Blue Jays Game State Explorer

Preset loaded: Illustrative scenario: home team trails by one in the bottom seventh, runners on first and second, one out. No specific game or Sportsnet post has been verified.
1712+
-10-1+10
012
Outs
0
1
2
—
1B
2B
3B
1B+2B
1B+3B
2B+3B
Loaded
Toy Win Probability0.342
Run Expectancy0.943
Toy Leverage Index1.87
Illustrative Wins426
Illustrative Losses821
Fixed Sample Size1,247

What does this game-state model actually tell you?

Change the score, outs and base runners to explore a fixed illustrative table and a simple probability heuristic. Its outputs are teaching examples; they are not verified Blue Jays game records, official MLB probabilities or betting advice.

Read the model, worked example and limitations

Follow the numbers

The three-character base code means first, second and third base, in that order. Code 110 with one out selects 0.943 from the fixed table. That value represents assumed runs remaining in the half-inning, not a guaranteed run or the chance of scoring exactly once. The table’s original claim of 284,731 historical games has no supplied dataset and has been removed.

For the bottom seventh and a home score difference of −1, the probability calculation is 0.5 − 0.06 + 0.02 − (1 − 0.943/2.3) × 0.15 = 0.3715, displayed as 0.372. The original static 0.342 label did not match its own function. The calculation clamps results between 0.001 and 0.999. It is linear with clipping, not a fitted logistic model.

Why the curve is flat

The function calculates an inning-dependent intermediate value but never uses it in probability. With score, runners and outs held constant, all nine points therefore have the same probability. Changing innings moves the marker and changes the separate toy leverage output, not probability. Toy leverage equals table expectancy × (1 − 0.07 × (min(inning,9) − 1)) × 1.8. The default is about 0.985, displayed as 0.98; its score-difference argument is unused. This is not official leverage index.

Understand the synthetic counts

The sample size is always 1,247. Illustrative wins equal round(1,247 × probability), with the remainder labeled losses. The default produces 463 and 784. These counts are generated from the formula, not queried historical observations. JSON keeps the historical_* field names for compatibility and includes an explicit model-scope statement.

Use real statistical context

MLB’s Game Strategy Explorer compares comparable historical situations and provides the dataset period and extra-inning rules. It is a useful next step when you need empirical game-state context. Its documentation does not validate this page’s fixed table or coefficients.

See the MLB leverage-index glossary for the separate official concept of situational importance. This page’s toy leverage formula has not been calibrated to that statistic.

Controls, persistence and export

The controls describe the home team batting in the bottom half; there is no half-inning selector. Innings 10–12 use ninth-inning toy leverage and a ninth-inning marker, without modeling an automatic runner. State saves in this browser’s local storage. Reload Preset and Reset both restore the same illustrative scenario. Export downloads JSON with the inputs and outputs; no import is provided. A specific Sportsnet post, player event, weather, pitcher, lineup or postseason condition has not been verified.

Super generates helpful tools and automates fact-checking across the internet proactively. If you enjoyed this tool, build your own with Super and share it with a friend.