LMD Rankings review · 2026
How our Week 4 rankings did
2026 season · Half PPR · Board published October 1, before kickoff
LMD vs. Expert Consensus vs. the previous-week results baseline
| Week 4 Half PPR | LMD | Expert Consensus | Previous-week results baseline |
|---|---|---|---|
| Start/sit calls correct | 62% | 62% | 50% |
| QB | 59% | 58% | 42% |
| RB | 67% | 66% | 62% |
| WR | 61% | 62% | 47% |
| TE | 53% | 53% | 36% |
| Starter picks correct | 54% | 50% | 47% |
| Typical rank miss | 10 spots | 11 spots | 13 spots |
| Average rank miss | 16.2 | 18.5 | 19.8 |
LMD was competitive with Expert Consensus this week. On the same 2,046 start/sit pairs, LMD put the higher scorer first 61.7% of the time, Expert Consensus 61.9% and the previous-week results baseline 50.0%. The previous-week results baseline ranks each position by the points players scored in Week 3, with no model and no adjustment, and grades that order on Week 4. It is not LMD's Week 3 rankings or the Week 3 review's figures. One week cannot separate two rankings this close. DeVonta Smith (WR12) and Justin Jefferson (WR15) were ruled out after the board's cutoff and still count as misses in every figure here.
Both lists were published before the review's cutoff. The Expert Consensus list was published about seven hours before LMD's board (October 1, 20:25 UTC) and may not reflect updates during that interval. Every section below grades LMD's whole board. One week is a small sample.
Expert Consensus: Half PPR lists for QB, RB, WR and TE, published October 1, 13:01 UTC. Expert Consensus source: borischen.co.
Top finish picks
| LMD | Expert Consensus | Previous-week results baseline | |
|---|---|---|---|
| Top 3 picks correct | 8% | 17% | 8% |
| Top 12 picks correct | 42% | 38% | 35% |
| Top 24 picks correct | 63% | 61% | 55% |
By position
| Starter picks | LMD | Expert Consensus | Previous-week results baseline |
|---|---|---|---|
| QB | 8/12 | 7/12 | 5/12 |
| RB | 17/24 | 15/24 | 16/24 |
| WR | 9/24 | 9/24 | 8/24 |
| TE | 5/12 | 5/12 | 5/12 |
| Typical miss | LMD | Expert Consensus | Previous-week results baseline |
|---|---|---|---|
| QB | 6 | 7 | 7 |
| RB | 11.5 | 12 | 11 |
| WR | 16.5 | 22 | 30 |
| TE | 5 | 7 | 7 |
Hits and misses
Ranked → finished
Hits
- Puka NacuaWR5 → WR5
- Harold Fannin Jr.TE10 → TE10
- Tony PollardRB24 → RB24
Misses
- Jaxon Smith-NjigbaWR1 → WR35
- Cam SkatteboRB13 → RB35
- Jeremiyah LoveRB12 → RB41
Limitations
At 2 clubs the quarterback who threw the most passes was not on the board at all: Athan Kaliakmanis (WAS, QB25) and Tyson Bagent (CHI, QB29).
Probability accuracy
Top-12 Brier score. The equal-chance baseline gives every player at a position the same probability.
- LMD
- 0.085
- Equal-chance baseline
- 0.097
Brier score measures how accurate probability forecasts are. 0 is perfect; lower is better.
Methodology & detailed stats
We graded the Week 4 board we published on October 1, 2026, the last one generated before the first game kicked off, against what happened. Nothing was regenerated, re-ranked or adjusted after the games, and the misses are listed the same way as the hits.
The comparison at the top of this page grades LMD, Expert Consensus and the previous-week results baseline on the same players. Every other figure grades LMD's whole board on its own.
Starter picks
A starter call is a player we ranked inside the starters a 12-team league fields at each position, not counting flex spots: 12 quarterbacks, 24 running backs, 24 wide receivers, 12 tight ends. It is a hit when he finished inside the same line at his position. A player we ranked as a starter who did not play counts as a miss, because the board's rankings already account for the chance a player sits.
39 of 72 players we ranked as starters finished as starters (54%).
Of the 72 players who actually finished as starters, we had ranked 39 inside the line (54%). That is the same count read the other way: we rank exactly as many starters as the league fields, and exactly that many finish there unless players tie.
