real biathlon
    • Athletes
    • Teams
    • Races
    • Seasons
    • Scores
    • Records
    • Blog(current)
    • More
      Patreon Content Course Profiles Explanations Shortcuts
      Error Report
      Privacy Policy About
    •     
  • Forum
  • Patreon
  • Twitter
  • YouTube
    Instagram
    Facebook

Recent Articles

  • Most improved athletes this winter
  • New biathlon point system
  • Historic biathlon results create expectations. But what about points?
  • What do you expect? Practical applications of the W.E.I.S.E.
  • Introducing W. E. I. S. E: the Win Expectancy Index based on Statistical Exploration, version 1

Categories

  • Biathlon Media
  • Biathlon News
  • Long-term trends
  • Statistical analysis
  • Website updates

Archives

  • 2022
    • December
    • June
    • May
    • March
    • February
    • January
  • 2021
    • December
    • November
    • September
    • July
    • June
    • May
    • April
    • March
    • February
    • January
  • 2020
    • December
    • November
    • August
    • June
    • March
  • 2015
    • December
  • 2013
    • August
    • July
  • 2012
    • July

Search Articles

Recent Tweets

Tweets by realbiathlon

Month: November 2020

Rule changes for the 2020–21 season

Posted on 2020-11-25 | by real biathlon | Leave a Comment on Rule changes for the 2020–21 season

The International Biathlon Union (IBU) announced several rule changes for the upcoming 2020–21 season – for the most part temporary changes to due to the COVID-19 pandemic. Here is a quick summary.

The one permanent change this winter is the introduction of the dark Blue Bib, which will be worn by the top competitor who has not turned 25 by December 31. The IBU Under 25 Award will be handed out at the end of the season, replacing the IBU Rookie of the Year award. Plans to introduce the experimental Super Sprint event on World Cup level were put on hold.

For the 2020–21 season only, the number of scratch results has been raised up to four for the World Cup Total Score. Scratch results are also introduced to the Nations Cup and Discipline scores as well as the IBU Cup Total Score. In case the total number of competitions are reduced, the number of scratched results will be deducted accordingly:

It’s quite the reversal for the IBU, because the sport had moved away from dropping results in the last two decades. The last time more than two results were dropped for the Overall World Cup Score was 2009–10, the last time four races were eliminated was the 1999–00 season. In Discipline World Cups, there hadn’t been any kind of scratch results since 2009–10 and no multiple scratched races since 1998–99. The Nations Cup had four dropped results until 2009–10, after that none.

Changing the number of dropped results from two to four might seem minor, but it could have a big impact on the outcome of the season. Last year, Tiril Eckhoff would have won the overall title if only 17 of the 21 races counted towards the final score (773 to 753 points); in reality, of course Dorothea Wierer won (793 to 786 points) because just the two lowest scores were eliminated.

The IBU obviously expects athletes to miss or intentionally skip more races this year, which seems reasonable. However, three dropped results for the sprint discipline for example is a lot (30% of an athlete’s results will not count), not to mention it’s quite disproportionate between the disciplines.

Number of races per score, if there are no cancelled events:

On a more technical level, the World Cup Qualification Criteria was adjusted as well, accounting for the late start of the IBU Cup season, which would be the normal route to the World Cup. Accordingly, for the first trimester only, each National Federation can register two athletes who have not fulfilled the qualification criteria. The start quotas for the World Cup remain the same.

You can find the full document about temporary changes here: Special Event and Competition Rules – COVID-19.

Posted in Biathlon News | Tagged 2020–21 season

Projection for the season opener

Posted on 2020-11-21 | by real biathlon | Leave a Comment on Projection for the season opener

The many statistics collected on this site allow to calculate a theoretical race time, solely based on performance data. I thought this might be an interesting exercise for the season opening 15/20km individuals – plus it’s a simple reminder where we left off. These aren’t meant to be serious predictions, of course. The individual is arguably the most unpredictable discipline and produces the most surprises anyway. Not to mention a lot usually changes during the off-season, maybe even more so this year.

