The four signals, in plain language.

WarCurve takes one player and returns one number: a defensible estimate of where their career sits today, and where it is most likely to sit six months from today. That number flows through every leaderboard on the site and underwrites every card valuation. This essay describes, in plain language, what goes into it.

1 · Career Wins Above Replacement

Wins Above Replacement (WAR) is the public-domain framework the baseball-reference community has refined over twenty years. It collapses everything a player has done — at the plate, on the basepaths, in the field, on the mound — into a single number expressed in wins above a freely available replacement-level player. Career WAR is the sum of every season a player has banked. It is the spine of the model and it is non-negotiable.

We do not invent our own WAR. We consume what the public record has already settled, then weight and project it.

2 · Peak three-year WAR

Career WAR alone rewards longevity at the expense of dominance. A compiler who played eighteen unremarkable years can accumulate more WAR than a generational talent who played twelve excellent ones, and the card market refuses to grant the compiler the same valuation. Peak three-year WAR — the best rolling three-season window of a player's career — captures the dominance the market actually pays for.

3 · Cultural pull

The honest part of the model is the WAR. The honest part of the market is that not every win is priced equally. A Yankee outfielder with a popular jersey number commands a valuation an equally productive player on a small-market club does not. We measure this with a popularity index built from Wikipedia traffic, Baseball Reference page views, and a small number of social signals — never from anything that could be manufactured by a card-side counterparty.

4 · Milestone proximity

Three thousand hits, five hundred home runs, three hundred wins. Three thousand strikeouts. A Hall of Fame chase tilts valuation in ways no rate stat captures. We compute the distance to each public milestone and the probability of reaching it given the player's current age and trajectory, then apply the resulting tilt to the projection.

The projection

We blend the recent three seasons heavily against the career line, apply an aging curve fit on the public WAR record of every active and recently retired player, then carry the resulting rate forward six months. The blend protects against a fading veteran scoring on lifetime résumé, and against a young phenom scoring on a single hot quarter.

What the model refuses to do

• It will not extrapolate from a single hot week.
• It will not credit a player for a season they have not yet played.
• It will not rank a 43-year-old veteran ahead of an in-prime talent on lifetime totals alone.
• It will not respond to grading-company press releases or breakdown counterparties moving in coordinated fashion.

From the player number to the card price

The player number is half the work. The other half is the cardboard hierarchy — set, parallel, grade, era — fit on one hundred and eighty days of real sales. That conversation continues in the rookie hierarchy essay.

Filed for the pre-launch · July 4, 2026