Fantasy managers have an unusual definition of “safe.”
We take a 30-year-old running back coming off 413 touches, stare at his enormous projection and say, Well, what could go wrong?
History has submitted several documents.
Because something strange happens around pick 4.
From 2016 through 2025, the first three selections in fantasy drafts posted severe bust rates of 23.3%, 20% and 20%.
Pick 4?
12%.
Pick 5?
8%.
Apparently safety exists in fantasy football. We have just been buying it in the wrong aisle.
And in 2026, that creates a fascinating decision at 1.04: do you preserve the historical advantage of this slot with Puka Nacua or Jaxon Smith-Njigba, or trade some of it back for Christian McCaffrey's absurd ceiling?
Pick 4 is where the first round changes
Here are the historical results around the top of Round 1:
- Pick 1 — 18.5 median PPG, 63.3% top-12 rate, 23.3% severe bust rate
- Pick 2 — 18.8 median PPG, 62.5% top-12 rate, 20.0% severe bust rate
- Pick 3 — 18.5 median PPG, 60.0% top-12 rate, 20.0% severe bust rate
- Pick 4 — 18.5 median PPG, 66.0% top-12 rate, 12.0% severe bust rate
- Pick 5 — 18.3 median PPG, 66.0% top-12 rate, 8.0% severe bust rate
Pick 4 did not produce more fantasy points than the players taken ahead of it.
It simply exploded less often.
That distinction matters.
The 1.04 delivered a 66% top-12 positional finish rate, tied for best in the first round, while cutting the bust rate almost in half versus 1.01.
Why?
Position is part of it.
Only 58% of historical 1.04 selections were running backs, compared with 70% at pick 1 and 72% at pick 2.
Across the entire first round, RBs produced a 24.6% severe bust rate.
Wide receivers?
8.3%.
Receivers also slightly beat running backs in both median scoring and top-12 hit rate.
Fantasy managers pay enormous prices for running back scarcity.
Sometimes what we are actually purchasing is anxiety with a rushing-attempt projection attached.
And the entire draft slot has been excellent
Pick 4 owns selections:
4, 21, 28, 45, 52 and 69.
Collectively, those six historical draft neighborhoods produced:
- 89.8 median PPG
- 2.62 expected top-12 finishes
- 1.92 expected severe busts
That ranked second-best among all 12 draft positions in both scoring and avoiding busts. Only pick 5 was better.
That matters because fantasy managers dramatically overrate the importance of their first selection.
Your first-rounder is one player.
The first six rounds are your roster.
Pick 4 has historically been good at both.
So who should you take at 1.04 in 2026?
| Your pick | Likely gone | 50/50 zone | Likely available |
|---|---|---|---|
| 1.04 | Jahmyr GibbsRB · 27% | Bijan RobinsonRB · 34% Ja'Marr ChaseWR · 42% Puka NacuaWR · 50% | Amon-Ra St. BrownWR · 73% Jonathan TaylorRB · 79% CeeDee LambWR · 88% |
| 2.09 | Drake LondonWR · 21% | Brock BowersTE · 37% Derrick HenryRB · 36% A.J. BrownWR · 43% | Nico CollinsWR · 76% Josh AllenQB · 71% Jeremiyah LoveRB · 85% |
| 3.04 | Nico CollinsWR · 24% Rashee RiceWR · 25% | Josh AllenQB · 36% Jeremiyah LoveRB · 36% Chris OlaveWR · 50% | Breece HallRB · 84% Javonte WilliamsRB · 72% Zay FlowersWR · 83% |
| 4.09 | Cam SkatteboRB · 21% Travis EtienneRB · 28% | Colston LovelandTE · 37% Ladd McConkeyWR · 32% Jaylen WaddleWR · 45% | David MontgomeryRB · 73% |
| 5.04 | Terry McLaurinWR · 25% Davante AdamsWR · 26% Bucky IrvingRB · 29% | Lamar JacksonQB · 44% David MontgomeryRB · 50% D'Andre SwiftRB · 36% | Tyler WarrenTE · 74% |
| 6.09 | Mike EvansWR · 21% Carnell TateWR · 27% | Christian WatsonWR · 31% Jayden DanielsQB · 40% Jalen HurtsQB · 48% | — |
Availability uses the same model as the app's Draft Plan: under 30% likely gone, 30–70% a coin flip, 71%+ likely available. Prices move; open the draft room for the live read.
The current market has formed a pretty clear top three.
CBS PPR ADP as of August 21:
- Jahmyr Gibbs — 1.30
- Bijan Robinson — 1.92
- Ja'Marr Chase — 3.98
- Puka Nacua — 4.91
- Jonathan Taylor — 5.29
- Jaxon Smith-Njigba — 6.98
- Christian McCaffrey — 7.20
Rotoworld is even cleaner: Gibbs, Bijan, Chase, Puka, JSN, McCaffrey.
If one of the top three falls?
Take him and enjoy whatever minor act of incompetence occurred directly in front of you.
Otherwise, Puka is my pick.
He caught 129 passes for 1,715 yards and 10 touchdowns last season, then added 105 rushing yards and another score.
JSN is not far behind after his own completely unreasonable 2025: 119 receptions, 1,793 yards and 10 touchdowns.
