comps
  • Dynasty rookie board
  • Track record
  • Past drafts, re-graded
  • Analyst leaderboard
My rankingsAccount
Get the weekly →
Draftanomics

Honest NFL Draft analytics. The model's receipts against actual NFL teams live on the track record.

Boards
  • 2027 board
  • Consensus board
  • Sleepers
Tools
  • Mock draft simulator
  • Plays Like
  • Dynasty rookie board
Track Record
  • Year-by-year results
  • Accountability Index
  • Our board, on the record
  • Past drafts, re-graded
  • Methodology
  • Developers · free API
  • About
Account
  • My board
  • Weekly email
Data + sources
RAS by Kent Lee Platte·Consensus Big Board (Arif Hasan)·nflverse·PFR·CFBD·All sources & credits·Methodology
Career-value model, live
Privacy·Terms·Press ⌘K to search·Draftanomics2026
  1. Draftanomics/
  2. Prospects/
  3. 2016 board/
  4. Compare

Side-by-side comparison

Backtest

3 prospects · 2016 draft

Each spoke: percentile within the prospect's own position (WR)
RL
Ricardo Louis
WR · Auburn
22.1 yo · 6'2" · 215 lb
DR
Demarcus Robinson
WR · Florida
21.6 yo · 6'1" · 203 lb
MM
Malcolm Mitchell
WR · Georgia
22.8 yo · 6'0" · 198 lb

Headline metrics

The four numbers a GM checks first.

Ricardo Louis
Projected range
Rounds 4-6
half went picks 115-207
Bust probability
14%
Consensus rank
—
no boards yet
Career arc
arrives after the combineThe forecast needs his measurables first.
2nd-deal AAV
$3.9M-$6.0Mmodel
Demarcus Robinson
Projected range
Rounds 3-6
half went picks 91-192
Bust probability
9%
Consensus rank
#166
on 1 board
Career arc
arrives after the combineThe forecast needs his measurables first.
2nd-deal AAV
$3.4M-$5.2Mmodel
Malcolm Mitchell
Projected range
Rounds 4-6
half went picks 143-225
Bust probability
32%
Consensus rank
#158
on 11 boards
Career arc
arrives after the combineThe forecast needs his measurables first.
2nd-deal AAV
$3.5M-$5.4Mmodel

Feature radar

Each axis is the prospect's percentile among prospects at his position (0–100). Bigger shape = stronger across the board.

ProductionAthleticismAgeConsensusPer-play impact (PPA)Breakout ageWorkloadRAS × production
Prospects
  • Ricardo Louis· RB / WR / TE
  • Demarcus Robinson· RB / WR / TE
  • Malcolm Mitchell· RB / WR / TE
Axis leaders
  • ProductionRicardo Louis0
  • AthleticismRicardo Louis86
  • AgeDemarcus Robinson92
  • ConsensusRicardo Louis27
  • Per-play impact (PPA)Ricardo Louis50
  • Breakout ageRicardo Louis50
  • WorkloadRicardo Louis100
  • RAS × productionRicardo Louis0

Comp overlap

Top five historical comps per prospect. Shared comps highlighted.

No shared comps in the top five. Each prospect has a distinct historical profile.
Ricardo Louis
  • comps arrive after the combineBefore the combine there isn't enough measured data to match him confidently against past players. Comparisons draw on recent NFL careers (2021–2024) and fill in once combine numbers and game-by-game stats land.
Demarcus Robinson
  • comps arrive after the combineBefore the combine there isn't enough measured data to match him confidently against past players. Comparisons draw on recent NFL careers (2021–2024) and fill in once combine numbers and game-by-game stats land.
Malcolm Mitchell
  • comps arrive after the combineBefore the combine there isn't enough measured data to match him confidently against past players. Comparisons draw on recent NFL careers (2021–2024) and fill in once combine numbers and game-by-game stats land.

Risk and opportunity flags

Top three concerns and top two upside signals from the model.

Ricardo Louis
Risks
  • One-year wonder
    Production concentrated in one season: best 716 receiving yards vs 4-season mean 334. Limited multi-year tape — small sample of true production.
  • Below-cohort college production
    Position-cohort z-score on production composite is -1.26 — well below the skill average. Raw box-score output didn't keep pace with peers.
Opportunities
No standout upside flags.
Demarcus Robinson
Risks
  • Consensus skeptic
    1 board have him at consensus #166; model has him #116 — a 50-spot gap. When boards fade a player this hard, the model is usually the one missing something.
Opportunities
No standout upside flags.
Malcolm Mitchell
Risks
No major risks flagged.
Opportunities
  • Consensus says higher
    11 boards have him at consensus #158; model has him #208 — a 50-spot fade. Analysts converging this much above the model is a tell that the model is undervaluing him.

Career trajectory forecast

Projected share of his team's snaps across the first four NFL seasons. Solid = seasons already played, dashed = the forecast.

0%25%50%75%100%Y1Y2Y3Y4Snap share
Ricardo LouisDemarcus RobinsonMalcolm Mitchell· solid = actuals, dashed = forecast

Verdict

The model's plain-English read.

Ricardo Louis

Draftanomics projects Ricardo Louis (age 22.1 at draft) as a mid-round flyer at #147 (range #145-#149, tight projection). Bust probability sits at 14% — low. Production concentrated in one season: best 716 receiving yards vs 4-season mean 334. Limited multi-year tape — small sample of true production. The call is a hold: no public board ranks him, so the model's #147 stands alone — and with bust risk at 14%, let the room catch up before paying a premium.

Demarcus Robinson

Draftanomics projects Demarcus Robinson (age 21.6 at draft) as a mid-round flyer at #116 (range #114-#118, tight projection). Bust probability sits at 9% — low. The call is a conviction buy: the model has him 50 spots ahead of the room (board #116 vs consensus #166), and at a 9% bust probability the edge is worth paying for.

Malcolm Mitchell

Draftanomics projects Malcolm Mitchell (age 22.8 at draft) as a Day 3 dart throw at #208 (range #206-#210, tight projection). Bust probability sits at 32% — real but not disqualifying. The call is a genuine fade: the room's consensus prices him around #158 and the model won't take him before #208 — a 50-spot markdown. The 50-spot markdown rests on thin verified signal rather than a named flaw — the model hasn't seen enough yet to match the room's conviction.