How many times have you heard the comment "It will take a lifetime best in heats to advance to finals?" The past two Olympics results show that if you are fast enough, you will not need a best time to win. You must simply win. Alexander Popov, the winner in both freestyle sprint events, swam this formula to the letter.
Following up on the 1996 Olympic statistics report from last month, I collected some results from the 1992 Olympics in Barcelona and the 1988 Olympics in Seoul. I was looking for a pattern in personal best times (PB). Many variables began to appear.
Variables such as age, gender, percent improvement, environment, and individual pre - Olympic rank all appeared as additional contributors to the importance of personal best times. Training methods, both physical and mental, have always been key factors in performance at any level, especially at the Olympic level. Yet there were many Olympic finalists not performing near their PBs. There were even two gold medallists from recent Olympics who had a PB dating back further than four years: Krisztina Egerszegi in 1991 and Pablo Morales (1992 winner in the 100 fly) in 1986. Egerszegi won her third consecutive Olympic title in the 200 backstroke, and Morales held the world record in the 100 Fly from 1986 to 1995. Both these swimmers won without a personal best. A look at the 1996 and 1992 Olympics showed 21 of 52 gold medals were won without a PB (40%). 68 of 156 gold, silver and bronze medals were won without a PB (43%). Swimming fast is one thing, but swimming fast enough to win at the Olympics is another.
What is the easy conclusion? At the Olympics, personal bests don't happen as often as I suspected. Each swimmer is on a different improvement curve, arriving at the world top 8 level in time for the Olympics. Jeff Rouse of the USA has been at the gold medal level in the 100 Backstroke since 1990! That is not a curve but a six year plateau. Agnes Kovacs of Hungary went from a rank of 100 in the 200 breaststroke in 1995 to the silver medal in 1996, a steep curve indeed. Let's have a look at some other variables.
In Atlanta, 35% of finalists swam to a PB, and just over half of all medalists achieved a best time. The rest of the field simply repeated a performance from the past. Winning was not about how fast they swam, but whom they beat. It helps when the world level does not improve to any great degree, as was often the case this year. The women's 200 backstroke is a good example. The bronze medal time has not improved since Egerszegi won her first title in 1988. The men's 100 fly is the opposite example. Every swimmer in the race achieved a PB. A previous best time here was of no value. Even the great Pablo Morales would have finished third with his best time. Why do some events improve while others plateau?
In 1992, 25% of finalists needed a PB to qualify, while 56% of medalists were better than ever before. These are similar results to 1996. Personal bests certainly help in the year of the Olympics, but a previous PB at this level has just as good a chance at a final spot. It is possible to predict the time needed to reach the final. Czech statistician Viktor Svoboda does a prediction for each Olympic event based on his own formula. He calculates the time to final and to win. For the win time he gives a prediction, a high and a low. This year he was almost 1% fast on his predictions. Swimming to at least his final times predictions would have made the top 8 every time.
I am using percent improvements, rather than time, to compare between events and gender more easily.
At age 27, British swimmer Sarah Hardcastle qualified for finals in the 800 freestyle in Atlanta. She swam 1.8% slower than her PB, from 1986! Had she equalled her PB, she would have won the event by 3 seconds. Janet Evans, the current 800 freestyle world record holder (1989) has a similar story. She finished sixth this year in a time 4.3% slower than her PB. At the other end of the curve, young Agnes Kovacs won the silver in the 200 breaststroke (PB). At the age of 14, she improved 6.7% in one year! Youth alone does not determine the probability of a PB. Sandra Volker, a 22 - year - old from Germany, improved 3.6% in the 100 freestyle to win the silver medal. In the same race, Angel Martino, 29, of the USA, swam to a bronze medal with a time just off her 1992 PB. This measure applies only in how the swimmer arrives at the Olympic final level. The chances of medalling are similar, regardless of the improvement curve.
