Predictability and Patterns

podcasts May 16, 2020

 

Marc
Welcome to another episode of Crunching the Numbers, you are joined by Marc Sophoulis from The Tennis Menu, myself and obviously our data expert Shane Liyanage from Data Driven Sports Analytics. Thanks again for joining us Shane.

Shane
Hi everyone. It's been nearly two months now in isolation so hopefully everyone's adjusting well.

Marc
Thanks again for being here and today's topic is about predictability. We've spoken about this at length a lot of times and obviously during the Australian Open we spoke about this also with the player we were working with and it's something that we agree to disagree about a lot of the time. The data and the science collide here a little bit and we want to look at: when is the right moment to be predictable and when is the right moment to change it up. But obviously the data tells us one thing, but my coaching opinion tells me another and we're going to have a little bit of a robust discussion about that there's no doubt. You're gonna give me some numbers that again can maybe think a little bit more but I'm gonna give you some thoughts that are gonna make the data look like crap. So let's start with what you've got for us in terms of your research around predictability in the game.

Shane
Yeah, I love this topic because I think usually Marc and I are pretty aligned on most things but predictability is one where we definitely diverged a bit. So, as Marc said, let's explore that and we might actually be a bit closer than we think if we align our definitions I think.

I want to throw a few metrics your way, that actually helped me sort of work out predictability and then I'll give you some tour examples of some patterns used by some of the big players and the aim, I think Marc, is for us to leave united on this, so it'll be interesting to see.

 

Marc

We'll see how we go and see what the listeners actually tell us in their feedback. We've had some great feedback actually Shane, so please keep that coming through. Through The First Serve, through the Tennis Menu, through Data Driven Sports Analytics. Please give us your opinions on any topics that you might want us to cover as well because we quite enjoy crunching the numbers weekly, and this is you know, this is gonna be no doubt one of our finest, robust discussion. So Shane give us some of your metrics that you've been working on over the last week.

Shane
So we're all familiar with point success and that's, you know, the amount of points played and winning it, so that's a simple metric but for me, when you're looking at predictability, you need to look at something called patent robustness so... I'm not going to go into all the details of the mechanics of the calculation, but it really centres around looking at how much a particular pattern for success diminishes if you use it multiple times. So I'll use the example of a serve-volley.So you might win that point. But if you use that point immediately after. Does your opportunity to win the point diminish because you used it recently?

So, the robustness score that the metric comes up with is a score out of 10, with a score of 10 being a really robust pattern that you can use all day, no problems. A score of zero is a pattern that loses value immediately after it being used once. I'll use the example and we gravitate to this one a lot - Nadal’s lefty pattern, where he uses the lefty serve, particularly on the ad side to a right hander's backhand, sets it up with a forehand, goes immediately back to the backhand. Now that has a robust score of 9.2, which really means he can use that all day. It's devastating for his opponents. And really, with the exception of maybe Djokovic, which we'll touch on a bit later that strategy works quite well. And I'll contrast this to Roger Federer SABR strategy. So the sneak attack by Roger. If you're not familiar with it. It's where he moves forward on the return and he ends up with a net really for that plus one return shot. He actually has a fair bit of success - he's winning nearly 53% of points on that, but it's actually got a lower robust score of 4.5, so that's something he can't use in close proximity. So I might start with those two Marc. Any thoughts on that?

Marc
Yeah, I think they're two really good examples to start with. I think the Nadal one is, you know, for me, one of the things that stands out because Nadal’s execution of his ball is exceptional. You know the pattern that he plays, his forehand lasso finish, where it's over the top of his head, allows for a steeper ball path off his racquet to allow the ball to jump higher on the backhand side of the opponent.

Now basically that's a pattern of play, or it's a shot that is very hard to execute as an opponent back no matter what. No matter if you know the ball is going there, no matter if you've seen him do it over and over again, even if he tells you the ball is going to go back in that corner - your ability to execute that shot back is really challenging because of the steepness of the ball bouncing at the other end. I feel that he has great success with that because it buys him time to utilise his forehand on the next shot as well. So he's obviously so well drilled in that pattern of play that it's going to be executed really well no matter if somebody knows it's going there. The other thing is from the serving perspective it's very hard to do much else with a backhand return sliding away from you on that ad side. So the lefty can really get the ball moving, get you out wide and it's very hard to execute much more than what you're probably doing anyway. When you hit that return though, this is what the flip side the challenging thing is, is that once you've gone out wide to play your backhand return from Nadal's serve, the one thing you want to do is you want to quickly get back and recover and come to the other side of the court because you're out of court.

But what Nadal does is he can mix it up really well and go the other way as well so you don't want to sit in that corner and just wait for him to play that that ball back in the backhand corner. So you try and recover and then you get caught off guard so then your second shot becomes defensive instead of offensive. So that's the Nadal pattern of play I think broken down a little bit from an opponent's perspective.

