An Overview of Ratings and Rankings
by Dallas Oliver, 5 October 2015
|Share: || || |
As a junior tennis website that rates and ranks tennis players, one of the most common questions we field at Tennis Recruiting is the following: What is the difference between ratings and rankings? Which one is better?
Many people in the tennis community want to draw fundamental distinctions between ratings and rankings. But there are no such distinctions. In the end, ratings and rankings are closely related - and the same systems can be used to provide both ratings and rankings.
This article explores ratings and rankings - their similarities and differences - and shows how they can be best used to answer different questions.
Rating systems are common throughout all walks of life. For example, movies are rated (G, PG, PG-13, R) to classify their content. Companies rate cars for metrics like performance and safety.
In a tennis context, ratings are values assigned to players for purposes of comparison against (1) some objective measure of skill level and/or (2) other players. Ratings can be used for level-based play - identifying players of similar skill levels to foster competitive and enjoyable matches.
Examples of tennis ratings include:
The values assigned by some rating systems have larger meanings. Both the Dynamic NTRP and UTR systems group players into skill levels based on their results. The TRN Star Ratings (boys, girls) group players by graduation year, while the TRN Power Ratings have properties that allow for calculating match win probabilities.
Rankings are ordered lists of items such that two items can be directly compared: one item can be ranked higher, lower, or the same as another item. Popular examples of rankings are the US News rankings of American colleges and the Popular Mechanics rankings of cars and trucks.
All rankings are based on ratings - ordering items by rating from top to bottom. In the above examples, US News and Popular Mechanics first evaluate each college and car using a set of criteria to come up with overall rating scores. The final rankings are mere orderings of all the items by rating score.
Tennis rankings are no different. Every ranking starts out with a rating - and every rating can be used to produce rankings. For example, prior to the US Open, we recorded every junior player's TRN Power Rating, UTR, and ITF Points total. From these numbers, we produced rankings of the players in the US Open. These rankings are simple orderings of the rating values.
Rankings can also use different filtering criteria. At Tennis Recruiting, our Power Ratings use all match results to come up with ratings for all players of the same gender. These Power Ratings can be used to compare any two players - regardless of age, graduation year, or nationality. We then use these Power Ratings to come up with many different rank lists. For example, we filter out each of the seven graduation years to come up with rankings for seniors, juniors, sophomores, and so on. These lists filter out players in different graduation years - and they also filter out players from other countries and players who have turned professional.
Examples of tennis rankings include:
- USTA and ITF junior rankings are based on points assigned to players in leveled tournaments. The points players earn can be viewed as ratings - and the rankings order players by their point totals.
- The Tennis Recruiting Class Rankings (boys, girls) are based on players' Power Ratings.
It's Not Ratings vs. Rankings - It's Ratings AND Rankings
As we explored in the previous sections, ratings and rankings are different ways of looking at the same data. Both are needed and necessary. Depending on the question you are asking, it will be more useful to use a rating or a ranking.
Here are some common questions for which there are clear advantages to using rankings or ratings...
(1) Will a match between these two players be competitive? How much better am I than that player?
If you are trying to compare the ability difference between two players, then you want a rating. UTR differences of less than one point are expected to be competitive. The TRN Power Ratings can give you expected win percentages for each player. On the other hand, knowing that one player is ranked No. 100 and another player is No. 150 would not be very helpful for this question.
(2) Where will I fit into a college lineup?
Ratings are more useful here. By comparing your UTR, TRN Power Rating, or TRN Star Rating with players on the roster, you can project where you might play your freshman year and beyond.
(3) My top choice college is only recruiting one player for next year. What are my chances?
Although it's good to know how you fit in with existing players on a roster, coaches are always trying to recruit the best players among who are available and interested. Even though the existing roster may be full of 3-Star recruits, a coach will try to land a 4- or 5-Star recruit if available. So knowing how you compare to the other players competing for that spot matters. You could compare your rating to all other players, but a rank ordering of the players is most meaningful in this situation.
(4) How should seedings be assigned in a tournament?
In this case, a ranking is needed. When you are trying to identify the Top 8 (or 16 or 32) players in a tournament, you need to order all the players from top to bottom so that you can assign seed numbers to the top players.
(5) Will I be able to gain entry into this national event?
Again, a ranking is needed. Players need to see where they rank against other players to see if they are high enough on the list to gain entry.
With the above framework in place, here are answers to several questions that we get about our rankings.
Since you rank by graduation year, are you able to compare kids between different classes?
Yes. As we describe above, our system first makes a pass across all
players in our system, rating all players relative to one another regardless of age or graduation year. Only then do we filter out the players by graduation year to come up with our fourteen rank lists of high school and middle school boys
Since you only rank Americans, are you able to evaluate international players?
Yes. Our system uses results from USTA- and ITF-sanctioned tournaments to come up with a much broader set of rankings than we display. Our company caters to an American market, and so we filter out the international players in our public lists. College coaches are able to see how international players fit in to the broader scheme with our Coaching Advantage service.
Isn't it true that rankings discourage play while ratings encourage play?
No, we certainly we don't think so. As we have discussed in this article, ratings and rankings are simply two different cuts of the same data.
How good are your rankings? Do college coaches really use them?
Our company focuses on rankings of college-bound junior tennis players, and we believe our rankings are the best in the business. We have evaluated our rankings against other systems using a number of sectional, national, and international events in a study that you can access by clicking here. As you can see, our rankings provide the highest level of accuracy.
College coaches absolutely use Tennis Recruiting in their initial evaluations of recruits. We have heard many anecdotes from players and families about how coaches had the TennisRecruiting.net website up on their computer screens during recruiting visits - and more than 900 college tennis programs from 48 states and the District of Columbia subscribe to our Coaching Advantage service.
How important is my ranking or rating to a college coach in the recruiting process?
When we talk to college coaches, we find that most of them use ratings and rankings as a starting point in the recruiting process. Coaches typically make preliminary lists based on the rating and ranking tools out there, and they try to use data from the different services available.
But keep in mind that these lists are only a starting point. Ratings and rankings can only capture so much information, and college tennis coaches do not recruit solely on computer lists. Coaches want to get to know their recruits - form close relationships, watch them play, understand whether their goals are compatible, etc. So while rankings have their place in the process, players and families should not get too hung up on them.
I can only see my ranking within my graduation class on the Tennis Recruiting web site. Why can't I see my Power Rating? It would also be nice to compare myself to players in other classes when I am playing a tournament.
We are working on new features at Tennis Recruiting that will give our Recruiting Advantage Members access to head-to-head comparisons and other information based on their Power Ratings. Stay tuned...
In the end, ratings and rankings are two sides of the same coin - they are different ways of looking at the same data. One is not "better" than the other, and both ratings and rankings have their uses.
We hope you have enjoyed this discussion of junior tennis ratings and rankings. Feel free to weigh in with your questions and comments below.
Leave a Comment
More Ranking Articles
Behind the Tennis Recruiting Rankings
This week marks the 569th consecutive week that Tennis Recruiting has
put out graduation-based rankings of American junior boys and girls.
Rankings are front and center on Tennis Recruiting, and people often ask
us how our rankings work. Today we describe our ranking system ...
What Is An Upset?
Tennis Recruiting is a website that rates and ranks junior tennis
players. One of the questions we get most often from our users is,
"What exactly is an upset?" There are many possible
definitions of an upset - this article explores the question and
puts forward an answer.
Revisiting the US Open Forecast
The US Open is the premiere event on American soil, and so Tennis
Recruiting pulled out its heat maps once again to forecast the
tournament. There were lots of upsets (Wait, Serena lost?), but let's
take a look at how our predictions played out...