Sandra Volker, GER, improved 3.6% since 1995 to win silver in 1996
For larger 64k photo click on image. Photo © Marco Chiesa
Canadian Stephen Clarke was ranked 11th in the 100 freestyle coming into Atlanta. At the meet, he won the second seeded heat in a PB of 50.14. This time would have placed him 5th on the pre - Olympic ranking. Normally he could expect to final with this performance. Clarke missed the final by 0.35 seconds to finish in 11th place! This means the event was very competitive. Five of the top 8 seeds did not make it through to finals. From the 1996 rank list last month, you can see how many of the top 8 are missing from finals. The men's backstrokes and the sprint freestyles are all missing 5 of the top 8 pre - Olympic qualifiers! Another SEVEN men's and women's events are missing half of the top 8 swimmers. In 1992, FIVE events were also missing at least half of the top 8. There are few guarantees in any event. Each Olympics appears to become more competitive regardless of the times. The pre- Olympic rank seems to matter only slightly.
Stephen Clarke, CAN, had PBs in three swims in two strokes
For larger 64k photo click on image. Photo © Marco Chiesa
How does age factor into performance? Remember the average age of women finalists in 1996 was 21.5 years, in 1992 it was 20 years; while in 1988 it was 19.3 years. This is a definite trend. Why is the average age rising? Is money keeping swimmers in the sport longer ? Are other opportunities allowing women to stay in longer? Based on the large number of women training in other countries, the opportunities must be rising for "older," post - university age women. Training centers could also allow greater numbers of swimmers to continue training at a higher level while also working for a living. This would apply to both men and women. There are still as many "juniors" (18 and unders) making finals, and the age spread (oldest to youngest) is as large.
The percent of juniors in the women's finals changed drastically from 1988 to 1992. Almost half the women's field in 1988 were juniors. This figure fell to 27% in both 1992 and 1996. Age may not be a key factor in performance, but the demographics are definitely changing. "Senior" finalists performed as many PBs as did the juniors. The 1996 women's meet was slower and the field was older. The reasons for this are still out there. The men's events show no difference between 1996 and 1992; the average age was the same for both meets.
Comparing age to percent improvement, I looked at every swim, from the previous PB to the current best. I also went back looking for large jumps (greater than 3%) sometime in the previous six years. There are no clear differences by age. Some swimmers improved in a gradual curve with 0.5% to 2.0% improvement steps. Others improved in a single spike improvement of over 3%. Some had their "spike" up to four years ago and maintained performance into 1996; some "spiked" in the 1995 - 96 season, timing the Olympics perfectly. Still others had a great meet in 1992 in Barcelona, only to slow down for a couple of years, rebounding right on the money in 1996. Jeff Rouse is a good example of this curve.
In 1991, Franziska Van Almsick, at 13, improved over 3% in her 200 freestyle. In 1992 she improved another 3% to win the silver medal at the Olympics. Inexperience cost her the gold, but still, she came out of nowhere. The spread between the slowest world ranked time in any event (150th) and 1st is usually between 6 - 7%. Any national finalist improving above 3% for 2 years in a row should be right near the top! It sounds easy. How is it done?
