Introducting Player Skyscrapers! - A new visual for comparing players

[El Dude]

So over the last week or so I've been spending WAY too much time playing with a new visual tool I created to analyze and compare players, and I wanted to share it with my friends here at Tennis Frontier. I was considering doing a blog article about this, and may at some point, but I'm also thinking this might work best as a thread.

Anyhow, it struck me while putting these together that as much as I like tennis statistics, I even more enjoy visuals that express or explain some aspect of tennis, particularly in relation to history and career accomplishments. This is the latest "fruit" of my ongoing experimentation.

So what is this, you ask? Well, I call them "skyscrapers" because that's kind of what they look like. But basically it is a way of depicting a player's major accomplishments in an informative and aesthetically pleasing manner. Here's how it works...

Each player skyscraper has a dark and light color. The dark color are titles, the light second week Slam appearances that aren't wins.

Each block represents (roughly) 250 ATP points, so for the titles an ATP 250 is one block, ATP 500 two blocks, Olympics gold is three blocks, Masters four blocks, World Tour Finals six blocks, and Slams eight blocks. I've also given five blocks for the defunct WCT finals and Grand Slam Cup. For Slam appearances (light color), I've given two blocks for a quarterfinal, three blocks for a semifinal, and four blocks for a final (I'm considering playing around with the ratio here, but this is what I'm going for now).

Just to clarify: these skyscrapers do not depict a player's entire career - they depict the span in which a player was winning titles and/or reaching the second week of Slams, what we could call their "contender phase" of playing.

Anyhow, I've done dozens of players going back to Bill Tilden, although almost all in the Open Era. The further back, the less accurate. For instance, the data is fuzzy on how the tournaments differentiate as recently as the 70s-80s, but certainly before the Open Era. For that I used TennisBase's tournament types, although had to stop as TB now requires an 85 euro per year signup fee (blah). So consider the older players to be less accurate, and all players to be more approximations than any final determination of their greatness.

But the bottom line is that I find this to be a very interesting and useful way to compare players. I found out some rather interesting things through doing this that I'll talk about as I share more players.

To start, here are the Big Four. I'll post my comments in a reply. Enjoy!

p.s. Part of the point of this thread is that I am taking requests! If you want to see a particular player, or players compared, please let me know and I'll be happy to oblige! Otherwise I'll probably just post sets of chronologically similar players going backwards in time over the next few days.

 

[El Dude]

So to start with the Big Four, on one hand there are no real surprises, but some interesting things are clarified.

*Different rates of ascendancy: One of the studies I'd like to do is on how quickly players go from winning their first title, to becoming elite players. Here we see very different "rates of ascendancy" - with Roger being steady but gradual, building up over the course of four years; Rafa was very quick - his first title in 2004 and then one of his very best years (fourth, by this chart) in the following year. Novak has that interesting spike and then fall, then a much higher spike as he reached a new level in 2011. Andy was on a much lower level for most of his career than the other three, and then looking to be undergoing rather typical age-relative decline in 2014-15 before having his greatest season ever in 2016, one that doesn't look out of place among the primes of the other three.

*Different "signatures": One of the interesting things about all of the skyscrapers is that every player is different. It is like a thumbprint or signature, each telling a unique story.

*Different styles of greatness. On one hand Roger's skyscraper is clearly the best, but on the other hand...is it? It is longer because he's been around for more years. He's also got that lovely four-year spike. But Rafa has that nice solid core, largely because of his consistency no clay, and Novak's peak is longer, with six seasons in a row of a height higher than all of Roger's except for the four.

*Best years. According to this it is Novak in 2015, then Roger in 2006, then Novak in 2011, then Roger in 2004, 2007, and 2005. And then we get to Rafa's best years. What I found a bit surprising was that Andy's 2016 was better than all but two of Novak's.

OK, enough for now.

 

[El Dude]

To add a bit more grist for the mill - here the best "second tier" players of Rafa's and Novak's generation (born 1984-88):

What is most striking is the drop-off from the Big Four to these guys. Comparing the two only further accentuates just how dominant the Big Four have been.

(I'll do another for Roger's generation a bit later)

 

[El Dude]

This might be a bit better in that it includes Novak, Rafa, and Andy so you can see them all side-by-side:

 

[El Dude]

If you look closely at the second image, you can really see how "weak sauce" Tsonga and Berdych have been, which such a large portion of their skyscrapers coming from second week Slam losses - meaning, not being able to pull a Slam win off despite numerous opportunities. Andy's early career was rather like that as well, but not as bad.

 

[mrzz]

You said you´ll do the Federer generation next, so no need to ask this, but just to register I would like to see Roddick.

 

[El Dude]

You said you´ll do the Federer generation next, so no need to ask this, but just to register I would like to see Roddick.

Your wish is my command. Here are the top players of Generation Federer (with apologies to Verdasco, Blake, Fish and a few others).