Note: Projected times are calculated based on ski speed, hit rates and range times in last season’s three individuals (IN). Hit rates are rounded to the nearest full shot. The top 30 median Course Time at the 2015 Kontiolahti IN is used as reference (last IN held in Kontiolahti) and multiplied with last season’s IN ski speed in percent. The “Time Loss Shooting” column follows the idea of the Shooting Efficiency score.

Men 20 km Individual

Johannes Thingnes Bø comes out on top, “winning” by a margin of over one minute, which probably isn’t surprising, especially since Martin Fourcade is no longer there. However, he’s projected to win mostly due to his extremely high 93.3% hit rate (rounded to 19/20 hits for this), not because of his ski speed, which wasn’t that remarkable in this event last winter. Quentin Fillon Maillet and Tarjei Bø are second and third – both skied faster, but were less accurate at the shooting range in 2019–20 individuals.

Race Projection based on 2019–20 IN statistics

RankFamily NameGiven NameNationRacesback from
Top30 median
(in %)
Projected
Course Time
Total
hit rate
(in %)
Projected
Time Loss
Shooting
Projected
Total
Race Time
Behind
RankFamily NameGiven NameNationRacesback from
Top30 median
(in %)
Projected
Course Time
Total
hit rate
(in %)
Projected
Time Loss
Shooting
Projected
Total
Race Time
Behind
1BoeJohannes ThingnesNOR
3-1.7143:23.895.004:30.447:54.2
2Fillon MailletQuentinFRA
3-3.8042:28.585.006:27.948:56.3+1:02.1
3BoeTarjeiNOR
3-1.7943:21.690.005:42.449:04.0+1:09.8
4ClaudeFabienFRA
3-0.8843:45.690.005:31.949:17.5+1:23.3
5JacquelinEmilienFRA
2-0.1544:05.190.005:27.349:32.4+1:38.3
6DollBenediktGER
3-1.5943:26.885.006:24.249:51.0+1:56.8
7EliseevMatveyRUS
2+1.0544:36.890.005:20.449:57.3+2:03.1
8WegerBenjaminSUI
3+0.1044:11.690.005:46.349:57.9+2:03.7
9NawrathPhilippGER
2+0.7944:29.990.005:37.450:07.3+2:13.1
10PidruchnyiDmytroUKR
2+1.2844:42.990.005:24.550:07.4+2:13.2
11LoginovAlexanderRUS
3-0.7943:48.285.006:19.850:08.0+2:13.8
12HoferLukasITA
3-1.6843:24.685.006:43.850:08.4+2:14.2
13DesthieuxSimonFRA
3-0.7743:48.585.006:25.950:14.4+2:20.2
14FakJakovSLO
3+1.8644:58.490.005:25.050:23.4+2:29.2
15DaleJohannesNOR
3-1.0043:42.685.006:47.150:29.7+2:35.6
16EderSimonAUT
3+2.3045:09.890.005:23.850:33.6+2:39.4
17MoravecOndrejCZE
3+2.2945:09.790.005:27.750:37.4+2:43.2
18HornPhilippGER
3+0.1344:12.485.006:30.550:42.9+2:48.7
19BjoentegaardErlendNOR
2+0.0144:09.385.006:38.250:47.5+2:53.3
20SamuelssonSebastianSWE
2+2.4845:14.690.005:38.550:53.2+2:59.0
21GaranichevEvgeniyRUS
3+1.2444:42.085.006:29.451:11.3+3:17.1
22PrymaArtemUKR
3+1.2344:41.585.006:37.751:19.2+3:25.0
23KuehnJohannesGER
3-0.9943:42.880.007:43.051:25.8+3:31.6
24NordgrenLeifUSA
3+4.0545:56.390.005:36.051:32.3+3:38.1
25RastorgujevsAndrejsLAT
3-0.5943:53.380.007:47.251:40.5+3:46.3
26ClaudeFlorentBEL
3+3.5145:42.090.006:04.851:46.8+3:52.6
27EberhardJulianAUT
3+0.9344:33.680.007:16.751:50.3+3:56.1
28LesserErikGER
2+5.1846:26.290.005:24.551:50.8+3:56.6
29LatypovEduardRUS
3+2.5645:16.785.006:37.851:54.5+4:00.3
30DombrovskiKarolLTU
3+6.5447:02.295.004:56.451:58.7+4:04.5