4 players we ranked as starters recorded no statistic and count as misses. DeVonta Smith and Justin Jefferson were ruled out in a status capture taken after the board went up, which a board published before the week cannot see: check the current board before lineups lock.
| Position | Starter line | Finished as starters | Hit rate | Actual starters | Capture rate |
|---|---|---|---|---|---|
| QB | Top 12 | 8 of 12 | 67% | 12 | 67% |
| RB | Top 24 | 17 of 24 | 71% | 24 | 71% |
| WR | Top 24 | 9 of 24 | 38% | 24 | 38% |
| TE | Top 12 | 5 of 12 | 42% | 12 | 42% |
| All four | By position | 39 of 72 | 54% | 72 | 54% |
Hit rate: of the players we ranked as starters, the share who finished as starters. Capture rate: of the players who finished as starters, the share we had ranked as starters, including players we did not rank at all.
Start/sit calls
For each position, the pool is every player LMD ranked near the starter line (QB and TE ranked 7 to 18 and RB and WR ranked 13 to 36), and every pair in it is compared once: a call is right when the player ranked higher scored more. A player who did not play scored zero.
LMD put the higher scorer first in 376 of 671 scored pairs (56%). Picking at random would be right half the time.
| Position | Ranks compared | Players | Pairs scored | LMD right | LMD rate |
|---|---|---|---|---|---|
| QB | 7-18 | 12 | 65 | 53 | 82% |
| RB | 13-36 | 24 | 275 | 172 | 63% |
| WR | 13-36 | 24 | 266 | 114 | 43% |
| TE | 7-18 | 12 | 65 | 37 | 57% |
| All four | By position | 72 | 671 | 376 | 56% |
684 pairs in all. Left out for every ranking: 13 pairs whose players scored exactly the same and 0 pairs LMD ranked equally.
Rank misses
How many places each player finished from where we ranked him, for the players we ranked as starters who recorded a stat line. A receiver we ranked WR8 who finished WR14 is 6 places off.
Across 68 players, the average miss was 16.2 places and the median 10. A few large misses pull the average up; the median is the typical player.
| Position | Players | Average miss | Median miss |
|---|---|---|---|
| QB | 12 | 6.3 | 6 |
| RB | 24 | 13.1 | 11.5 |
| WR | 20 | 27.5 | 16.5 |
| TE | 12 | 13.5 | 5 |
| All four | 68 | 16.2 | 10 |
A player with no stat line has no finish, so he is left out here and counted as a miss in the starter calls above.
The same measure over the players ranked inside the top 24 at their position who recorded a stat line, as earlier reviews reported it. A receiver ranked WR8 who finished WR14 is a miss of 6 places.
| Position | Ranked top 24 | With a stat line | Median miss | Within 6 | Rank corr. (top 24) | Rank corr. (whole board) |
|---|---|---|---|---|---|---|
| QB | 24 | 23 | 8 | 48% | 0.38 | 0.42 |
| RB | 24 | 24 | 11.5 | 25% | 0.43 | 0.71 |
| WR | 24 | 20 | 16.5 | 20% | 0.10 | 0.54 |
| TE | 24 | 24 | 7.5 | 50% | 0.09 | 0.72 |
| All four | 96 | 91 | 10 | 36% | -- | -- |
Rank correlation is Spearman's: 1 is the right order, 0 is no relationship. Inside the top 24 it is a hard test, because those players are close together; over the whole board it is an easier one, because it includes backups who were never near a start. It is not pooled across positions, which would correlate four different orderings.
Top 3, 12 and 24 picks
The same count at every finish line the board prints: of the players we ranked inside the top 3, 12 or 24 at their position, how many finished there, all four positions together. The chances printed beside each rank are graded as probabilities further down.
| Threshold | Finished inside | Hit rate | Actual finishers | Capture rate |
|---|---|---|---|---|
| Top 3 | 1 of 12 | 8% | 12 | 8% |
| Top 12 | 20 of 48 | 42% | 48 | 42% |
| Top 24 | 60 of 96 | 63% | 96 | 63% |
Top 24 is a wide net at quarterback and tight end, where a 12-team league starts 12, so the starter calls above are the better read there.
