Virtual Par Time

What is the Virtual Par Time?

In its simplest terms, it's a par time calculated independently of the pacesetter's time.

It's determined by analyzing all racers in the event and using their collective racing experience to calculate a par time.

Methodology

The process outlined below wasn’t created by chance. With over 100 million results, more than 1 million competitors, and over 100,000 races in the NASTAR database, this methodology is the result of extensive simulations and comparisons to determine the most statistically sound approach.

One alternative often suggested is calculating a race’s Par Time based on medal distribution. However, after reviewing this method in practice, it was quickly dismissed for several reasons:

  • Insufficient Competitors: Many races don’t have enough participants to create meaningful medal-based calculations.

  • Non-Representative Groups: Many races are not a cross-section of NASTAR competitors. Some events may feature predominantly beginners, where most participants earn bronze or participant medals, while others might include only elite racers earning platinum medals.

Adjusting these races to match an "expected" medal distribution would skew results and fail to reflect the actual performance of racers. Instead, our approach ensures accuracy and fairness across all race types and participant levels.

 

How is the Virtual Par Time Calculated?

The Virtual Par Time is calculated for each course in every race. Here's how it works:

  1. Determining Each Competitor’s Expected Handicap
    For every competitor with a racing history who participated in the course, an "expected handicap" is calculated. This is based on their previous race results, typically from the past two years or their most recent races. The calculation takes into account:

    • Time since last race: More recent results carry greater weight.

    • Statistical consistency: Competitors with a lower standard deviation in their results are considered more reliable.

    • Trend line: Tracks whether the competitor’s performance is improving or declining over time.

  2. Calculating Individual Par Times
    Using each competitor's expected handicap and their best result from the race, a hypothetical par time is calculated as though they were the pacesetter for the course.

  3. Assigning Weighting Factors
    A numerical weighting factor is assigned to each competitor to determine how much their data contributes to the final Virtual Par Time. Through extensive data analysis, two primary criteria were identified as indicators of reliability:

    • Number of race results: Racers with more race history produce more consistent data.

    • Lower handicap: More skilled racers typically exhibit greater consistency.
      Using these criteria, a formula assigns higher weighting to competitors with extensive race history and lower handicaps.

  4. Calculating the Non-Adjusted Virtual Par Time
    A weighted average of the calculated individual par times is computed, factoring in each competitor’s weighting. This produces the "non-adjusted virtual par time" for the course.

  5. Applying the Adjustment Factor
    Although Virtual Par Times are statistically very close to pacesetter-derived Par Times, a slight difference remains. To align these values, an adjustment factor is applied, producing the final "Virtual Par Time".

    • Current Adjustment Factor: As an example of how closely Virtual Par Times align with pacesetter-derived Par Times, the adjustment factor as of December 2024 is 0.988. This means that Virtual Par Times are, on average, only 1.2% different from those set by pacesetters.

    • Monitoring and Updates: This factor will be regularly reviewed and may be adjusted over time by analyzing medal distribution across all races.