Women 15 km Individual

Olympic champion Hanna Öberg failed to win an individual last season, however, she still won the discipline World Cup title; her as the projected winner is no surprise either. Marte Olsbu Røiseland and Monika Hojnisz-Staręga round out this theoretical podium. What’s maybe most noteworthy is the fact that eight athletes are inside a minute of the winning time (none for the men) – rather emblematic of the gender divide when it comes to competitiveness at the very top of the field in the last couple of seasons.

Race Projection based on 2019–20 IN statistics

RankFamily NameGiven NameNationRacesback from
Top30 median
(in %)
Projected
Course Time
Total
hit rate
(in %)
Projected
Time Loss
Shooting
Projected
Total
Race Time
Behind
RankFamily NameGiven NameNationRacesback from
Top30 median
(in %)
Projected
Course Time
Total
hit rate
(in %)
Projected
Time Loss
Shooting
Projected
Total
Race Time
Behind
1OebergHannaSWE
3-0.9538:43.790.005:20.844:04.5
2RoeiselandMarte OlsbuNOR
3-2.6438:04.285.006:31.644:35.8+31.3
3Hojnisz-StaregaMonikaPOL
3-0.8638:45.890.005:52.344:38.1+33.6
4HerrmannDeniseGER
3-2.9737:56.385.006:42.544:38.8+34.3
5WiererDorotheaITA
3-1.6038:28.485.006:22.044:50.4+45.9
6KuklinaLarisaRUS
3+1.4739:40.490.005:17.344:57.7+53.2
7PreussFranziskaGER
3+1.3039:36.590.005:21.544:57.9+53.4
8BraisazJustineFRA
3-2.8237:59.985.007:00.044:59.9+55.4
9HinzVanessaGER
3+0.9339:27.990.005:38.445:06.3+1:01.7
10DzhimaYuliiaUKR
3+1.4039:38.790.005:41.045:19.7+1:15.2
11StarykhIrinaRUS
2+1.9439:51.690.005:41.945:33.5+1:29.0
12DavidovaMarketaCZE
3-1.3438:34.585.007:07.245:41.8+1:37.3
13BrorssonMonaSWE
2+2.4340:03.190.005:40.145:43.3+1:38.8
14LunderEmmaCAN
2+3.4540:26.990.005:20.445:47.3+1:42.8
15TandrevoldIngrid LandmarkNOR
3-0.1839:01.885.006:54.145:55.9+1:51.4
16VittozziLisaITA
3+0.7839:24.385.006:39.346:03.6+1:59.1
17EckhoffTirilNOR
3-2.0138:18.880.007:52.746:11.5+2:07.0
18SimonJuliaFRA
3-0.4438:55.880.007:17.346:13.1+2:08.6
19Yurlova-PerchtEkaterinaRUS
3+1.4239:39.385.006:35.046:14.3+2:09.8
20Kristejn PuskarcikovaEvaCZE
3+3.9340:38.390.005:45.546:23.7+2:19.2
21TodorovaMilenaBUL
2+4.2640:45.890.005:58.546:44.3+2:39.8
22EganClareUSA
3+1.9439:51.485.006:56.946:48.3+2:43.8
23HauserLisa TheresaAUT
3+3.1240:19.385.006:31.146:50.4+2:45.8
24BescondAnaisFRA
3+0.0239:06.680.007:44.346:50.9+2:46.4
25GasparinAitaSUI
3+3.0840:18.385.006:33.546:51.8+2:47.3
26FialkovaPaulinaSVK
2+0.0139:06.280.007:46.446:52.6+2:48.1
27MerkushynaAnastasiyaUKR
3+4.1540:43.485.006:18.547:02.0+2:57.5
28RiederChristinaAUT
3+5.9141:24.790.005:52.047:16.8+3:12.3
29SchwaigerJuliaAUT
3+4.0940:41.885.006:37.947:19.7+3:15.2
30ZdoucDunjaAUT
2+6.9741:49.490.005:33.047:22.5+3:17.9
Posted in Statistical analysis | Tagged 2020–21 season, projection, results