| Position | Finished inside | Hit rate | Actual finishers | Capture rate |
|---|---|---|---|---|
| QB | 0 of 3 | 0% | 3 | 0% |
| RB | 0 of 3 | 0% | 3 | 0% |
| WR | 0 of 3 | 0% | 3 | 0% |
| TE | 1 of 3 | 33% | 3 | 33% |
| All four | 1 of 12 | 8% | 12 | 8% |
| Position | Finished inside | Hit rate | Actual finishers | Capture rate |
|---|---|---|---|---|
| QB | 8 of 12 | 67% | 12 | 67% |
| RB | 3 of 12 | 25% | 12 | 25% |
| WR | 4 of 12 | 33% | 12 | 33% |
| TE | 5 of 12 | 42% | 12 | 42% |
| All four | 20 of 48 | 42% | 48 | 42% |
| Position | Finished inside | Hit rate | Actual finishers | Capture rate |
|---|---|---|---|---|
| QB | 18 of 24 | 75% | 24 | 75% |
| RB | 17 of 24 | 71% | 24 | 71% |
| WR | 9 of 24 | 38% | 24 | 38% |
| TE | 16 of 24 | 67% | 24 | 67% |
| All four | 60 of 96 | 63% | 96 | 63% |
Every hit and miss list
Hits
Ranked inside the starter line (Top 12 at QB and TE, Top 24 at RB and WR) and finished inside it, closest to the rank first.
- WR Puka NacuaLAR Ranked WR5 · WR5 · 23.2 pts · Top 24: 51%
- TE Harold Fannin Jr.CLE Ranked TE10 · TE10 · 10.2 pts · Top 12: 36%
- RB Tony PollardTEN Ranked RB24 · RB24 · 12.0 pts · Top 24: 41%
- TE Brock BowersLV Ranked TE2 · TE3 · 17.6 pts · Top 12: 50%
- TE George KittleSF Ranked TE3 · TE4 · 15.0 pts · Top 12: 46%
Misses: ranked as starters, finished outside
Ranked inside the starter line and finished outside it or recorded no statistic, the ones we gave the best chance first. Players ruled out in a later status capture are listed separately below.
- WR Jaxon Smith-NjigbaSEA Ranked WR1 · WR35 · 10.1 pts · Top 24: 68%
- RB Cam SkatteboNYG Ranked RB13 · RB35 · 7.1 pts · Top 24: 65%
- RB Jeremiyah LoveARI Ranked RB12 · RB41 · 6.0 pts · Top 24: 64%
- WR Amon-Ra St. BrownDET Ranked WR3 · WR28 · 11.5 pts · Top 24: 61%
- RB D'Andre SwiftCHI Ranked RB10 · RB38 · 6.4 pts · Top 24: 61%
Misses: finished as starters, ranked outside
Ranked outside the starter line and finished inside it, the ones we gave the smallest chance first.
- WR Darius CooperPHI Ranked WR173 · WR12 · 16.8 pts · Top 24: 0%
- WR Jahan DotsonATL Ranked WR102 · WR21 · 12.4 pts · Top 24: 4%
- WR Tyquan ThorntonKC Ranked WR92 · WR4 · 25.6 pts · Top 24: 5%
- WR Keon ColemanBUF Ranked WR93 · WR9 · 20.6 pts · Top 24: 6%
- WR Isaiah WilliamsNYJ Ranked WR76 · WR17 · 14.3 pts · Top 24: 7%
Ruled out after the board was published
Ranked inside the starter line, then ruled out in a status capture taken after the board went up. Listed apart from the misses because the status evidence the board was built from did not rule him out.
- WR DeVonta SmithPHI Ranked WR12 · Did not play (ruled out) · -- · Top 24: 44%
- WR Justin JeffersonMIN Ranked WR15 · Did not play (ruled out) · -- · Top 24: 41%
Finished as starters without being ranked
- Ian Thomas (LV): TE10, 10.2 pts
Probability detail
Beside each rank the board prints a chance of finishing in the top 3, 12 and 24 at his position. If those chances are right, a group of players each given 60% should get there about 60% of the time, and the total number of finishers should land near the sum of the chances. They are graded as probabilities, not as yes-or-no calls at 50%.
| Threshold | Expected finishers | Actual | Typical range | Brier | Same % for all | Rank as yes/no |
|---|---|---|---|---|---|---|
| Top 3 | 11.9 | 12 | 5-18 | 0.027 | 0.029 | 0.055 |
| Top 12 | 47.6 | 47 | 36-59 | 0.085 | 0.097 | 0.138 |
| Top 24 | 94.3 | 95 | 80-108 | 0.111 | 0.156 | 0.177 |
Typical range is the expected count plus or minus two standard deviations of what the chances imply. Brier is the mean squared gap between each chance and what happened; lower is better. "Same % for all" gives every player at a position the same chance, and "rank as yes/no" says 100% inside the threshold and 0% outside: two simple forecasts the chances should beat. Groups under 10 players are shown faded: too few to read.
- The number of Top 3, Top 12 and Top 24 finishes landed inside the range the percentages implied (12 against 11.9, 47 against 47.6 and 95 against 94.3).
- At every threshold the percentages scored better than reading the rank as a yes-or-no call, so the odds columns said something the rank alone did not.