In 1987, at the age of 13, Krisztina Egerszegi improved 5.9% in the 100 backstroke to reach the world level. In 1988, she improved 1.8% at the age of 14 to win the silver at the Olympics. Michelle Smith, 27, won three events in 1996. She was ranked 2nd in 1995 and 17th in 1994. Prior to 1996, Smith did not rank in the 400 freestyle. No comparison can be made, but she did improve considerably. Is Smith too old to be improving so much? Her improvements in the 400 IM for the past three years were; 1.3% (1995 - 96), 1.7% (1994 - 95), and 3.1% (1993 - 94), for a three year improvement of 6%. This is not a steep curve by any measure. To compare, here is a list of swimmers from 1996 who had improvement over 3% sometime in the past 6 years:
|Michelle Smith, IRL||24||400 I.M.||93-94||4:47.89||3.12|
|Dagmar Hase, GER||23||400 Free||91-92||4:07.18||3.78|
|Samantha Riley, AUS||22||200 Breast||93-94||2.25.53||4.36|
|Sandra Volker, GER||22||100 Free||95-96||54.88||3.68|
|Sandra Volker, GER||22||50 Free||95-96||25.14||3.49|
|Penelope Heyns, RSA||21||200 Breast||94-95||2.26.98||3.69|
|Penelope Heyns, RSA||20||100 Breast||93-94||1.09.79||4.65|
|Vera Lischka, AUT||20||100 Breast||95-96||1.09.24||3.88|
|Minouche Smit, NED||19||200 IM||94-95||2.17.66||3.94|
|Christin Petelski, CAN||18||200 Breast||93-94||2.36.34||5.95|
|Hitomi Kashima, JPN||16||100 Fly||95-96||59.43||3.27|
|Cristina Teuscher, USA||15||200 Free||93-94||1.59.71||4.28|
|Agnes Kovacs, HUN||15||200 Breast||95-96||2.25.57||6.70|
|Brooke Bennett, USA||14||800 Free||93-94||8.31.30||4.65|
|F. Van Almsick, GER||14||100 Free||91-92||54.94||3.49|
|Brendan DeDekind, USA||20||50 Free||95-96||22.59||5.04|
|Danyon Loader, NZL||18||400 Free||93-94||3.48.62||3.81|
|P. V.d.Hoogenband, NED||17||200 Free||94-95||1.48.78||3.53|
|Nessier Bent, CUB||19||100 Back||95-96||56.83||3.51|
|Pavlo Khnykin, UKR||21||100 Fly||90-91||54.43||3.39|
|Khynikin was world ranked freestyler since 1988|
|Dennis Silantiev, UKR||18||200 Fly||93-94||1.59.81||4.16|
|Silantiev does not show up on either 1993 or 1992; % improvement is from 150 th for 1994|
|Note: age is for the year of measured improvement|
All these swimmers made finals in Atlanta and they all showed over 3% improvement sometime in the last five years. There were some examples of swimmers who showed up on the world charts the year before the Olympics (1995) with a time fast enough to final in 1996, yet are nowhere to be seen in the 1994 rankings. They tended to be younger swimmers. All men who improved better than 3%, were under 22 years old.
I looked at the results from the 1992 Olympics for similar improvement rates.
|Dagmar Hase, GER||23||400 Free||4:07.18||3.78|
|N. Mersheryakova, EUN||20||50 Free||25.47||4.49|
|Susan O'Neill, AUS||19||200 Fly||2:09.03||4.40|
|Kyoko Iwasaki, JPN||14||200 Breast||2.26.65||7.96|
|F. Van Almsick, GER||14||100 Free||54.94||3.49|
|F. Van Almsick, GER||14||100 Fly||1.00.02||3.82|
|Alexander Popov, EUN||21||50 Free||21.91||3.80|
|Mark Foster, GBR||22||50 Free||22.52||3.00|
|Danyon Loader, NZL||16||100 Fly||1.57.93||3.60|
This is a list of many of the best swimmers in the world. It is easy to see that improvement rates can easily top 3% in a single season, regardless of age. Every improvement curve may not include a 3% spike at the international level. However, you should find more than one 3+% improvement season, however far back, for every swimmer in the finals.
Getting back to the idea of personal bests as a performance indicator, I have to admit there is more analysis to be done. There are so many factors involved. The stress of the Olympic Games is very large. Those who are able to control their focus and somehow step outside of the "hot zone" while they compete, may be the ones who perform the best.
Getting a performance to the Olympic Top 8 level as early in a career as possible is probably a good bet. Regardless of the shape of the curve, you will need to get there at some point.
Maintaining a world performance level is difficult, but no more difficult than performing a personal best at the Olympics.