Key finding
FantasyPros Consensus led the 2026 Week 3 half-PPR kicker accuracy comparison with a skill score of 10.6, ahead of Eric Moody by the stated 7.1 points. The field contains 14 eligible tracked ranking products. FantasyPros Consensus and Moody scored above zero, Draft Sharks finished exactly at zero, and all remaining products scored below zero. That distribution frames the result: the leader outperformed this week’s measured field, but its position needs to be read alongside the random-order baseline and each product’s own through-week record. [1][2]
What the leading score actually measures
Skill score measures how much better than a random order a ranking matched how players actually scored. On this scale, 0 is random and 100 is perfect; most rankers land between about 5 and 35 because weekly fantasy scoring is noisy. FantasyPros Consensus’s 10.6 therefore represents the highest observed score here, not a near-perfect ordering or a prediction success rate. The comparison concerns named kicker ranking products in this half-PPR weekly sample, rather than an overall assessment of the services associated with them. [2][1]
The leader’s own record adds context that its placement alone cannot provide. FantasyPros Consensus averaged 0.0 through Week 3 over 3 eligible weeks out of 3 published weeks, putting this week’s result 10.6 points above that average. Eric Moody’s 3.5 also exceeded his through-week average of 1.6, by the supplied comparison of 2.0 points. Moody’s average likewise covers 3 eligible weeks out of 3 published weeks. Both positive scores were above their own observed averages, although that does not turn the stated 7.1-point lead into a forecast. [1]
Why an above-average week can still fall below random
Draft Sharks provides the clearest distinction between the weekly baseline and a product’s own record. Its 0.0 matched the random-order point, yet stood 17.6 points above its through-week average of -17.6. That average covers 3 eligible weeks out of 3 published weeks. In other words, a result can be above its own average without exceeding the random-order benchmark. This distinction matters when reading the negative scores below Draft Sharks: placement, performance against random, and performance against a product’s own average answer different questions. [1][2]
The next stretch of the field reinforces that distinction. Matt Bowen scored -5.9, Mike Clay -7.1, Eric Karabell -8.2, and Daniel Dopp -9.4. Each result was below random on the stated scale, but each also exceeded its own through-week average: Bowen by 9.4 points, Clay and Karabell by 9.0, and Dopp by 6.7. Those comparisons make the middle more informative than a simple ordering of negative scores. They describe above-average weekly results for these products without implying that the rankings beat the random-order baseline. [1][2]
Where the same-product comparisons diverge
Farther down the field, Liz Loza and Tristan H. Cockcroft offer a useful comparison because both had a through-week average of -17.6 over 3 eligible weeks out of 3 published weeks. Loza’s -14.1 was 3.5 points above that reference, while Cockcroft’s -18.8 was 1.2 points below it. Their shared average does not make their weekly results interchangeable. It instead shows how the same historical reference can frame an above-average result for one product and a below-average result for another. [1]
Field Yates and Footballguys (PPR board) illustrate the complementary point: different places in the weekly field can still share the same direction of comparison with their own records. Yates scored -12.9, a stated 7.8 points above his -20.8 through-week average. Footballguys (PPR board) scored -22.4, a stated 5.9 points above its -28.2 average. Both averages cover 3 eligible weeks out of 3 published weeks. Their negative weekly scores describe performance against the baseline; their positive differences describe performance against their respective observed averages. [1][2]
What the lower end adds to the overall spread
The contrast continues near the bottom. ESPN app projections scored -31.8, above its through-week average of -35.7 by 3.9 points. FFToday scored -38.8, above its -50.2 average by 11.4 points. Both records cover 3 eligible weeks out of 3 published weeks. FFToday’s larger stated difference from its own average did not place it ahead of ESPN app projections this week. That is why the size of an above-average showing should not be substituted for the weekly accuracy comparison itself. [1]
FF Toolbox (PPR board) finished lowest at -48.2, placing the stated highest-to-lowest spread at 58.8 skill points across the 14 products. Its weekly result was also 13.5 points below its through-week average of -34.7. Unlike the other displayed products, that average covers 2 eligible weeks out of 3 published weeks. The differing sample scope belongs alongside the comparison, not as an explanation for the score. Neither this spread nor the leader’s 7.1-point edge establishes statistical significance or future separation. [1][2]
Published status and the limits of this weekly snapshot
The publication reports 0 provisional results among the 14 displayed products; all are published. The supplied evidence does not define the criteria for a provisional designation, so it should not be treated as a separate accuracy judgment. Nor does non-provisional status mean the evidence is immutable: these results remain subject to the publication and correction process. Results use the weekly lock policy and permitted published observations, and corrections or permission changes can change the available evidence. Status describes the current publication, not a guarantee against revision. [1][2]
For a fantasy manager, the useful takeaway is a distinction between this week’s ordering and each product’s observed record. FantasyPros Consensus led the measured kicker field, while Draft Sharks and several negative-scoring products also exceeded their own through-week averages. Those findings help explain this particular comparison without identifying why any provider performed as it did. A single week cannot establish season-long or future superiority, and the seasonal participation gate is separate. This sample supports a reading of Week 3 accuracy, not a promise about the next rankings. [1][2]
Sources
- Fan50 evidence record · Checked October 29, 2026
- Fan50 methodology · Checked November 5, 2026