How We Decide What Runs. And What We Are Arguing About.

One issue a day means every story competes with every other. This page is the argument, written down: what is in the pile, what it scored, what is already scheduled, and where we need you.

155

Issues published

Days of queue

Stories under consideration

0/15

Forecasts graded so far

The metric

Most sports desks pick the biggest event and hunt for numbers inside it. We do the reverse: pick the most interesting number, then explain the sport around it. Stats driven by sports, not sports driven by stats.

NEWSINESS = (GRIP ÷ 10) × (0.35 TWIST + 0.25 CLOCK + 0.15 STACK + 0.25 CARRY)

GRIP multiplies rather than adds. That is the editorial position, not a modelling convenience. A story with no number scores zero however good it sounds.

Score each 0–10. The anchors are below so two people scoring the same story land close — you should be able to do this from this page, on your phone, in five minutes.

GRIP	the number itself — multiplies everything 0 no number · 2 a number someone else published · 5 our number, one source · 8 computed from primary data, verified · 10 computed, verified three ways, with an interval and a stated limit
TWIST	distance from what a reasonable fan expects 0 confirms what everyone assumes · 3 mildly counterintuitive · 6 contradicts a reasonable prior · 9 the obvious answer is wrong and we can show why
CLOCK	what today buys over next month 0 true forever · 3 good for a season · 6 good for a fortnight · 9 decays within days
STACK	audience, in printed pages — one page is a thousand readers 0 under a page · 3 a few hundred · 6 a ream · 9 a tree or more (a Super Bowl is about fourteen trees)
CARRY	does the lesson survive outside sport? 0 sport only · 4 transfers to one other domain · 7 a general habit of thought · 10 changes how you read any number “A deficit is a fact, superiority is a rate” carries. “The Rays are lucky” does not.

Answerability is a gate, not a term. If the data cannot settle it, the story is held rather than ranked — however well it scores.

The pile, ranked

	score	story	grip	twist	clock	stack	carry
1	5.88	After One Week, What Do We Actually Know? CFB is about to take over and the queue has nothing for Sep 12; the honest answer is probably 'less than you think', which is the house thesis the number: how much a week-1 result predicts final record, 2015-2025 game logs already pulled	8	7	8	6	8
2	5.12	Nine of Notre Dame's Twelve Games Cannot Help Them. ND is the paper's first priority and this is the structural cost of independence, not a complaint about the schedule the number: resume weight by opponent: three games carry 89%	8	7	4	8	7
3	4.88	The Athletics Ruined Somebody. Did We Call It? a metric that never checks itself is advertising; this is the follow-up #144 promises the number: scoring #144's leverage claim against what actually happened	8	5	9	4	6
4	4.58	Do First-Year Coaches Beat Their Spreads? Campbell to Penn State at 8% returning production is the live case; markets are known to price transitions badly the number: every FBS coaching change vs closing spread, CFBD era	7	8	5	5	7
5	4.09	Twelve of Fifteen AL Clubs Are Alive. Nine of Fifteen in the NL. reader-noticed, verified on every measure tried, and decays as September resolves the number: SD of win pct .057 AL vs .070 NL, survives the Pythagorean check	7	6	7	5	5

Held — good stories the data cannot settle yet:

We Told You 71% Thirty Times. Here Is How Often It Happened. would score 5.7
held: needs 25+ resolved picks; the ledger has 15 and none final
The Biggest Game in Sport Is Fourteen Trees. would score 3.27
held: Stack anchors are illustrative and not yet verified to three sources

What is already scheduled

Mon Aug 31	MLB	hot · days	The Mets Cannot Ruin Anybody. The Athletics Can Ruin Four Seasons.
Tue Sep 1	Methods	keeps · any time	We Built 252 Ways to Predict the Playoffs. The Best One Lost to the Simplest One.
Wed Sep 2	MLB	dated · has a deadline	The Mets Must Pass Seven Teams. They Play Two of Them.
Thu Sep 3	Methods	keeps · any time	How Many Stars Does It Take to Flip a Team? Baseball: Four. Basketball: One.
Fri Sep 4	CFB	dated · has a deadline	Two Teams Have Everything the Contenders Have. They Are Ranked 33rd and 35th.
Sat Sep 5	MLB	slow · weeks	The Best Season to Watch Is Not the Best Season to Have.
Mon Sep 7	NFL	dated · has a deadline	Kickers Make 80% From 43 Yards. That Told Chicago Nothing.
Tue Sep 8	MLB	keeps · any time	Every Team Has Its Babe Ruth. The Mets Have Two.

Decay tags govern the order. 3 evergreen pieces are banked as buffer — that is what lets a breaking story take tomorrow’s slot without leaving a hole.

Where we actually need you

Pick one story above and give it five numbers. That is the whole ask — one story, five scores, sent back however you like. Then tell us which number of ours you would move and why: “your TWIST on that is a 4” is worth more than a paragraph of encouragement. A metric nobody argues with is a metric nobody is using.
Bring questions, not topics. The best issues this year came from someone asking a question they could not settle at the table. Gene’s produced a concept primer and an issue with his name on it.
Tell us when a piece was boring. Nobody reports this and it is the most useful thing you can say.

The open questions

Readership. The target is one reader per issue published — 155 today, and roughly a thousand before we hang a shingle. We have no weekly tracker for this yet; it is the next thing to build.
Our own calibration. We keep telling readers a thing is 71% likely. The picks ledger will eventually say how often those happened. 0 of 15 graded so far; it needs 25+ before it can speak.
Publishing the rejects. This page shows what we passed over. That record should itself become an issue — readers seeing the selection, not just the selected.
The Stack’s anchors. The audience scale is built but its readership figures are illustrative, and every one must be verified three ways before it appears in an issue.