Federer's losses in Grand Slams post-prime compared to Samrpas/Nadal

[McEnroeisanartist]

Federer's losses in Grand Slams post-prime compared to prime Sampras/Nadal

It is rather interesting to note that the average ranking of Federer's opponents that he has lost to since 2012 Wimbledon is higher than the average ranking of the opponents that Sampras and Nadal lost to during their primes.

Since 2012 Wimbledon, 11 Grand Slam losses for Federer to: 7, 3, 8, 116, 22, 1, 17, 2, 16, 46, 9. Average: 22.45

Sampras - prime - 1993 Wimbledon - 1998 Wimbledon, 11 Grand Slam losses to: 7, 23, 2, 24, 40, 7, 13, 65, 16, 20, 97. Average: 28.55


Nadal - prime - 2008 French Open to 2014 French Open, 11 Grand Slam losses to: 6, 25, 6, 4, 7, 2, 1, 1, 100, 135, 8. Average: 26.82

 

[ollinger]

Of course one of the problems with averages is that one outlier piece of data can skew the entire thing in a way that's misleading. good scientists actually LOOK at the data before computing a statistic. In such cases where there's an outlier (in this case, the 135th ranked player) it's useful to look at the median rather than the mean. In this example, the median ranking for Nadal's losses is 6th, for Federer's losses 9th.

 

[Boom-Boom]

It is rather interesting to note that the average ranking of Federer's opponents that he has lost to since 2012 Wimbledon is higher than the average ranking of the opponents that Sampras and Nadal lost to during their primes.

Since 2012 Wimbledon, 11 Grand Slam losses for Federer to: 7, 3, 8, 116, 22, 1, 17, 2, 16, 46, 9. Average: 22.45

Sampras - prime - 1993 Wimbledon - 1998 Wimbledon, 11 Grand Slam losses to: 7, 23, 2, 24, 40, 7, 13, 65, 16, 20, 97. Average: 28.55


Nadal - prime - 2008 French Open to 2014 French Open, 11 Grand Slam losses to: 6, 25, 6, 4, 7, 2, 1, 1, 100, 135, 8. Average: 26.82

Impressive how Goaterer is even more consistent post-prime than Sampdal in their prime :shock:

 

[cknobman]

Of course one of the problems with averages is that one outlier piece of data can skew the entire thing in a way that's misleading. good scientists actually LOOK at the data before computing a statistic. In such cases where there's an outlier (in this case, the 135th ranked player) it's useful to look at the median rather than the mean. In this example, the median ranking for Nadal's losses is 6th, for Federer's losses 9th.

Your computation is even worse than OP's LMAO.

Median is never a good statistical analysis for this type of comparison as it puts no weight on the numbers of either side of the median.

 

[JMR]

What I find curious here is the apparently arbitrary application of the slippery concept of "prime." 2011 is included within Federer's prime even though he won no slams that year, made just one slam final, and finished No. 3. This is a player who had four straight multislam years and spent a longer time at No. 1 than anyone else. Well, perhaps the OP is saying that one slam final is still pretty good. No. 3 is still pretty good.

Yet Nadal's 2007 is not part of his prime even though he won a slam that year, made another slam final, and finished No. 2. I suppose it must remain a mystery.

 

[Bobby Jr]

Your computation is even worse than OP's LMAO.

Median is never a good statistical analysis for this type of comparison as it puts no weight on the numbers of either side of the median.

Yeah, agreed. Maybe an Olympics ice-skating style calculation where you discount the highest and lowest results and then calculate would iron out the kinds and be the best of both worlds. :!:

 

[sdont]

Your computation is even worse than OP's LMAO.

Median is never a good statistical analysis for this type of comparison as it puts no weight on the numbers of either side of the median.

Arithmetic mean is not a good measure.

People should look at geometric or even harmonic mean.

 

[spinovic]

As soon as I saw "post-prime" in the title, my eyes glazed over.