Improvements season-to-season, putting it to use

Posted on 2020-11-19 | by biathlonanalytics | 2 Comments on Improvements season-to-season, putting it to use

To highlight what a great site Real biathlon is, and how easily the data can be used to give some great insights, below is a step-by-step on how to make a quick interactive chart based on the data referenced in the previous article. I used Google Sheets and Tableau Public, but you can use any of these kinds of tools that you are comfortable with, and publish a lookup chart within an hour. That’s how easy the real biathlon site is to collect biathlon data!

Step 1 – download & store the data

For the created chart I used the data of the last five season, for both men and women, limiting the data to only those athletes with at least 10 races. Since that never resulted in more than 100 athletes per gender and season, I did not set a filter on that.

After selecting the Season Statistics, Performance Score (for example, the 2019-2020 season for women), you can just select the table, copy, open a blank Google Sheet, select cell A1 and paste. In my case, my first row showed twice so I just removed one of them. Do this 10 times (five per gender), name the sheets appropriately (W1920, W1819, M1920, M1819, etc.) and export from Google Sheets to an xlsx file. You can export to text, but you would have to do that per sheet, where exporting to excel (xlsx) exports all sheets at once.

Step 2 – open data in Tableau

In Tableau Public connect to the spreadsheet by:

  • clicking on the Data menu > New Data Source
  • click on Microsoft Excel in the Connect window
  • select the xlsx file you just exported from Google Sheets
  • at the bottom of the list that shows your sheet names, double click New Union
  • drag all 10 sheets into the Union window
  • click OK

Step 3 – create some calculated fields

  • Full Name = [Given Name] + ” ” + [Family Name]
  • Gender =
IF LEFT([Table Name],1) = "M" THEN "Men"
ELSEIF LEFT([Table Name],1) = "W" THEN "Women"
ELSE "Unknown"
END
  • Season = MID([Table Name],2,5)

Step 4 – create the chart

Depending on what you want to show in your chart, the following differs, but to replicate the chart I made, drag the following pills in the Filters, Marks, Columns and Rows:

Step 5 – publish to Tableau Public

Once you are happy with your chart, just save the file to Tableau Public.

Now users can use highlighters to see how their favourite athletes stack up against the field, or see how certain Nations fare.

Posted in Statistical analysis | Tagged data use examples, data visualization, Tableau

Improvements season-to-season

Posted on 2020-11-18 | by real biathlon | Leave a Comment on Improvements season-to-season

The new website allows you to look up basic biathlon data on your own (for different disciplines, periods, categories, etc.), so I won’t be posting too many of the regular statistical updates that I have done in the past. If you are interested in a specific statistic or ranking, you can always check out:

  • 2019–20 Shooting hit rates: Men | Women
  • 2019–20 Ski speed: Men | Women
  • 2019–20 Shooting Times: Men | Women
  • 2019–20 Range Times: Men | Women
  • 2019–20 Shooting efficiency: Men | Women
  • 2019–20 Overall Performance Score: Men | Women

These will be updated after each race. I thought it would still be interesting though to take one high-level look at last season’s performances. Below I listed the season-to-season changes in the Overall Performance Score of regular World Cup athletes (at least 14 races in the last two seasons).

Note: The scores are standard scores (or z-scores), indicating how many standard deviations (SD) an athlete is back from the World Cup mean (negative values indicate values better than the mean). The Total Performance Score is calculated by approximating the importance of skiing, hit rate and shooting pace using the method of least squares (for more details, see here and here), and then weighting each z-score value accordingly.

Men

Émilien Jacquelin was the most improved athlete last season, getting better in all major aspects of the sport: 5.5% higher hit rate, 1.8% faster skiing and 1.8s lower range time. Vytautas Strolia improved by the same amount, albeit on a much lower level, earning his first career top 20. They are followed by Johannes Kühn, who almost halved his average ski rank (12.3 to 6.7), and Erlend Bjøntegaard, who managed to increase his hit rate by 7.4%. On the flip side, Lukas Hofer‘s and Benjamin Weger‘s performance scores declined the most; both skiing over 1% slower; Hofer also hit 4.5% less of his targets.