The AutoDraftYears model prefers Puka slightly:
- Puka: 76.8 Upside / 35.0 Risk
- JSN: 71.9 Upside / 34.3 Risk
- McCaffrey: 88.9 Upside / 65.0 Risk
And there is the decision.
McCaffrey has the highest upside score on the entire board.
He also carries roughly double the risk of the receivers.
McCaffrey is the fun answer. Puka is the one I'd make.
It would be easier to dismiss McCaffrey if he had stopped being ridiculous.
Unfortunately, he has not.
Last season he rushed 311 times for 1,202 yards, then caught 102 passes for 924 more. Seventeen total touchdowns. 413 touches.
That is not a workload.
That is an OSHA investigation.
He also turns 30 this season and has recently been held out of practice with soreness.
Puka isn't completely clean either. He is currently dealing with psoas soreness, and Sean McVay said this week that Los Angeles is being cautious and wants him fully ready before returning to practice.
So this is not “risky player versus risk-free player.”
Those do not exist.
It is 35 Risk versus 65 Risk in the AutoDraftYears model.
Given that pick 4's historical superpower has been avoiding first-round disasters, I wouldn't voluntarily surrender that advantage.
Give me Puka.
Pick 21 might be the real prize
This is where drafting fourth becomes more interesting.
Your second selection comes at 21.
Historically, pick 21 produced 16.4 median PPG and a 56% top-12 rate.
By pick 24, those numbers had fallen to 15.1 PPG and 50%.
Three picks does not sound dramatic.
Draft tiers do not care.
And the current board shows exactly why.
Rotoworld ranks A.J. Brown 21st, Malik Nabers 22nd, Rashee Rice 23rd, George Pickens 24th, Omarion Hampton 25th and Trey McBride 26th.
That is a legitimate tier.
The AutoDraftYears availability model currently gives Nico Collins a 76% chance of making pick 21, George Pickens 66%, Trey McBride 50%, A.J. Brown 43%, Brock Bowers 37% and Derrick Henry 36%.
If I start Puka and one of the premium running backs falls, great.
If Collins or another elite receiver is clearly best?
I am perfectly comfortable starting WR-WR.
Do not draft a worse running back because your roster page looks asymmetrical after two rounds.
It will survive.
Picks 21 and 28 are a fantastic pair
I also prefer this setup to drafting directly at the turn.
At 21, you make a pick.
Six teams make decisions.
Then you return at 28.
That matters because the board has genuinely changed.
You can take the player you prefer at 21 without trying to predict the next pick simultaneously.
At 28, current rankings put players such as McBride, Chris Olave, Travis Etienne, Kyren Williams, Josh Jacobs and Josh Allen in the neighborhood.
My preference?
Keep attacking RB/WR unless McBride represents a clear tier advantage.
I would not take Josh Allen here.
The AutoDraftYears historical quarterback data continues to scream patience: Round 4 QBs produced 20.7 median PPG, while Round 6 QBs produced 18.7 PPG with a 78.6% top-12 rate. Nearly 60% of QB1 seasons came after Round 6.
Josh Allen is wonderful.
So are beachfront hotels.
Price still matters.
The mistake I would avoid from this slot
Starting WR creates a predictable emotional response.
I need running backs now.
That is where things get dangerous.
Historical severe bust rates for running backs:
- Round 3: 43.2%
- Round 4: 51.2%
- Round 5: 65.7%
- Round 6: 62.5%
Comparable receiver bust rates in Rounds 3 and 4 were just 26.9% and 33.9%.
So if a good RB reaches 21 or 28, take him.
What I would not do is spend picks 45 and 52 forcing running back simply because you started WR-heavy.
Pick 4's entire historical advantage is avoiding unnecessary failure.
There is no reason to spend four rounds carefully preserving it and then wander directly into the RB dead zone carrying a shopping list.
How I would draft from pick 4
Take Puka Nacua at 1.04.
If the psoas issue meaningfully worsens before your draft, JSN becomes an extremely easy pivot.
Let somebody else take McCaffrey's enormous ceiling along with the age, workload and AutoDraftYears 65 Risk Score.
Then attack pick 21.
That selection comes just before both our historical Round 2 results and the current 2026 board begin getting noticeably thinner.
At 28, stay flexible.
Do not force RB because you started WR. Do not force quarterback because Josh Allen is sitting there looking handsome in the draft app.
The surprising thing about pick 4 is that it doesn't require heroics.
Historically, this slot worked because fantasy managers stopped paying quite as much for certainty and accidentally ended up with more of it.
Sometimes the smartest thing you can do with a good draft position is avoid trying to improve it.
Sources: Pro Football Reference — fantasy finishes; FantasyPros — PPR Overall ADP; Auto Draft Years — 2026 Guide to the Fantasy Football Markets. Sample: 6,434 player-seasons, 2016–2025; per-pick figures pool players drafted within ±2 ADP ranks of that pick (n≈50 per pick). Outcome definitions match the 2026 Guide to the Markets. Availability uses the same model as the app's Draft Plan. League assumed: 12-team, full PPR, 1QB/2RB/3WR/1TE/1FLEX. For informational and entertainment purposes only.