The week ahead, written up for your group chat. Every matchup with its win chance and projections, who each side is leaning on, and the players to watch. Every number is the one Matchup shows; Copy preview puts it on your clipboard with a link.

The week, written up for your group chat. Once every game of a week is over, each matchup gets a sentence or two and the week gets its awards. The scores and the awards are worked out from the box scores; only the sentences are written, and each one is checked against those numbers before it appears. Copy recap puts the whole thing on your clipboard, with a link to the site.

Where the week actually stands. Not the score — the score cannot tell a thirty-point lead with their whole slate still to come from the same lead with one man left. Still open is the chance the week finishes the other way from how it leans now: 45% when you are a 55% favorite, nothing once it is settled. Riding on and Who can beat you split that one amount between the men still to play, each in proportion to how much his game can still move the margin — so a card reading 15% with 45% open means he carries a third of what is left to decide, and the cards on both panels add up to what is still open. A man projected fourteen in a game you are winning by forty matters less than a man projected six in a game you are losing by four.

The currency depends on the system. In a budget league what matters is what is left to bid; in a rolling priority league it is your place in the queue, and a claim only resets when you spend it.

Three numbers, and they are not the same number. Proj is what a man was projected at kickoff and never moves after it. Scored is what he has put up. Edge is where the slot ends up — what each player has banked plus the part of his projection his game has not reached yet — so a man who is done contributes what he actually got and a man who has not started contributes all of it.

The eight edges add up to the total at the bottom, and to the big number in the middle of the scoreboard — the difference between the two on pace figures, which is where the week is heading and what the percentage beside each squad is worked out from. The smaller figure under it is the plainer thing: the gap between the two scores either side of it, right now. Slot by slot is still the better read of all of them — one enormous mismatch is worth more than being slightly ahead everywhere.

Two different problems, shown separately. Availability is a health question — whether he plays at all, from a designation that may clear by Sunday to a stay on IR that will not. Role is a usage one: nothing is wrong with him, he simply is not the one getting the ball, and no amount of recovery time changes that. Collapsing them into one badge hides the distinction that matters most. Risk runs 1 to 5. These figures are deliberately not folded back into the projection — ESPN already prices designation to a degree, and penalising the same injury twice would be worse than showing it plainly.

What is being written about your league, whether or not anybody has been given a designation. An injury report is a lagging thing: it says a man is out after somebody has decided he is out. Coverage is where you find the ankle he walked off on, the coach declining to name a starter, the charge, the holdout, the trade being reported before it is announced. None of it is a designation and all of it is a reason to look again before Sunday.

Adds, drops and trades only. Draft picks and lineup changes are not transactions and are left out. A trade shows both sides and which way each player went; an offer that was declined shows as much as ESPN keeps, which is not the players.

A percentage does not answer the question you actually have. “You make the playoffs 62% of the time” is true and unusable; win two of your last four and you are in is the same simulation read along the axis a manager can do something about. Need is the fewest wins that put you past 95%, Alive the fewest that keep you above 5%. Nothing here says “clinched”, and that is deliberate: this league breaks ties on points scored, so a berth is never a matter of wins alone, and a line claiming otherwise would be wrong in exactly the close seasons where it mattered. A bye costs one week, not the season. Each squad plays this week’s game on this week’s lineup and every game after it with the lineup it can field that week, so a star resting this Sunday counts against his team this Sunday only, and a bench player who covers a later bye counts in the week he plays. Proj/wk is the rest-of-season level. This week counts what has already been scored. This week’s matchup is played from where it stands — points already banked plus what is still to come — the same figures the Matchup scoreboard shows, so a forty-point Sunday lead counts as the lead it is.

Graded against a forecast that was sealed first. Each week’s projections are written down the first time the week is seen and never rewritten, because a forecast that could be revised afterwards would quietly converge on the result and report perfect accuracy. Nothing appears here until games have been played.

Two numbers a week, because one of them is mostly luck. Left on bench is the best lineup you could have fielded knowing the final scores, minus what you did field — everybody leaves points there, and nobody knew the backup would go for thirty. Your calls is the honest one: what you fielded minus what the page’s own lineup at kickoff would have scored, from the projections sealed before the games. A plus means your calls beat the numbers; a minus means going against them cost you. It grades the page as much as it grades you.

Who is actually good at this, and who has been lucky. Efficiency is what a manager's starting lineup scored against the best lineup that roster could have produced that week, knowing the final scores. Nobody reaches 100% and nobody should — the hindsight lineup was unknowable at kickoff — so it is only worth reading against the other managers in this league. All-play is how each week would have gone against every squad rather than the one the schedule picked, which is the fairest measure of a team there is. Luck is actual wins minus the wins that all-play record deserved: a team at 3-0 with the worst scores in the league has taken three the schedule handed it, and this says so with a number.

Who is strongest right now — which is not who is top of the table. A record is a receipt for games already played. This is a valuation, and it has two halves, both of them columns.

Past is what they have done: all-play, how each week would have gone against every squad rather than the one the schedule picked, with Last 3 and scoring level inside it. Ahead is what is in front of them: Roster left, the rest-of-season points of the best lineup that squad can still field — with the weeks injured men are missing already taken off them — adjusted for Rest sched, how good the opponents they have yet to play have been.

Ahead counts for more than Past. Nick: “If I am 5-0 and my entire team gets hurt for the rest of the year, I am not the best team in the Power Rankings.” At an even split that squad would still sit mid-table; weighted this way it falls to where it belongs, and its Past column still reads what it earned.

Two things this does not claim. It is not a forecast of next Sunday: measured week by week against this season it predicts a single week barely better than the standings do. And Move is what the results did — ESPN does not keep what a squad looked like a week ago, so last week’s order is worked out with this week’s rosters, and a trade shows up in Roster left straight away rather than as a move of its own.

Ours is the number the page stands on. It starts from the middle of the experts’ projections, ESPN’s among them, each scored under your league’s rules, and then leans on what they do not fold in: the Vegas team total, the game script from the spread, who he is actually lining up against, the wind and rain where there is any, and his own season rate — as a bounded tilt. What it will not do is overrule ESPN on whether a player has a role: a near-zero ESPN projection means inactive or buried, and those are gated out rather than conjured into starters. Edge is ours minus ESPN’s: the sign is the call, the size is how strongly. The columns after the season numbers show where each week’s number came from.