The team page shows every active club in the league you're viewing, with season-long totals and rate stats sourced from public play-by-play data. Records and point differentials are the headline numbers, but the columns that travel further across predictive models are the rate stats — points per possession, point differential per game, and home/away splits.

A useful read of a team page is rarely the team at the top. It's the team whose point differential looks much better or worse than their record suggests. Those teams tend to regress toward their underlying numbers over the back half of a season, and knowing where the gap is between record and run differential is the closest thing there is to a free predictive signal.

Want more on what to look for? Read the QB stat that actually predicts winning, the best defensive linemen never get sacks, or how load management broke the All-NBA vote.

The player page is the largest table on the site. Every active player in the currently selected sport, with their season totals and per-game rate stats, sortable by any column. The default sort is whatever scoring stat that sport's box score uses most often — points in basketball, passing yards in football, OPS in baseball — but the column you sort by is usually the column you should be most suspicious of.

Most box-score columns are noisier than they look. Steals and blocks reward gambling that worked. Save percentage treats every shot as the same. Batting average has been obsolete for decades. The page lets you sort by any of them, but the more useful read is usually to look across two or three rate stats at once — efficiency next to volume, counting stats next to per-possession context. A leaderboard isn't an evaluation; it's the starting point for one.

Deeper reads on individual stats: true shooting vs. FG%, save percentage is broken, why batting average refuses to die, how to actually read an NBA box score.

Standings are the most-watched table in any sport and the most misread. The win-loss column is a coarse summary of a season's worth of probabilistic outcomes, and the teams at the top in March are often not the teams the underlying numbers favored. Look at point differential next to record. The team whose record outpaces its differential by more than a couple of wins is usually getting lucky in close games; the team whose differential outpaces its record is usually due for positive regression. Both are predictive over the season's back half.

Conference standings deserve a separate read. Conference imbalance is real and chronic across most leagues; the eighth-place team in a strong conference is often a better club than the third-place team in a weak one. Use the standings table to identify the bubble teams, then check their underlying differentials and remaining schedules before drawing conclusions about who's actually a playoff contender.

Related reads: the relegation six-pointer is a myth, going for it on fourth down.

Stat leaders are the league's most-circulated numbers and the easiest to misread. Counting-stat leaders (points, rebounds, yards, goals) reward volume and usage; rate-stat leaders (percentages, efficiencies) reward consistency on smaller samples. The minimum-games filter is there to keep small-sample noise out of the rate-stat leaderboards — a player who's gone 4-for-6 from three isn't shooting 67%, they're shooting on a tiny sample, and the filter prevents that from being mistaken for excellence.

Swap categories liberally. The same player can look very different across the counting and the rate version of the same stat. A scorer who leads the league in points and finishes 30th in true shooting percentage is producing volume, not efficiency. A scorer who's 12th in points and second in true shooting is probably the better offensive player. Leaderboards tell you a fraction of that story by themselves. Crossing two columns is usually where the real read lives.

On reading the columns themselves: true shooting and the death of FG%, why batting average refuses to die, the best defensive linemen never get sacks.

Side-by-side player comparison is the most useful tool on this site and the easiest to misuse. The temptation when looking at two stat lines is to declare a winner — Player A scored more, therefore Player A is better. Most of the time that conclusion is unsupported by the same comparison's other columns. Two scorers can produce identical points-per-game on very different shot profiles; two pitchers can post identical ERAs while pitching very different innings against very different opponents.

The right way to read the comparison panel is to scan for divergences. Where do the two players' columns disagree most sharply? That's where their styles actually differ. Sometimes the divergence is the answer to a question scouts are asking — a player whose volume column is high and efficiency column is low has a different role than one with the opposite profile, and a smart coach will use them differently regardless of which one scores more. Don't look for a verdict. Look for the gap.

Background reading on comparing stats responsibly: how to actually read an NBA box score, the mid-range jumper isn't dead, the QB stat that actually predicts winning.

Favorites is a tracking tool, not a leaderboard. The teams and players you star here are stored locally in your browser — no account, no server. The point is to give yourself a private dashboard for the storylines you actually care about, whether that's your favorite team's playoff push, a rookie having a breakout season, or two players you want to compare across years.

Sports analytics work better when you have something specific to watch. Following a single player across a hundred-game season teaches you more about how to read stats than glancing at five different leaderboards every week. Favorites is built for that kind of deep tracking. Star a team and watch how its differential tracks its record over a few weeks. Star a player and notice which columns are stable and which are noisy.

For more on what to look for as you watch over time: how to actually read an NBA box score, why goalies are impossible to evaluate, what one full year of the shift ban actually proved.