Best bets
Tight end touchdown picks (anytime and first TD) from the model and this week's sportsbook prices.
Track record
Every pick is saved at the odds snapshot and locks at kickoff, then gets graded after the game. Each pick is a $10 bet at the listed price; Units is total profit in $10 units. Total adds up every category, so a player picked in three categories counts three times. Unique bets counts each player once per market (anytime TD, first TD) each week, which is what you'd have bet at the book. CLV (closing-line value) is how many percentage points the market's chance moved toward the pick, on average, between the first odds snapshot that had it and the last one before kickoff. Steady positive CLV means the picks are finding prices before the books move, a faster sign of a real edge than wins and losses. Picks seen at only one snapshot (like Thursday games) aren't counted.
Book is the best price across US sportsbooks, with the book under it. Model is our anytime-TD chance (first-TD chance in the First TD category); Market is the median sportsbook's implied chance, which includes the books' margin. Edge is model minus market in percentage points. EV / $10 is the expected profit on a $10 bet at the best price if the model is right. A model that only knows targets, field position and team scoring will be wrong in ways the books aren't (injuries, game plans), so treat these as a starting point, not betting advice. Players listed as Out, Doubtful or on IR are never picked. A Q tag means Questionable. CC rows are close calls: the next-best players, shown in lighter text to fill a category to five when fewer meet its rule. They aren't official picks. They're graded in a separate shadow record, which shows whether the rules are too strict or too loose, and never count toward the official record. Longshots and First TD value also require at least 2 targets in the team's last 2 games. Players that rule cuts from the top five are graded in their own shadow table, Cut by the usage rule, which shows whether the rule is costing winners. Grading: a pick wins if the player scores a receiving or rushing TD. A player who didn't play is voided, the way books handle inactive players. A player who played but never caught a pass counts as a loss. Fades aren't bets (the "No" price isn't tracked), so they're kept out of the totals. Each fade shows ✓ No TD or ✗ Scored, and the line under each table compares how many faded players scored with how many the market and the model expected. Fewer scorers than the market expected means the fades were right. The market figure includes the books' margin, so it runs a little high. First TD bets win only if the player scores the game's first touchdown, by either team. They lose if someone else scores first or nobody scores a TD. Books take a much bigger margin on this market, so its Market column has the margin removed. The model's chance is the player's share of both teams' expected TDs.