Toronto's +1.42 run-diff surge over the last 30 games gives them favorable late-inning win probability in tight AL East rivalry contests.
Toronto's +1.42 run-diff surge over the last 30 games gives them favorable late-inning win probability in tight AL East rivalry contests.
Move four levers to inspect an authored heuristic. No dated performance feed, broadcast schedule or calibrated forecast has been verified.
Toronto's run index is (OPS/0.720)^3.8 times4.6; New York uses times4.5. Assumed runs allowed are (bullpen ERA/4)^1.1 times4 for Toronto and times4.1 for New York. Each pair feeds scored^1.83/(scored^1.83 + allowed^1.83). Their resulting values T and N enter (T-T*N)/(T+N-2*T*N), then an authored .035 home adjustment is added and the result clipped to .20-.80. These mappings and coefficients are uncalibrated.
Exactly at ERA3.15/3.60 and OPS.785/.772, a special case replaces the formula with .542. Without that override the same inputs give about .6505. Moving an input away from the exact fixture exits the override: a sudden jump reflects implementation, not measured team-strength change. The .772 fixture is also off the OPS slider's .005 step grid; native slider values need separate verification.
The thirty-point chart accumulates linear changes and sine/cosine terms, then forces final values to +1.42 and+1.15. It does not use the levers, recent records or actual box scores. These run-differential values also do not enter the probability formula.
The series buttons select fixed annotations. A3-0 head-to-head sweep gives a three-game relative standings swing and2-1 gives one, holding other results fixed. Playoff shifts +8.4/+2.8/-2.9/-8.1 are authored constants. The switch does not recalculate single-game probability or a postseason simulation.
Read MLB's Pythagorean winning-percentage glossary, OPS definition and ERA definition for statistical context. These do not validate bullpen ERA as all-team run prevention or the custom OPS conversion.
Reset restores only the four levers and retains the selected series annotation. JSON exports current inputs and all fixed scenarios, not the selected scenario or chart points. The text report is a summary; no import or browser persistence is provided. Rounded precision is formatting, not forecasting accuracy.