Martin Fourcade ended his record-breaking career with his highest ever hit rate (91.8%), while his ski speed was almost back to his previous best (after a big decline in 2018–19): he had an average Course Time rank of 6.0 last winter – in 5 of his 7 title winning seasons his average ski rank was in the 4.5-5.0 range. The improvement of the French men really stand out (three in the top 10 below). Quentin Fillon Maillet became the second-fastest skier overall (1.5% faster). Johannes Thingnes Bø‘s ski speed declined slightly (on the highest possible level), yet he managed to set the best shooting percentage (92.1%) for a World Cup winner ever.

2019–20 z-Scores compared to 2018–19 | Non-Team events

NoFamily NameGiven NameNationRacesSki Speed
Score
Hit Rate
Score
Range Time
Score
Total
Performance
Score
Change
NoFamily NameGiven NameNationRacesSki Speed
Score
Hit Rate
Score
Range Time
Score
Total
Performance
Score
Change
1JacquelinEmilienFRA
20-1.27-0.83-1.36-1.15-0.44
2StroliaVytautas LTU
14-0.68-0.050.30-0.38-0.44
3KuehnJohannesGER
21-1.510.02-0.17-0.90-0.28
4BjoentegaardErlendNOR
18-1.38-0.77-0.40-1.08-0.28
5SeppalaTeroFIN
17-0.980.15-0.55-0.60-0.26
6FourcadeMartinFRA
21-1.63-1.26-0.71-1.41-0.21
7PrymaArtemUKR
20-0.79-0.53-0.72-0.71-0.20
8Fillon MailletQuentinFRA
21-1.68-0.84-1.05-1.36-0.15
9IlievVladimirBUL
16-1.140.45-0.44-0.60-0.12
10EliseevMatveyRUS
20-0.63-0.68-0.85-0.67-0.11
11BoeTarjeiNOR
21-1.50-0.93-0.51-1.21-0.09
12BoeJohannes ThingnesNOR
17-1.89-1.30-1.01-1.62-0.08
13GuzikGrzegorzPOL
14-0.18-0.19-0.81-0.26-0.08
14KrcmarMichalCZE
20-0.84-0.44-0.59-0.69-0.08
15BormoliniThomasITA
15-0.59-0.62-0.47-0.59-0.06
16YaliotnauRamanBLR
14-0.920.240.18-0.46-0.05
17PidruchnyiDmytroUKR
19-0.83-0.50-1.35-0.80-0.05
18DollBenediktGER
21-1.42-0.26-1.04-1.04-0.03
19ClaudeFlorentBEL
17-0.58-0.730.78-0.46-0.02
20FakJakovSLO
20-0.77-1.00-1.04-0.87-0.02
21SamuelssonSebastianSWE
16-0.94-0.22-0.44-0.67-0.01
22ChristiansenVetle SjaastadNOR
21-1.18-0.70-0.69-0.98+0.03
23NelinJesperSWE
17-1.050.00-0.28-0.66+0.03
24LoginovAlexanderRUS
19-1.22-1.03-1.15-1.15+0.03
25LeitnerFelixAUT
19-1.00-0.370.16-0.68+0.04
26BauerKlemenSLO
15-0.59-0.10-1.46-0.55+0.04
27MoravecOndrejCZE
17-0.54-1.06-0.71-0.71+0.05
28DesthieuxSimonFRA
21-1.33-0.76-0.81-1.10+0.08
29EberhardJulianAUT
18-1.350.24-1.11-0.86+0.13
30GaranichevEvgeniyRUS
16-0.80-0.86-0.64-0.80+0.16
31GuigonnatAntoninFRA
15-0.96-0.52-0.72-0.80+0.17
32RastorgujevsAndrejsLAT
17-1.160.11-0.26-0.68+0.18
33PeifferArndGER
20-1.15-0.97-0.79-1.06+0.18
34WindischDominikITA
21-0.990.13-0.04-0.55+0.22
35DohertySeanUSA
14-0.49-0.33-0.61-0.46+0.25
36EderSimonAUT
15-0.67-0.86-1.12-0.78+0.28
37WegerBenjaminSUI
16-0.85-0.34-0.19-0.62+0.33
38HoferLukasITA
20-1.160.12-0.06-0.65+0.35

