TF Rank moves ranking authority from one editor to a community of verified experts. Here is the full methodology — who votes, how much their vote counts, and how verification changes your score.
Most player rankings come down to whoever is holding the pen. One person watches some film, forms an opinion, and publishes a list. Flag football is growing too fast internationally for that to hold up — no single person sees enough games across enough countries to rank this sport honestly.
TF Rank is built the other way around. It moves the authority from one voice to a community of verified experts, and it shows its work.
The Talkin Flag player database indexes more than 400 athletes across high school, college, and national team levels. Every profile comes from one of three pipelines:
1. Official rosters — USA Football, IFAF, and national federation published rosters. We verify against primary sources and do not invent athletes. 2. Public indexes — Reference profiles sourced from publicly available databases (like flagsonly.com), tagged with their origin and marked unclaimed until an athlete verifies ownership. 3. Self-submitted profiles — Athletes submit their own profiles via our Submit Profile form. Stats can be verified by coaches or scouts.
The goal isn't to be exhaustive from day one. It's to be accurate first.
Rather than us deciding which qualities make a great flag player, the people who actually watch the games do. Coaches, expert scouts, and the hosts each complete a structured evaluation, and their answers set the weight of every ranking dimension. The three groups blend like this:
Coaches carry the most weight because they see the most football.
Within the coaching group, each vote is multiplied by that coach's credibility — their Coach IQ quiz score, level coached, win percentage, years of experience, and postseason record. Influence runs from a standard 1.00x up to a capped 2.00x. A coach who hasn't taken the Coach IQ quiz, or who scores below the threshold, votes at exactly 1.00x. Nobody is silenced; demonstrated knowledge just counts for more.
Every player is scored 0-10 on the same ten dimensions: competition level, production, athleticism, football IQ, ball skills, defence, clutch performance, intangibles, versatility, and consistency.
Scores are derived from the stats on record, adjusted for the difficulty of the league the player competes in — a touchdown against a national-team defence is not a touchdown in a low-level rec league. Where a player has verified measurables (40-yard dash, vertical jump), those are used directly. Where they don't, the model falls back to a documented proxy rather than guessing.
Data quality is part of the score, not an asterisk beside it:
That factor multiplies the final score. Claiming and verifying your profile is the single fastest way to move up the board — and it is free.
blended_weight[dim] = coaches(55%) x coach_weight[dim] + experts(30%) x expert_weight[dim] + hosts(15%) x host_weight[dim]
raw_score = the sum of blended_weight[dim] x player_score[dim], divided by max_possible, x 100 tf_score = raw_score x verification_factor
Players are ranked within their own cohort — high school (18U) and college/world are separate pools — both nationally and by position.
Rankings recompute automatically once a week, on Sundays at 02:00 UTC, and any time an admin triggers a manual recompute. Claim and verify your profile on a Tuesday and you'll see the effect that Sunday.
This post is the summary. The complete, always-current methodology — including every dimension definition and the exact weighting maths — lives at How Rankings Work. If the two ever disagree, that page is the one to trust.
Rankings are editorial opinion. Reasonable people disagree about them, and they change as data and evaluations improve.
The Talkin Flag Show
Explore all podcast episodes and listen now.
Stay in the Game
Weekly Flag Football News
Episodes, player rankings, events, and stories from around the world.
Subscribe Free →