5 comments on “A New Way to Look at Break Points

  1. Hello,
    I enjoyed reading your thoughts on break points, but I feel there are several issues that I would like to mention, although by no means am I certain that I’m right, simply posing a response to your theory.
    You argued that 1/1 break ratio and 1/4 in the same game is the same in the end because it achieves the same result. This is true if we were simply going to look at the result for that game, but a long term look at the stats makes a big difference. It’s why casinos make money instead of going out of business. They look at stats in the long run. If I want to be an effective returner, it would be helpful to know on average how many break points I would need before I could achieve the break. It tells me how well I play big points. It should not be the defining statistic in determining a returners effectiveness, but like every statistic it has it’s place and usefulness.
    The other issue I have is not the usefulness of EBP, but rather it is not measuring a player as a returner, but a player as both a server and a returner. I don’t believe it’s fair to discount break point percentage (measuring player as returner) because EBP measures more. Your EBP is a fascinating number and concept but instead of being a replacement for break point percentage, it would be an incredible complement. If I have a player I would take the break point percentage as a measure of returners effectiveness, and then use EBP to determine the effectiveness Of holding serve after breaking.
    Again, I think your EBP is a great concept and I will begin to use it in my own coaching, but I am not willing to throw out break point percentage and ratio. Raw numbers tell us only facts. Wisdom allows us use them to our advantage. I would rather have as many numbers as I can at my disposal. Thanks for sharing your theory and I would appreciate your thoughts on my thoughts

    • First off, thank you very much for the well thought out comment and encouraging words.

      I’ve come to rethink a few things since writing this post, and I tend to agree with most of your points.

      I’m still a proponent of the idea that number of break points in a game is mostly irrelevant and that it comes down to whether you convert or not. I guess a better way to look at it would be as ‘break games’ instead of break points. That way you still get the long term stats; how many games with break opportunities, on average, do you need to finally convert that break. Points alone still do more to skew that, in my opinion. Especially if you want to look at skill in playing ‘big points’. Say you’re in a tight back and forth deuce game where you have six or seven deuces and four break point opportunities. Converting on that fourth break point and winning that game is a huge win, and means you did well in playing the big points, just to give yourself an opportunity to win that game. It would misrepresent that game to say you had a 25% conversion percentage (which is relatively poor), when you still did well to win the game.

      I also agree that as I continued to think about this stat, it became a bit more than just return effectiveness. I definitely think it serves as a good complement to other break related statistics, and perhaps its not as useful as I had originally imagined on its own. It’s true that if you had a break point percentage and an EBP side by side, you’d likely be able to glean significantly more about player performance than you would with either of those on their own. As I mentioned in the original post, I’d love to run a regression and see if EBP was any more revealing about who wins the match than traditional break point percentage and other similar stats. I think it may be, but that may be because more than just returning is involved in the number.

      Thanks for reading and I hope you’ll come back to check out future posts!

  2. Hello sir,
    As I look back at what you and I have written, I think we both have fallen into the deceptive (not with a bad connotation) trap of what I would call being good coaches:). What I mean by that is we have entered the realm of taking quantitative facts and trying to explain their usefulness with subjective perspectives. It is in doing this that neither of us is rght or wrong, but rather we are taking the numbers and using them to our advantage. There is nothing wrong with that I think, but it is a matter of perspective. You are very correct in that winning the 4th break point in a long game I major mental victory, and could just as easily show the player to be effective at playing big points as it would showing them being ineffective. I think our difference is that my focus is not in one game but on the long term. A 25% break ratio in one game is in actuality a 100% games broken ratio for that one game as you very correctly suggested. My concern in the long run is really what the flip side of that stat tells us: that the opponents have 75% chance of getting another chance to win the game, and I will again use the example of the casinos as one of the reasons I am hesitant to throw out this stat.
    After a bit of thinking, I would then propose this: adding another statistic (haha). My theory would be that we would first need to determine ratio of break points to total return points played. In other words, getting your first break point in the second deuce is different than at 0-40. If we could determine then the average number of points to get a break point ( I.e. 1 break point every 9 points) we would have the average frequency of break points. Then, we determine the break point conversion ratio, followed by, but used in conjunction with, the break games won to break games played ratio, followed by EBP. These numbers would (after you recover the dizzying vertigo of staring at numbers), give us a better and more quantitative foundation from which we could then begin applying subjective feedback to the player.

    • I definitely like your idea of looking at average frequency of break points. That’s a very useful stat, and I may go see who the current leaders on the tour are in such a stat. Should be interesting.

      I’d love to create something that incorporated all these numbers, but at the risk of making things more muddled I think I’ll just leave it as is for now.

      Thanks for challenging me on some of these points and expanding my thoughts about the subject. That was my main goal of this blog (to learn), so I’m glad it’s working so far.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s