Women

Tang Jialin improved the most among regular starters, skiing an impressive 2.5% faster. Emma Lunder increased her hit rate from 74.3% to 82.1% and lowered her average Course Time rank by 7.4. Baiba Bendika improved virtually by the same amount, mostly thanks to skiing 1.7% faster. Not far behind was Tiril Eckhoff, who went on an incredible run of 6 wins in 8 races, in large parts thanks to a career-best hit rate (83.1%); her already high ski speed also increased slightly, however, she had been faster in 2015–16.

One of the pre-season favorites, Lisa Vittozzi, had a winter to forget: her overall shooting percentage fell by 7.6%, while her ski speed declined roughly to its 2017–18 level. Susan Dunklee proves that aggregate data isn’t everything, winning world championship silver in one of her worst seasons statistically. Dorothea Wierer claimed her second overall title, shooting minimally worse (-0.9%), but skiing faster than ever (career-best average Course Time rank: 10.0). Kaisa Mäkäräinen ended her long World Cup career (358 individual top-level races, 3rd all time) on a slight uptick; although her hit rate stayed below 80% for a second consecutive year, she managed to improve her ski speed in her final season.

2019–20 z-Scores compared to 2018–19 | Non-Team events

NoFamily NameGiven NameNationRacesSki Speed
Score
Hit Rate
Score
Range Time
Score
Total
Performance
Score
Change
NoFamily NameGiven NameNationRacesSki Speed
Score
Hit Rate
Score
Range Time
Score
Total
Performance
Score
Change
1TangJialinCHN
14-0.36-0.941.15-0.35-0.49
2LunderEmmaCAN
18-0.50-0.57-1.29-0.61-0.46
3BendikaBaibaLAT
17-0.77-0.52-0.88-0.71-0.45
4EckhoffTirilNOR
20-1.57-0.67-0.40-1.17-0.42
5BraisazJustineFRA
21-1.450.09-0.18-0.85-0.35
6FialkovaIvonaSVK
16-0.800.25-0.30-0.44-0.33
7SemerenkoValentinaUKR
16-0.70-0.65-0.77-0.69-0.30
8ZukKamilaPOL
15-1.060.480.38-0.44-0.28
9SanfilippoFedericaITA
15-0.69-0.420.03-0.52-0.26
10PreussFranziskaGER
17-0.95-1.17-1.40-1.07-0.25
11DavidovaMarketaCZE
20-1.22-0.370.33-0.79-0.23
12RoeiselandMarte OlsbuNOR
14-1.47-0.82-1.04-1.23-0.23
13Hojnisz-StaregaMonikaPOL
18-1.09-1.100.02-0.96-0.22
14TandrevoldIngrid LandmarkNOR
21-1.15-0.68-0.23-0.90-0.21
15AymonierCeliaFRA
15-1.42-0.080.59-0.79-0.21
16MakarainenKaisaFIN
21-1.56-0.240.42-0.94-0.20
17GasparinElisaSUI
15-0.46-0.47-0.58-0.47-0.18
18HerrmannDeniseGER
21-1.66-0.21-0.30-1.07-0.18
19Kristejn PuskarcikovaEvaCZE
16-0.48-0.64-0.67-0.55-0.18
20ZbylutKingaPOL
15-0.35-0.37-0.35-0.36-0.17
21SimonJuliaFRA
21-1.05-0.24-1.69-0.89-0.16
22HaeckiLenaSUI
18-1.040.49-1.49-0.65-0.15
23OebergHannaSWE
19-1.14-0.94-1.57-1.14-0.14
24Yurlova-PerchtEkaterinaRUS
19-0.82-0.74-1.01-0.82-0.13
25BescondAnaisFRA
20-1.03-0.55-0.34-0.81-0.11
26PerssonLinnSWE
18-0.89-0.91-0.13-0.81-0.10
27WiererDorotheaITA
21-1.24-0.73-1.43-1.11-0.09
28BrorssonMonaSWE
18-0.80-0.74-0.03-0.69-0.06
29HorchlerKarolinGER
14-0.41-1.01-0.59-0.61-0.06
30HauserLisa TheresaAUT
18-0.63-0.99-1.06-0.78-0.04
31FialkovaPaulinaSVK
19-1.01-0.62-0.39-0.83-0.02
32HinzVanessaGER
19-0.85-0.69-0.45-0.75+0.01
33EganClareUSA
14-0.79-0.640.09-0.64+0.01
34OjaReginaEST
15-0.06-0.29-0.79-0.21+0.02
35KryukoIrynaBLR
18-0.69-0.690.06-0.60+0.07
36DunkleeSusanUSA
14-0.720.20-0.61-0.44+0.14
37VittozziLisaITA
21-0.92-0.40-0.82-0.76+0.24
Posted in Statistical analysis | Tagged 2019–20 season, overall performance, shooting, skiing