So that's that's one example, the Federer SABR is the other one that you brought up. I believe, and this is just a belief and opinion - I'm no numbers guru Shane. Federer does this purely because he tries to put pressure on. Now, he doesn't expect to always win the point, but it's a changeup and it puts pressure on and what I feel he does it for is to see what the opponent will do if he does it down the track at a later moment in time. So you might do two in the same game like you're saying and have a small percentage of chance of winning that point, but looking further down the track, he's exposed and seen what the opponent can and can't do off the SABR. He says okay well, if at a 30-30 point, or at 5-5, if I SABR, which side do I SABR to? Do I hit it to the forehand side? To the backhand side? Do I go straight up the middle? Do I try and get myself in and cover a certain particular spot? So I think he does it to look further down the track as to what it can and can't do. So they're my two little examples of how I may look at from a coaching perspective.

 

Shane

Yeah, yeah. So, the reality though Marc is not a lot of players actually have a pattern or strategy that they've got a really high robust score. It's a bit lower than Nadal’s 9.2 for instance. So how do we make sure that we are not giving up our best strategy so the opponent can sit on it? But still, being able to use it when we really need it?

Marc
Yeah, it's a really good question. It's something that I try and impart on the players a lot. So, what I try and get the players to look at is at a big moment, where it 30-30, deuce, break-point down, game point up - play the pattern that you most feel comfortable doing. What can you execute and nine, nine and a half times out of 10, and you know that your side of the game will not break down. If that's the serve out wide and the forehand back behind them, so be it. If the opponent knows it's going to happen, it's fine. Like Nadal, if you could execute it well enough to put pressure on, then to me that's a really good strategy because you need to be comfortable enough to make the play at the big moments.

On the flip side, I give my players something like - if you're two points up in the game, if you're 30-0 up, 40-15 up, 40-0 up for example, you can change that pattern of play to keep them guessing. Right, so it gives the player the confidence to say I'm 30-0 up, I'm okay sliding the ball wide now instead of going down the tee. So it gives them a little bit of an opportunity where they're not worried, if they lose that point they've still got another point up their sleeve to be ahead. So that's the kind of strategy and game planning I look at with a lot of players at the top level and even at junior level - is when you don't feel as confident, play whatever you feel the most comfortable. And then that takes away the stress of, will I make this a wont I make this. Will I make this first serve or am I going to miss this first serve because I'm trying something that I don't normally do at this moment. And that just gives the player a comfort of - I know I've executed this many times before, and I will be able to do at this big moment.


Shane
Yeah, okay. So I mean, I know when we've actually gone into matches and one of the things that I usually do in my match preview or preparation pack for you is I'll often go and say, don't be too predictable on this, don't be too predictable on that and you're like, but that's his strength, that's their strength I want them to use that. So, I suppose that's something we've grappled with, but one thing I think we are sort of in agreement with is that in those under pressure moments, we do want to be using that strength. More often than not, so...

Marc
I think using your strength is a real key component of winning matches, no doubt. So in the big moments, I think it's important even if it is predictable. The other thing is obviously you go in with a game plan of which side of the player you want to try and target the most. You know whether it's the forehand or the backhand, and that can come into play in the big moments. If you've got a player that can execute any serve, which hopefully from a coaching perspective you have, and you've been able to produce that and create that in your athlete, then it's okay to change your most comfortable pattern if you know that they're comfortable serving to any part of the court or returning to any part of the court. So you know there's obviously taking the individual into account which is really important whenever you're game planning. So data and the art and science are one thing, but it's the players feeling that's the most important part in any kind of decision that you have going forward, as to where to play the ball.

Shane
So let's go back to one of those sort of patterns and strategies that actually brings a lot of one off sort of success for a lot of players and that's a volley. It's not not necessarily a robust strategy for them, but how do you sort of coach your players to maybe use the serve volley as a surprise tactic?

Marc
Well, it's a good question because I feel like serve and volley was one of the four game styles that we have grown up with and that we think is part of the game, but I feel like serve volleying now has become a tactic rather than a game style. So you know, you use it as the surprise element, you throw it in and that's the kind of thing that you throw in it 30-0 up, at 40-0 up, at 40-15 up as a surprise tacticical element, to be able to keep the opponent guessing as to what you're going to do. Now you don't throw in once and never do it again, it might be something that you throw in each time that you're a couple of points up to keep that opponent guessing a little bit more.

Serve and volley is... the one thing we need to understand about serve and volley is it's a tactic that needs to be practiced as well. So it's not like you go into a match, a player is a baseliner and you say to them - just every now and again throw in a serve volley if you haven't practised it, because you still want them to have success at the game plan. So you know I think it's something that is a practiced skill, it's an art, you know, the ability to serve to the spot as you're moving forward and hit that first volley from on or inside the service line. To be able to close the net, to be able to split step, to be able to read the ball coming back. There all really key elements that need to be taught and not just thrown into a game plan for the sake of throwing it in as a changeup.

Shane
And I think the other thing that we sometimes discusses when I talk predictability and sort of mention in a game plan, and I think you do highlight to me that perhaps the player's not thinking about that, as it's an added stress to them. So, I suppose, how do you communicate that predictability element to a player?