- One position at a time, the percentages lost to a simple forecast at Top 3 QB (0.086 against 0.085 for the same chance for all). The pooled figures above do not show it.
- No group of 20 or more players finished outside the range chance allows for the percentages it was given.
- Of the 58 players given a 30-50% chance of a Top 12 finish, 34% got there. A percentage on the board is a rate across many players like him, not a call on one.
- 104 of the 400 ranked players recorded no statistic. Their chances are part of every percentage above, which is why the whole board is graded rather than only the players who played.
- Of the Top 24 finishes the board expected, players who recorded no statistic were given 3.9 of 23.9 at wide receiver (56 players) and 3.8 of 22.9 at tight end (28 players). Chances spread onto players who did not take the field are chances the starters' percentages did not get, so read a small percentage beside a backup's name with that in mind.
| Threshold | Expected finishers | Actual | Brier | Same % for all | Rank as yes/no |
|---|---|---|---|---|---|
| Top 3 QB | 3.0 | 3 | 0.086 | 0.085 | 0.188 |
| Top 3 RB | 3.0 | 3 | 0.028 | 0.030 | 0.062 |
| Top 3 WR | 3.0 | 3 | 0.015 | 0.017 | 0.034 |
| Top 3 TE | 3.0 | 3 | 0.027 | 0.031 | 0.042 |
| Top 12 QB | 11.8 | 12 | 0.211 | 0.234 | 0.250 |
| Top 12 RB | 12.0 | 12 | 0.096 | 0.108 | 0.186 |
| Top 12 WR | 12.0 | 12 | 0.059 | 0.064 | 0.091 |
| Top 12 TE | 11.8 | 11 | 0.079 | 0.102 | 0.137 |
| Top 24 QB | 23.6 | 24 | 0.185 | 0.188 | 0.375 |
| Top 24 RB | 23.8 | 24 | 0.096 | 0.186 | 0.144 |
| Top 24 WR | 23.9 | 24 | 0.104 | 0.118 | 0.171 |
| Top 24 TE | 22.9 | 23 | 0.114 | 0.184 | 0.158 |
| Board's chance | Players | Average chance | Actually finished there | 95% range |
|---|---|---|---|---|
| 0-5% | 315 | 1% | 0% | 0%-1% |
| 5-15% | 71 | 9% | 14% | 8%-24% |
| 15-30% | 14 | 20% | 14% | 4%-40% |
| Board's chance | Players | Average chance | Actually finished there | 95% range |
|---|---|---|---|---|
| 0-5% | 212 | 1% | 1% | 0%-4% |
| 5-15% | 69 | 9% | 10% | 5%-19% |
| 15-30% | 52 | 21% | 25% | 15%-38% |
| 30-50% | 58 | 38% | 34% | 24%-47% |
| 50-70% | 9 | 56% | 44% | 19%-73% |
| Board's chance | Players | Average chance | Actually finished there | 95% range |
|---|---|---|---|---|
| 0-5% | 108 | 2% | 2% | 1%-7% |
| 5-15% | 111 | 9% | 5% | 3%-11% |
| 15-30% | 57 | 21% | 23% | 14%-35% |
| 30-50% | 49 | 39% | 45% | 32%-59% |
| 50-70% | 42 | 60% | 57% | 42%-71% |
| 70-85% | 32 | 78% | 84% | 68%-93% |
| 85-100% | 1 | 88% | 100% | 21%-100% |
Were firmer calls more reliable?
No confidence rating was published with the Week 4 board, so there is none to grade. What the board did publish is how one-sided each player's chance was. Here the players ranked in the top 24 at each position are grouped by their Top 12 chance: the further from an even split, the firmer the call. The median rank miss counts only those who recorded a stat line.
| Top 12 chance | Players | Expected on the favoured side | Actually on it | Median rank miss |
|---|---|---|---|---|
| 35-65% | 50 | 59% | 64% | 9 |
| 15-35% or 65-85% | 46 | 74% | 67% | 12 |
| Under 15% or over 85% | 0 | -- | -- | -- |
In every group large enough to read, the share landing on the favoured side was within the range its percentages implied. Firmer calls missed by more places (median 12 against 9), which in one week is not evidence either way.