Time Behind Score: comparing fruit, rather than apples and oranges

Posted on 2020-11-18 | by biathlonanalytics | Leave a Comment on Time Behind Score: comparing fruit, rather than apples and oranges

As IBU ranking point systems vary over time and per level (Junior, IBUcup and Senior) and typically awarded only to the top 30 athletes per race, I created the Time Behind Score to compare performances between races in different seasons and at different levels.

The Time Behind Score is based on the idea that at every level, every athlete is trying to be the fastest and wants to avoid being the last athlete crossing the finish line. As not all historic data, nor the data for all levels include skiing and shooting details, this Score only uses the final time per race, regardless of the balance between skiing-time and shooting-results. Although this leads to a lack of depth for further analysis, it is the only way to compare between different level races from different eras, and in the end, the balance between skiing and shooting is less relevant when only interested in performance based on which athletes cross the finish first.

Calculation

For the Time Behind Score calculation, all total race times per race are converted to a 0-100 scale, where the fastest athlete gets a score of 100, the slowest athlete gets a score of 0, and all other athletes get a score based on the relative position between the fastest and slowest athlete. This also gives points based on relative times rather than a rank-score that ignores how much time difference exists between positions.

The figure below demonstrates the process of converting a race result to the Time Behind Score: the top half shows the race results of all athletes with the winner on the left and the last finisher on the right; the orange dots representing each athlete are placed depending on how many seconds they finished behind the winner (so the further to the right, the more seconds behind). Those “seconds behind the winner” are converted to a percentage between the winner and last finisher in the bottom half of the image (“Percentage time behind compared to maximum time behind”) with the winner being 0% and the last finisher 100%. The Time Behind Score is the inverse of this percentage, shown on the horizontal axis of the graph, so 100 for the winner and 0 for the last finisher:

Converting race results to Time Behind Score

When comparing race results between seasons and levels, I will be using the Time Behind Score as the measurement. I hope the above will sufficiently explain the reasoning and process to calculate these values. I understand that there are (as with any other scores) pro’s and con’s but I like the pragmatic idea of scores based on how the athlete did, compared to the rest of the field. However, any comments or feedback are appreciated!

Posted in Statistical analysis | Tagged Puck Possessed, Ranking, Score, Time Behind Score

Posts navigation

Older posts

Recent Articles

  • Most improved athletes this winter
  • New biathlon point system
  • Historic biathlon results create expectations. But what about points?
  • What do you expect? Practical applications of the W.E.I.S.E.
  • Introducing W. E. I. S. E: the Win Expectancy Index based on Statistical Exploration, version 1

Categories

  • Biathlon Media
  • Biathlon News
  • Long-term trends
  • Statistical analysis
  • Website updates

Archives by Month

  • 2022: J F M A M J J A S O N D
  • 2021: J F M A M J J A S O N D
  • 2020: J F M A M J J A S O N D
  • 2015: J F M A M J J A S O N D
  • 2013: J F M A M J J A S O N D
  • 2012: J F M A M J J A S O N D

Search Articles