Ranking benchmark
Ball Ranks vs Yahoo's fantasy basketball analysts
Model Zero beat Andy Behrens and Dan Titus across their archived preseason boards and finished nearly even across the full Yahoo analyst field. Every result is published.
Each preseason board is graded against completed-season per-game results. First-year players are excluded on both sides, so every player compared had at least one prior NBA season.
The field
4 analysts, 13 seasons, every board written before tip-off.
The result
Across every board, the gap is 0.083 of a rank.
Ball Ranks averaged a weighted rank miss of 31.517. The analysts averaged 31.6. That is 0.26% in our favor across 4,209 comparisons. Model Zero finished slightly ahead overall, effectively even with the full analyst field.
Bars right of zero mean Ball Ranks was closer across that analyst's whole sample. Bars left mean the analyst was.
Seasons where more than one analyst published, restricted to players every one of them ranked. Shorter is better. The best board each season is marked.
★ marks the smallest weighted rank miss that season.
The split
The average hides the interesting part.
The nearly even overall result combines clear analyst-level wins and losses. Ball Ranks beat Andy Behrens in 4 of 4 seasons and Dan Titus in 5 of 6. Against Dalton Del Don it won 4 of 9.
Brandon Funston's single season is one board and should not be read as a trend. Del Don's nine seasons are a different matter. That is the longest sample here, and he is ahead across all of it.
The per-analyst numbers are not directly comparable, because each analyst is scored on their own pool of players. So for the seasons where more than one of them published, we restricted the comparison to the players every analyst that season ranked. Same players, same season, no advantage from a deeper board. Del Don takes 4 of those 5, and in 2022-23 he beat Ball Ranks by 3.6 ranks on identical players. That is not noise.
Among players who actually finished in the top 50, Ball Ranks' weighted miss was 17.951, compared with Del Don's 18.444. Outside that group, the weighted misses were 37.973 for Ball Ranks and 36.6 for Del Don. That lower part of the pool is where he built his overall edge; Model Zero was closer among the actual top 50.
Bars above zero are seasons Ball Ranks was closer. Colour identifies the analyst.
Solid bars are seasons Ball Ranks won. Faded bars are seasons the analyst won.
The rules
Every board faces the same completed-season test.
Every board predates opening tip
All 20 boards were published before that season began. 5 appeared on opening day; even the closest arrived 4 hours before tip.
Same players, same pool
Analysts rank different numbers of players. For each comparison we kept only players on that analyst's board who reached 20 games and had a prior NBA season, then re-ranked the analyst, Ball Ranks, and the actual finish inside that shared pool.
First-year players excluded, both sides
Between 4 and 21 first-year players were removed from each matchup before either side was scored.
Misses on players who mattered count more
The best actual finisher in each shared pool counts 2x, and that weight declines smoothly to 1x for the last player. Missing on the fourth-best player in fantasy still matters more, without arbitrary cliffs between neighboring ranks.
What we take from this
Model Zero beat 2 analysts and finished ahead overall.
Across 4,209 comparisons our board is 0.26% ahead. It beat Andy Behrens and Dan Titus across their available boards. Dalton Del Don beat Model Zero across the longest analyst record, while Brandon Funston won the one season available for him.
The full result is the value: every preseason board is graded against completed-season results, and the losses sit beside the wins. Model Zero finished nearly even with this field while posting clear aggregate wins against two established analysts.
That is a fantasy basketball board with a public record, not an ungraded preseason opinion.
Method notes
Boards were recovered from the FantasyPros expert feed, the same source Yahoo Sports embeds in its own rankings articles. FantasyPros labels them overall rankings rather than strictly 9-category. Every matchup still scores both sides on the same players with the same per-game method.
Every board's publication time was checked against that season's opening tip. Any result published at or after the opener was excluded.
Model Zero and each analyst are re-ranked inside the same eligible player pool before their preseason ranks are compared with the completed-season finish.