# Special Calculations for Variation Score, Peak Performance Risk and Over training Risk

N.B. These special calculations only use Numeric fields that ARE set up with a data direction (e.g. higher or lower is better) and Option based fields (e.g. single selection or dropdown) that ARE set up to have a Best Value. If this information is NOT set up on your Event Forms then the these calculations may not be running optimally.

## Variation Score, Peak Performance Risk and Over training Risk calculate the difference of a newly entered event against all of the past history for that athlete. These are best used as part of daily review and monitoring forms.

See the steps below to find out more about the scores and risk equations that automatically detect how different the athlete is to their "normal".

## Variation Score calculates a score (out of 10) based on how DIFFERENT a new entry for an athlete is compared to the athlete's normal state (based on their history). It is essentially showing how different the new entry is.

Variation Score

The calculation can only use Numeric fields that ARE set up with a data direction (e.g. higher or lower is better), and Option based fields (e.g. single selection or dropdown) that ARE set up to have a Best Value. Based on this information the system, learns how an athlete usually responds to the questions,and then calculates a score based on how different the new entry information is to the athletes learned normal state and produces a risk profile score. This score utilises the information that you have set up about the questions (such as higher or lower is better, and preferred options etc) and produces the variation score to show how different they are from normal (with no positive or negative direction as with the peak performance and over training risk algorithms).

1. Click on Variation Score ( the artificial intelligence based algorithm automatically uses ALL of the questions in the form. It does NOT selection of fields like the sum and average Calculation Option set up)

2. Set up the Properties

This field is usually used on forms where you are regularly capturing lifestyle or lifestyle based training data such as recovery forms, daily monitoring form etc. The score is from 1 to 10 and the results are as follows:

1 means the athlete is normal, 10 means they are extremely different, in no specific "direction".

## Peak Performance Risk calculates a risk score based on how DIFFERENT (e.g better than normal) the new entry for the athlete is compared to the athlete's normal state as peak performance score

Peak Performance Risk

The calculation can only use Numeric fields that ARE set up with a data direction (e.g. higher or lower is better), and Option based fields (e.g. single selection or dropdown) that ARE set up to have a Best Value. Based on this information the system learns how an athlete usually responds to the questions,and then calculates a score based on how different the new entry information is to the athletes learned normal state and produces a risk profile score. This score utilises the information that you have set up about the questions (such as higher or lower is better, and preferred options etc) and produces the score to show how different they are from normal in a Positive direction.

1. Click on Peak Performance Risk ( the artificial intelligence based algorithm automatically uses ALL of the questions in the form. It does NOT selection of fields like the sum and average Calculation Option set up)

2. Set up the Properties

This field is usually used on forms where you are regularly capturing lifestyle or lifestyle based training data such as recovery forms, daily monitoring form etc. The score is from 1 to 10 and the results are as follows:

1 means they are as per normal, 10 means they are extremely different today (in a good way).

## Over training Risk calculates a risk score based on how DIFFERENT (e.g worse than normal) the new entry for the athlete is compared to the athlete's normal state as performance risk score

Over Training Risk

The calculation can only use Numeric fields that ARE set up with a data direction (e.g. higher or lower is better), and Option based fields (e.g. single selection or dropdown) that ARE set up to have a Best Value. Based on this information the system learns how an athlete usually responds to the questions,and then calculates a score based on how different the new entry information is to the athletes learned normal state and produces an over training risk profile score. This score utilises the information that you have set up about the questions (such as higher or lower is better, and preferred options etc) and produces the score to show how different they are from normal in a Negative direction.

1. Click on Over training Risk (the artificial intelligence based algorithm automatically uses ALL of the questions in the form. It does NOT selection of fields like the sum and average Calculation Option set up)

2. Set up the Properties

This field is usually used on forms where you are regularly capturing lifestyle or lifestyle based training data such as recovery forms, daily monitoring form etc. The score is from 1 to 10 and the results are as follows:

1 means they are as per normal, 10 means they are extremely different today (in a bad way).

For each athlete, the questions in the specific Event Form:

1) If the question is categorical it has a frequency distribution created for their history for this question.

2) If the question is numeric it is discretized then has a frequency distribution created for this question.

3) The amount of information is calculated according to standard information-theoretic principles (Traditional Shannon Entropy):

The results determine the total amount of information by summation and scale the result to fit in the 0-10 range.

In the case of overtraining or peaking calculations the information is weighted by the "data direction". I.e. using knowledge of what values

are good or bad, weight the information content relative to how bad it is (for overtraining) or how good it is (for peaking). For example in

overtraining calculations, "high information" and "bad" values have a higher weighting towards the total than "high information" but "great"

values. In peaking calculations, "high information" and "great" values have a higher weighting towards the total than "high information" but

"poor" values.