Two players, one answer.
MarShawn Lloyd vs Savion Williams
Any league · this week start/sit · Half PPR
Stats through week 8 of 2026
This week
No call this week: one of them has no game or a projection under 4 points
Rest of season
MarShawn Lloyd by 38% (Redraft half PPR)Dynasty
Savion Williams by 5% (Dynasty Superflex)
Week 9 · Half PPR · 52% held up in 2025
A toss-up
On Sleeper's projections, Savion Williams is projected 0.4 points ahead of MarShawn Lloyd. In the 2025 season, a player projected that far ahead outscored the other 52% of the time. That's close to a coin flip: on these projections neither is the better start, and an injury report or a projection update before kickoff can tip it.
WR · GB
Savion Williams
4.4 projected
@ NE · Sun 4:25 PM ET
5.2 a game this season · last 3: 6.1, 6.6, 2.8
RB · GB
MarShawn Lloyd
4.0 projected
@ NE · Sun 4:25 PM ET
5.7 a game this season · last 3: 1, 4.5, 2.1
What the call rests on
- Savion Williams: Sleeper's projection, last changed Tue, Nov 3, 10:40 PM ET
- MarShawn Lloyd: Sleeper's projection, last changed Tue, Nov 3, 10:40 PM ET
- Sleeper's projections read Tue, Nov 3, 10:42 PM ET.
- Injury statuses read Tue, Nov 3, 10:42 PM ET, with the projections (hourly). A fresh projection time doesn't mean every status is: game-day inactives, about 90 minutes before kickoff, can come after it.
| Same projections, scored | Williams | Lloyd |
|---|---|---|
| PPR | 5.6 | 4.5 |
| Half PPR | 4.4 | 4 |
| Standard | 3.2 | 3.4 |
What could change it before kickoff
- A projection update of 0.4 points or more in the other direction flips the call; Sleeper keeps adjusting its projections until kickoff, and this page reads them again at most an hour after it last did.
- In standard scoring the call goes the other way: MarShawn Lloyd 3.4, Savion Williams 3.2.
Open in Who should I start (your league, more players)
This week’s projections are Sleeper’s, scored with Half PPR settings. The percentage is how often a projected gap like this held up across the 2025 season; checked against 2026 games (not used to fit it), the stated chances so far run a few points high, so read a 60% call as nearer 55%.