Other scoring formats
| Scoring | Starter calls | Start/sit calls | Median rank error (starters) | Top 12 Brier |
|---|---|---|---|---|
| Standard | 49% | 55% | 9.5 | 0.085 |
| Half PPR | 54% | 56% | 10 | 0.085 |
| Full PPR | 58% | 57% | 11 | 0.087 |
How this review was graded
The comparison. LMD, Expert Consensus and the previous-week results baseline are graded on the same players, finishes and start/sit pairs, by the same calculations as the rest of this page. No game kicked off before the Expert Consensus lists, so every player is graded. Every rank is the one the list printed. The start/sit pool is every player any of the three put near the starter line; a pair is scored for all three or for none, and a player a list did not rank counts as ranked below everyone it did. 2,046 of 2,156 pairs were scored: 33 pairs scored the same, and 78 pairs were tied on at least one list.
Expert Consensus. The Half PPR tier lists for QB, RB, WR and TE, published October 1, 13:01 UTC. A player's rank is his place in the list, across tiers, in the last list published no later than LMD's board was generated. That list can be hours older than the board, so it may not reflect news LMD's board had. 151 listed players were matched to our catalog by name and position; all of them were matched.
Previous-week results baseline. Each position ranked by the points players actually scored in Week 3, Half PPR, from the same statistics as the rest of this page. Nothing is adjusted for injuries, matchups or roles. A player with no stat line that week is not on the list, and players who scored the same share a rank. Matthew Hibner (TE15) is left off the list because our catalog does not hold him, so his Week 4 result cannot be read. Nobody else's rank changes.
The board. Five boards were archived for Week 4. This review grades the final pre-kickoff v1.4.1 board. It was generated 3.8 hours before the first kickoff (October 2, 00:15 UTC), and it is the board readers saw going into the week. The October 4 refresh was generated after games had started and is not graded here.
The population. All 400 players the board ranked at QB, RB, WR and TE. A finish is counted over every player in the league who recorded a statistic (362 stat lines, including 66 players the board did not rank), priced with the same Half PPR scoring the board projected. Players who tie share the better place.
Starter calls. The starter line is Top 12 at quarterback and tight end and Top 24 at running back and receiver. The hit rate counts the players ranked inside it who finished inside it; the capture rate counts the players who finished inside it, ranked by us or not, whom we had ranked inside it. Top 3, 12 and 24 use the same arithmetic.
Start/sit calls. The pool is every player LMD ranked inside half a starter line either side of it (QB and TE ranked 7 to 18 and RB and WR ranked 13 to 36), and every pair in it is formed once. A pair is scored only when the two players' points differ and LMD put them in a strict order. The band is set by the starter line, not chosen after the games.
Rank error. The number of places between a player's published positional rank and his finish, over the players ranked inside the starter line who recorded a stat line. Earlier reviews reported it over the top 24 at every position, and that table is kept under the rank error section.
Players who did not play. 104 ranked players recorded no statistic. 23 were ruled out in a status capture taken after the board was published (October 4, 11:05 UTC) and 81 had no ruling. A player we ranked as a starter who did not play is a missed starter call, and scores zero in a start/sit comparison. Players who did not play are left out of the rank error, which needs a finish, and counted as misses in the probability checks, because the board's rankings and percentages already include the chance a player sits.
What is not adjusted. Nothing is corrected for in-game injuries, game script or garbage time: a player who left early is graded on what he scored. Player pages chart the last board before each player's own kickoff, so for later games their Week 4 projection can come from a newer board than this one.
Sample size. This is one week. A few players' afternoons move every figure on this page, so read it as one data point rather than a verdict, and nothing here has been used to retune the model.
Sources and evidence
- Model rankings-v1.4.1, configuration 91a0267e0790c386, board of 2026-10-01+r77967ba0 generated October 1, 20:25 UTC.
- Archived board (Half PPR): rankings/2026/week-04/artifacts/rankings-v1.4.1/2026-10-01+r77967ba0/weekly-half.json, SHA-256 86ca609bfedcdcd48c12d68fdc7f341c034ad1c9f8b6287c2cbdfe96678eb2a4.
- Results: nflverse weekly player statistics, published October 6, 15:55 UTC (CC BY 4.0). 4 stat lines could not be matched to our player catalog and still count toward finishes.
- Rulings: balldontlie status capture of October 4, 11:05 UTC, used only to label players who did not play.
- Expert Consensus: the Half PPR lists published October 1, 13:01 UTC, 151 rows. Comparison inputs SHA-256 df57076091a622b0c6747fd46db7462847c49cf74fdbec2eff1d8dda82f08340.
- Per-player evaluation SHA-256 abbb5db8d15cdc643dedc6c74708b0cdc5aa9a6c523c8d743e8d960dbe5a4d89.
See this week's rankingsPrevious: 2026 Week 3 reviewNext: 2026 Week 5 review