Marc
Look, I feel like I don't even use the word predictability and I know you put into your reports and I love when you put it in gives me an idea of what works and what doesn't. But one of the things I try not to talk to my players about is the what ifs. I think if we, as coaches focus on the what if, then we put a lot of doubt into the mind of the player because that means we're doubting. You know, what if they know you're serving there? Or what if they know you're going to play the ball there, you know? I think that adds a little bit of negativity to the game plan so I generally try to steer away from the word predictability and focus more on what can we control? You know, we can control where we put the serve, we can control what we're trying to do to the opponent, we can control where we hit our ball, how hard we hit it, how heavy we hit, how deep we hit it. That's in our control, let's focus on the things in our control, more often than thinking about the what if and, you know, if they read the play, if they know I'm serving there, so be it.

You know, I think the one thing about predictability that I don't know is in your data, and we're going to agree to disagree once again but, you know, if I'm serving to the ad side, how have I served it there? If I'm going, particularly out wide for example, one serve might be a flat serve, once serve it might be a kick serve, one serve it might be a slow serve, one serve might be a slider that comes back into the backhand, as opposed to going away. So that predictability might be - My opponent knows I'm going there, but I'm going to change up the way that I do that to make sure that it's unpredictable in the way that I've done it as opposed to where I've gone.

Shane
Yeah that's that's a really good point. So I think that's probably what distinguishes good sort of tagging to more comprehensive tagging where you are starting to actually look at the different spins, so the kick serve, the flat serve, the slider. You also factor that into the predictability calculation. So I actually think we do agree a fair bit here, and I think it's in around that definition again of what's a successful strategy and for me that successful strategy is one that's winning points, but it's one that is winning points consistently no matter when you use it, so I think when you factor that in, we're actually quite similar on our viewpoint on this. Communication is a big thing and Marc touched on why he doesn't even use the word predictability and and that's why he's a coach and I'm not, because I think I would confuse them if I went in with that.

Marc
Oh look, it definitely comes in language doesn't it. No, I think the ones that are delivering the messages need to have the right language around how we utilise the stats and the data. A lot of that predictability though Shane, and this is one thing that I look at with data, just to finish this off is, I think data tells you the average. Whereas, you might be playing a predictable pattern, but win the point on a moment that you need to win the point, even though the averages might not tell you that you're going to win the point. So, you know it could be ...averages give you a guessing tool, whereas sometimes a player might just be feeling that one moment and go - I can hit that serve, and I can execute this point right now. And this is where we've got to back the athlete and be individual in our planning and our thoughts, and ensure that they're comfortable doing what they're doing and back them 100%. If they play that predictable pattern and it doesn't work, so be it. They felt it at that moment. There's nothing a coach can say or do that can change a player's feeling.

I always tell my athletes that; hey you feel it, I see it, let's bring it together in the match and decide whether it was the right decision or not. And let's talk about it, rather than; hey you played the wrong pattern. I told you that that was a predictable way that you were playing. They knew you were going to serve out wide. Why did you do it? It happens and players will make decisions as they make them, and the one thing I don't want to put into my players mind is fear or doubt. And the one word that creates fear and doubt to me is that predictability and the what ifs. You know, I try and steer away from the terminology, but I use it in different ways. So, obviously, you give me a lot of that data and I try to use that as much as I possibly can. But in the simplest form so that I can understand it and that's where the art and science definitely collide and work really well together and amalgamate extremely well, so and that's where your work comes into my work and vice versa.

So it's it's a great topic, it's one of multiple answers could have been said and different opinions will be had on this topic and if you are listening to this podcast, please make sure you you do send us a message, send us an email or touch base with us on social media because we love hearing what other people think and your opinions and everybody sees it differently. So again, once again Shane you know your your research in these topics is exceptional and I do appreciate your time. Obviously you balance your time really well, Cricket Australia and Data Driven Sports Analytics, and obviously with our podcasting and I do appreciate what you do and the time you put into this sort of stuff.

Shane
Thanks again and anyone listening, if you'd be interested in this kind of analysis for your athletes, come and talk to Data Driven Sports Analytics and subscribe to The Tennis Menu. We do try and put this kind of data driven content out there as well.

Marc
Absolutely. And you know, thank you very much for listening, I we've had a lot of a lot of great messages on social media so thank you so much for your support. We continue to hopefully bring you some sanity in these insane times and, thanks for being part of Crunching the Numbers and hopefully we can continue to provide you with some stimulation and some excitement for tennis and tennis in the forefront of your mind. I've been Marc Sophoulis from The Tennis Menu. Shane Liyanage from Data Driven Sports Analytics, feel free to click on, have a listen to any more, the 1-8 podcasts that we've done already and stay in touch with us, we'd love to hear from you. So thanks again for joining us here on Crunching the Numbers.

 

Shane

See you all.

 

For a more detailed look at Crunching the Numbers, don't forget to checkout Shane and Marc within the Coaches Cupboard, where they do a deep dive into the latest trends in the world of tennis.