What normalising three scores does and does not show

A worked example using a hypothetical scenario to show how the reasoning applies.

The scenario

For an illustrative field with positive scores 40, 35 and 25, dividing each by their total produces shares of 40%, 35% and 25%. Those shares sum to 100% because of the arithmetic. That alone gives no evidence about how frequently the runners would win. A different transformation could produce a sharper distribution without improving accuracy.

What to take from it

Use the calculator to inspect the transformation. Probability claims still need testing against outcomes on data unavailable during model development.

What to check next

• Identify whether the number is a rank, a raw score or a calibrated probability.
• Check whether runner scores change when the field changes.
• Compare like-for-like evaluation periods and retain the model version.

Keep the conclusion specific

Separate the observation, the interpretation and the unresolved question. A worksheet is most useful when its conclusion stays within the evidence entered into it.

This site trains no model and publishes no AI picks. The score tool is an arithmetic demonstration.