2011 US Open men's tennis champion. Can I predict it?
It's all in the numbers. The more the numbers, the better. The more fiddling with them, the fancier. Fanciness will lead to concealment. From there the plausibilities are endless. Of course, I joke: I am eager for you to understand. To that end, I hope to make these crazy charts understandable.
To get at the numbers and what I'll conclude from them, it will take some explaining first. Hold tight; it'll all make sense and lead to a thrilling conclusion.
I describe the performance of Roger Federer, Rafael Nadal and Novak Djokovic over the period 2008 to the middle of 2011. The three men have been selected because together they have held the number one ranking in men's professional tennis from 2004 until today; by 'today' I mean this very morning. Roger himself maintained the equivalent of this 'Superbowl champion' status -- that's what an achievement #1 is -- for five years running.
In the charts below I record results in the four 'majors' over the years, 2008 through the middle of 2011. For a player to enter and to be seeded properly depends upon points he acquires in the year dated from the last day of the prior November. The players that play in the four 'major' tournaments are selected by the total number of points listed on the ladder or by special selection (top amateurs, for example).
Here are the major tournaments' names and locations (plus the headers I'll be using). In calendar order: the Australian open (Melbourne, Australia - AUS), the French Open (Paris - FRE), the Championships at Wimbledon -- really, please -- Wimbledon (London -WIMBL, WIMBLE) and finally the US Open (New York -- USOP).
<September 19, 2011 -- While the tournament has passed, the mechanism I developed and the predictions were already written; so, just to show the method, I reproduce here the results and comparisons to show how well the what was predicted. >
We would conclude from the numbers that Federer is a long shot among the three. Further, that if Nadal and Djokovic play the final the match will be ultra-tight and a toss-up. What actually happened, that Federer took Djokovic to two Federer match points before being overwhelmed after Djokovic finally relaxed (at a joke he himself made that brought much merriment and applause from the crowd).
Therefor the 4.3 versus 2.3 matchup came at one point to a razor's edge.
The final pitting Nadal against the #1 Djokovic was a blister-barrage of Novak over Rafael, a straight sets victory never in question over the three sets. A four of seven set match, taking no account of fatigue, might have gone to Nadal, as he was conceding less and less each set (6-1, 6-2, 6-4); however, this scenario takes no account of the distress he'd experience over how hard it was to win service games.
My poor prediction came down to a couple of errors. I gave Nadal a point in intangibles, for his likely extra drive to regain the number #1 ranking. This was a thin rationalization in that this drive was clouded over by a spoken pessimism of Nadal's I can reproduce with some accuracy, "I have not year found a solution for Djokovic." This, based on five straight defeats. Roger had been one of the only two players on tour to have beaten Novak this last year.
I ommited one more measure: to account for the chance of the field to win. Using an arbitrary total of 12.0 for winning for all players, let's complete the arithmetic. Federer (4.3), Nadal (3.4), Djokovic (2.4)
account for 7.0 of the probability. Tsonga, Soderling, Isner and the rest of the field together would stand a 5.0 chance.
Interestingly, yet to be proven or even examined, to reduce each of our three players numbers by 2.0 yields a number corresponding to the coding system): Federer (2.3 = semifinalisist), Nadal (1.4 =finalist ), and Djokovic (0.4 = winner). How this really stands up to the method might just be an anecdote. But??
That's it. Enjoy the charts.
The chance of each player is expressed in a prediction number given in decimals to one place, a lower number showing a better chance.
From the charts below, the chance of Roger Federer is 4.3. Rafael Nadal and Novak Djokovic emerge with only the barest difference in likelihood to win (2.4 to 2.3). The chief determinants between the two are in reductions and US Opens with Nadal scoring a 1 for US Opens won and Djokovic having 3 points for reductions to Nadal's with 1 point.
Abbreviations used below:
yr-maj = total for all scores for the majors for one of the years in the timeframe
maj-tot= of scores for the majors for a given year, the average
avg-major = average of all the scores for a specific major over the four-year span
tot-avgs = sum of averages for all majors over the four-years
factor = amount by which avg-major (avg of majors' score over 4 years) will be multiplied
tot-wtd = finally, the sum of all weights derived by using (factors times yearly weights) minus the number of US Opens won over period)
-->
-=-=-
| Player | |||||
| FEDERER | |||||
| Majors | |||||
| AUS | FREN | WIMBL | USOP | yr-maj | |
| Yrs | |||||
| S 3 | F 1 | F 1 | W 0 | 5 | |
| 2009 | F 1 | W 0 | W 0 | F 1 | 2 |
| 2010 | W 0 | Q 5 | Q 5 | S 3 | 13 |
| 2011 | S 3 | F 1 | Q 5 | prediction^ | 9 |
| ` | |||||
| maj-tot | 5 | 7 | 11 | 4 | 29 |
| yr 2008 | avg-maj | 1.25 | minmzg 2008 | factor 1/5 | .25 |
| yr 2009 | .5 | lessng 2009 | 1/3 | .17 | |
| yr 2010 | 3.2 | cutting 2010 | 1/2 | 1.6 | |
| yr 2011 | 3 | leaving same | 1/1 | 3 | |
| tot-avgs | 7.25 | tot-wtd | 5.29 | ||
| minus Opens | 1 | ||||
| 4.3 | |||||
| maj-tot | 29 | ||||
| yr-avg | 7.3 | ||||
| total-wtd | 4.3 | ||||
| reductions* | reductions* | 0 | |||
| momentum | 0 | ||||
| intangibles | 0 | Prediction^ | 4.3^ | ||
-=-=- |
| Player | |||||
| Raf NADAL | |||||
| Majors | |||||
| AUS | FRE | WIMBL | USOP | yr-maj | |
| Yrs | |||||
| 2008 | S 3 | W 0 | W 0 | S 3 | 6 |
| 2009 | W 0 | 10 | --- | S 3 | 13 |
| 2010 | Q 5 | W 0 | W 0 | W 0 | 5 |
| 2011 | Q 5 | W 0 | F 1 | prediction^ | 6 |
| By Major: | 13 | 10 | 1 | 6 | |
| yr 2008 | > | 1.5 | minmzg 2008 | factor 1/5 | .3 |
| yr 2009 | 4.2 | lessng 2009 | 1/3 | 1.43 | |
| yr 2910 | 1.25 | cutting 2010 | 1/2 | .63 | |
| yr 2011 | 2 | leaving same | 1 | 2 | |
| avg | 7.25 | tot-wtd | |||
| 4.36 | |||||
| minus Opens | 1 | ||||
| 3.4 | |||||
| maj-tot | 29 | ||||
| yr-avg | 7.25 | ||||
| total-wtd | 3.4 | ||||
| reductions* | reductions* | 1 | |||
| momentum | 0 | ||||
| intangibles | 1 | Prediction^ | 2.4^ | ||
-=-=-
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
