Now that the Realbiathlon website has some IBU data as well (available to Patreon supporters), I wanted to do a comparison of metrics for athletes that competed at both levels for a minimum of five races (leaving 437 observations). Although there are different race disciplines for the two levels, in this first look I included all disciplines and combined the categories (genders). I looked at all athletes that competed in the current or last season at the IBU level, and at the World Cup level since the 16-17 season (661 athletes).
These charts are nothing fancy, just comparing athlete’s average metrics for the two levels and drawing a trend line to see what relationship exists between the results for the following metrics:
- Total Shooting Percentage
- Prone Shooting Percentage
- Stand Shooting Percentage
- Course Time
- Shooting Time
- Range Time
As the P-value for all charts is smaller than 0.0001 we can say all relationships are statistically significant, or, that there is a relationship between the results at the two different levels. But how strong is the relationship?
For the Total Shooting percentage the relationship is very strong: if a shooter has a certain percentage at the IBU level, almost 93% of the time the shooter is has a similar percentage at the World Cup level.
Prone and Stand don’t differ much, with Prone having a 0.907 R2, and Stand a 0.916 R2.
Since Course time, Shooting time and Range time are not expressed in percentages but in Z-values, which looks at how how much the times differ from the average (negative is faster, positive is slower), we see different patterns. But the relationships are still very strong:
In general, we can say that metrics from IBU level races translate very well to World Cup level races, but since the level of competition at the World Cup level would likely be higher than on the IBU level, the results for the athletes could, and likely will be very different. To make this more clear, here are the averages for the used shooting metrics at the two levels for all 661 athletes (rather than the subset of 437 athletes that have a minimum of five races at both levels):
- Shooting Percentage: 80.1% (WC) versus 77.5% (IBU)
- Prone 82.8% (WC) versus 80% (IBU)
- Stand 77.34% (WC) versus 75% (IBU)
Note: Course Time, Shooting Time and Range Time are all Z-scores so on average they are the average (0).
From the shooting averages, we can conclude that the average shooting is better at the World Cup level than at the IBU level, which is what you would expect. So an athlete who had the same score on both levels can be above average at the IBU level and below the average at the World Cup level.
I hope to do further research on the IBU data once more seasons become available in the summer. To be continued.