NSWC Slalom Ranking Explanation for 1997

SLALOM RANKINGS

The algorithm used to compute these rankings is the same as that used in 1996.

Alternative algorithms will be evaluated during 1998. Don't Panic! The algorithms being considered give substantially the same results as the current one; the major difference is that they represent attempts to utilize the difficulty of the race rather than its importance. At high-level races, this makes almost no difference: important races like Team Trials are hard. Where it does make a difference is at mid-level races, which fluctuate in difficulty due to course design, changing water conditions, etc.

So don't change your training or travel plans: the way to move up in rankings will remain unchanged: go to harder races and beat stronger paddlers.

If you are interested in participating in the discussion or development process, then contact me -- my info's below.

Here's a step-by-step explanation of the algorithm used for 1997 rankings. If you don't want to read all this, skip to the end.

0. The data used is that which is available right here, on the 1997 Race Results page.

1. Each race's data is reshuffled into the same format:

Class Name(s) Time-1 Penalty-1 Total-1 Time-2 Penalty-2 Total-2 Better-Score  Total-Score

Obviously, not all of this information is available for every race; fields which are not available from the race results are left blank.

2. The classes for each race are translated from the many names that show up in results to a list of canonical racing classes. Here's an example of part of the translation table used to do this:

	C-1		C-1
	C-1W		C-1W
	C-1W_expert	C-1W
	C-1_(A/B)	C-1
	C-1_A/B		C-1
	C-1_C/D		C-1
	C-1_Cadet	C-1
	C-1_Expert	C-1
	C-1_Jr		C-1

3. Results for classes which don't currently get ranked -- that would be open boats, recreational/plastic/cruiser decked boats, K-2, squirt boats, sit-on-tops, etc.; are dropped.

4. Scores with DNS, DNR, or DNF are interpreted numerically -- this is just used to gather race statistics. Speaking of which, statistics for every race used in rankings are available.

5. If a race was scored with 2 seconds for touches/50 seconds for misses, nothing happens in this pass. But if it was scored with 5 seconds/50 seconds, the penalties are recalculated to the 2/50 system.

6. Every paddler's name is converted to canonical form, which I've hopefully spelled correctly. Here's an example:

	Carleton Goold		Goold, Carleton
	Carleton Gould		Goold, Carleton
	Carlton Gould		Goold, Carleton
	Goold, C		Goold, Carleton
	Goold, Carlton		Goold, Carleton
	Gould, Carleton		Goold, Carleton

7. Every race result is converted into this form:

Class=Name Score Ratio

where "Score" is their combined-run total score for the race, and "Ratio" is the ratio of their score to the best-score-of-the-day. For example, from the Riversport Slalom:

K-1W=Thomas,_Natalie	471.65	1.904
K-1W=Potochny,_Evy	462.75	1.868
K-1W=Gelblat,_Renee	531.46	2.145
K-1W=Hearn,_Cathy	270.49	1.092
K-1W=Weld,_Kara		272.07	1.098
K-1W=Beakes,_Nancy	317.22	1.280

8. The ratio from the previous pass is inverted to give a competitor's race ratio: this number reflects how far off they were from the best-score-of-the-day. (The boat with the best score of the day has a race ratio of 1.0.) Two lookups happen: last year's rank class (A, B, C, D or Unranked), and membership on the national A team. (The reason for this is that the strength-of-field assignment is based on this.)

K-1W=Thomas,_Natalie	0.525	C
K-1W=Potochny,_Evy	0.535	U
K-1W=Gelblat,_Renee	0.466	C
K-1W=Hearn,_Cathy	0.916	ATEAM
K-1W=Weld,_Kara		0.911	ATEAM
K-1W=Beakes,_Nancy	0.781	B

9. Each race result is weighted by the race weight; the race weight is given by

(Field Strength + Importance Factor)
------------------------------------
              20

K-1W=Thomas,_Natalie	0.367
K-1W=Potochny,_Evy	0.374
K-1W=Gelblat,_Renee	0.326
K-1W=Hearn,_Cathy	0.641
K-1W=Weld,_Kara		0.638
K-1W=Beakes,_Nancy	0.547

10. All results from all races are combined. If a paddler has done more than three races, their best (highest) three race weights are selected. If a paddler has done only two races, they're assessed a 5% penalty; if only one race, a 10% penalty. These results are then averaged to give the competitor's rank ratio. Same example as before:

K-1W=Thomas,_Natalie	0.350	Bellefonte,Riversport
K-1W=Potochny,_Evy	0.366	Lehigh,Riversport,Bellefonte
K-1W=Gelblat,_Renee	0.369	Lehigh,Codorus,Farmington
K-1W=Hearn,_Cathy	0.841	Trials-3,Trials-2,Nationals
K-1W=Weld,_Kara		0.824	Trials-3,Trials-2,Nationals
K-1W=Beakes,_Nancy	0.654	Trials-2,Trials-1,NOC-DBH-2

11. The results are sorted by rank ratio and separated by racing class. Each boat's percentile rank within its class is assigned, based on the boat having a value of 100.0. Letter classes (A,B,C,D) are assigned based on percentile rank. Two more lookup tables are used: one to assign agegroups (e.g. Jr, Sr, Ms) and one to mark non-US paddlers.

Here's the table used to decide the assignment of letter classes based on the percentile rank:

"A" Ranked		85% to 100%
"B" Ranked		65% to 84%
"C" Ranked		40% to 64%
"D" Ranked		below 40%

12. That's it. Please note that although all the calculations were done to several decimal places, that does not mean that rankings are accurate to that degree. For example, the difference between a rank ratio of .453 and .456 falls well within the variability of manual timing systems.

Also, please note that paddlers who did well at important races with high field strength numbers may be ranked ahead of paddlers who beat them head-to-head.

For those of you who want the techie details, all the scripts are written in Perl and run on a Unix system.

Summary

best score of race   (field strength + importance factor)
------------------ X ------------------------------------ X penalty
competitor's score                   20