Tuesday 10 May 2011

Guest Post: Streaks and Rankings

During the last few weeks, all we heard about were the different streaks, present and past. It seemed that each day, the ATP decided on different numbers to represent Lendl's and Borg's records from the 1980s.

The next few weeks will surely feature a lot of talk about rankings and points. With Rome and Madrid switching places on the calendar, figuring out the next rankings might give anyone an headache (and that's before you consider ATP's elaborate rules for them).

So, in order to bring some order into this mess, I'm delighted to present this guest post by Yolita, who knows the numbers much better than all of the tennis journalists and ATP's stats people combined ;) You can follow her on twitter - @Yolitatennis.

Anna.

******************************************************************************************
I think Novak's achievements over the last 5 months surpass anything even the most die-hard Nole fan would have dared to hope. We've been living in a dream and when we wake up (because wake-up we will) we'll have memories to last us a lifetime. But not only that, after so many records by Roger and Rafa, one would have thought that there was no hope of new records being achieved. Not so. One of the most amazing things about Novak's season is his winning streak, which started last December, when he won two Davis Cup matches for Serbia. Just how good is this winning streak of Novak's? It should be straightforward, but it isn't, apparently.

I hate it when a rumour becomes the truth by virtue of repetition. I got tired of hearing the girl commenting the match between Novak and Guillermo Garcia-Lopez saying that Nole was about to tie Lendl for the third best start of the season. He wasn't. He had already passed Lendl. :roll: :roll:

It's not difficult to understand. In the Northern Hemisphere, an academic year is not the same as a calendar year, right? Why? Because the academic year starts in August, September or October.

In the same way, before 1987, the tennis season didn't start at the beginning of the year. The end-of-season championships (known then as Masters Cup and now as World Tour Finals) were played in January of the following year. And the season started later. So start-of-year is not the same as start-of-season. Not so difficult, right?

I have exchanged e-mails with Greg Sharko (a very nice man, by the way), and he told me that a decision had been made to count all matches played in January, even if they were the end-of-season championships for the previous season, as belonging to the next season. I'm, sorry, you can have your own opinions, but you can't have your own facts. You cannot "decide" that matches from one season really belong to another.There's nothing wrong with having 2 records: start-of-year and start-of-season. You can also have winning streaks in general, starting at any point in the year. So we have three different records relating to winning streaks.


BEST START-OF-YEAR WINNING STREAKS IN THE OPEN ERA
1. John McEnroe 42 (1984)
2. Novak Djokovic 32 (2011)
3. Bjorn Borg 31 (1980)
4. Ivan Lendl 29 (1986)
5. John McEnroe 23 (1985)
6. Pete Sampras 17 (1997)
7. Roger Federer 16 (2006)
8. Andre Agassi 15 (1995)
8. Stefan Edberg 15 (1987)

BEST START-OF-SEASON WINNING STREAKS IN THE OPEN ERA
1. John McEnroe 39 (1984)
2. Novak Djokovic 32 (2011)
3. Bjorn Borg 26 (1980)
4. Ivan Lendl 25 (1986)
5. John McEnroe 20 (1985)
6. Pete Sampras 17 (1997)
7. Roger Federer 16 (2006)
8. Andre Agassi 15 (1995)
8. Stefan Edberg 15 (1987)


Starts of season including the Australian Open.
This is relevant because before 1987 the Australian Open took place in November, so it was easier to get a winning streak to start the season, as can be readily seen. Novak has the longest winning streak to start a season since the Australian Open was incorporated, with 88.23% advantage over 2nd place Sampras and 100% advantage over 3rd place Federer. I think that looking at the streaks in red, and seeing the names involved, puts Novak's streak in context. It's beyond amazing.  :o :)

LONGEST WINNING STREAKS IN THE OPEN ERA
1. Guillermo Vilas 46 (1977)
2. Ivan Lendl 44 (1981-82)
3. John McEnroe 42 (1984)
4. Roger Federer 41 (2006-7)
5. Bjorn Borg 38 (1979-80)
6. Thomas Muster 35 (1995)
6. Roger Federer 35 (2005)
8. Novak Djokovic 34 (2010-2011)
9. Rafael Nadal 32 (2008)

As you can see, at this point in time the difference between start-of-year and start-of-season is irrelevant. The list of players at the top is the same in both. But it may become relevant if Nole's streak ends at 40! Or for future players.

So, with this season, Novak has already made sure his name is in the record books. But there is something else happening. Since January 2004, when Federer won the Australian open, the #1 player has been either Federer or Nadal. For the first time in over 7 years, there is a real possibility of that changing. Many people are confused about how the ranking system works. There are many little rules that I won't go into now. What's important to know is that on May 16th. the players will drop the points they earned in Madrid 2010 and add the points they earn in Rome 2011. Below you see the Top 10 players, with the points they have on Monday May 9th., the points they'll drop and the minimum and maximum points they'll have on May 16th., depending on their performance in Rome 2011.


Rankings as of 9 - May - 2011

Heading to Rome

Current points.
Points to be dropped on May 16.
Net starting points.
Interval of possibilities at the end of the Rome tournament.
Possible rankings after Rome.


1. Nadal 12470 1000 11470 [11470,12470] 1-2
2. Djokovic 10665 0 10665 [10665,11665] 1-2

3. Federer 8900 600 8300 [8300,9300] 3

4. Murray 5905 180 5725 [5725,6725] 4-5
5. Soderling 5265 10 5255 [5255,6255] 4-5

6. Ferrer 4420 360 4060 [4060, 5060] 6-7
7. Berdych 4035 0 4035 [4035,5035] 6-7

8. Melzer 3020 180 2840 [2840,3840] 8-10+
9. Monfils 2645 180 2465 [2465,3465] 8-10+
10. Almagro 2495 360 2135 [2135,3135] 8-10+


Remarks:
If Nole wins Rome and Rafa loses before the semis, Nole will take the #1 spot on Monday May 16th. :o :o
If that happens, Nole will have 11665 points and Rafa at most 11470+180=11650.
I know it's a long shot and the probability of Rafa losing before the semis on clay is very small indeed, but it's mathematically possible.
Remember also that even if Nole doesn't get the #1 spot, he'll get closer to Rafa, and Rafa defends 2000 points in Roland Garros, to Nole's 360. A much bigger chance there.
We're living in interesting times. :)

No comments:

Post a Comment