 

[El Dude]

I guess I'll just keep posting charts by generation, then maybe some comparisons if there is any interest. If not, well, I'm having fun

By my estimation, this is the third weakest five-year generation - after 89-93 and 39-43, in my Generation Theory. No real all-time greats, but several Slam winners and a solid second tier cast. This is an interesting generation in that there's no abrupt drop-off, just a gradual fading from second to third tier type players. There are a bunch of players that aren't far behind Grosjean and Norman: Arnaud Clement, Felix Mantilla, Nicolas Kiefer, Sjeng Schalken, Guillermo Canas, Jiri Novak, Radek Stepanek, Rainer Schuttler...


Some unusual careers here - like Tommy Haas, for instance. And if only Marcelo Rios had been more consistent - his "middle finger" would have been a strong fist.

Comparing to Federer's generation, I think Kuerten is comparable to Hewitt and Roddick but not really any greater. In fact, Kuerten's chart looks a bit like Wawrinka's.

 

[El Dude]

And one more, because I'm on a roll.

Later I'll post comparisons with players of similar accomplishments, but if you eyeball this it looks like Sampras is slightly below the careers of both Rafa and Novak.

This chart really illustrates Andre's "two careers." Jim Courier has a similar pattern to Kuerten: strong early peak, weak later years.

 

[El Dude]

Well, these skyscrapers weren't the catalyst of historical conversation I'd hoped they would be, but oh well. :cry I'll keep posting them by generation, so those who might be interested can see them.

To me what stands out is how weak Wilander is, relatively speaking - that is, for a seven-Slam winner. Clearly his career was not quite on par with Edberg and Wilander, and maybe not much better than Murray.

I know this is just one lens of looking at career greatness, but the visual does say a lot. This group is pretty easy to rank, with a big drop-off after the "Big Three" and then Muster and Stich above the rest of the field. Mecir is kind of the Rios of his generation - another what-could-have-been story.

 

[mrzz]

The part I liked most is how Rio´s Skyscraper looks like someone giving the middle finger. It fits his style.

 

[Kieran]

It's incredible work, Dude!

Just out of curiosity, do these charts take long to construct, and are they simple to update once they're made? It's a very interesting and clear system, I dunno how you do it! :clap

 

[El Dude]

The part I liked most is how Rio´s Skyscraper looks like someone giving the middle finger. It fits his style.

Haha, I noticed and chuckled at that as well.

It's incredible work, Dude!

Just out of curiosity, do these charts take long to construct, and are they simple to update once they're made? It's a very interesting and clear system, I dunno how you do it! :clap

Thanks, Kieran. It is pretty simple and easy to update once they're made. I have one big spreadsheet with all of them on it, by generation. I also have another that I make comparison charts, copying and pasting them over. After I finish with the generations, I might do other groupings by "tier" of accomplishment, so we can see them side-by-side.

The thing I like about this system is that it is visual - unlike some of the other systems, which end up with number ratings. That said, I'm tempted to tweak it...it seems like it under-values Slams. What I might do is add in Slam wins for the light color as well, so they're double-counted in both dark and light colors - dark being titles, light being Slam results.

 

[El Dude]

Here's another.

What this brings to light, for me, is that despite "only" winning 8 and 7 Slams, respectively, Lendl and McEnroe are really closer to Sampras, Nadal, Djokovic, and Federer than they are the Becker, Wilander, Edberg group, with Agassi being the player in-between the two groups. This also has such a huge drop-off from those two to the rest of the pack, with Noah being the "best of the rest."

 

[El Dude]

Probably just one more for today.

Aside from Borg, a pretty weak generation. It just doesn't have the depth of second tier talent some of the other generations have. It also shows you just how dominant Connors, Borg, and McEnroe were in the late 70s and early 80s.

I think this chart also shows us that despite the "what could have been" mystique of Borg, he was clearly on a down-turn. Maybe he would have bounced back somewhat, but I'm guessing he would have finished with 14-15 Slams at most.

 

[mrzz]

I´m impressed with Lendl´s figure...

Maybe one thing that could be interesting for perspective, is to group players by Slam count. The first group could be Federer, Sampras, Nadal and Djokovic. And so on...

Other thing that might confuse a bit is that (I believe) that every time you group players, you re-scale the graphs, right? So this way you lose a bit of perspective...

 

[El Dude]

I´m impressed with Lendl´s figure...

Maybe one thing that could be interesting for perspective, is to group players by Slam count. The first group could be Federer, Sampras, Nadal and Djokovic. And so on...

Other thing that might confuse a bit is that (I believe) that every time you group players, you re-scale the graphs, right? So this way you lose a bit of perspective...

That's a great idea, mrzz - I'll do that. It will be a great way to illustrate how different the overall careers are of different tiers of Slam titles.

As for the re-scaling, that is unintentional and I'm not sure how to get around that. I'm using my Snipping Tool to copy bits of the spreadsheet, then uploading the image to Imgur, then linking it here. Any suggestions of a better way, and one that would keep the size the same?

 

[mrzz]

...hmmmm, maybe you should always have some reference, for example a line (or even one column) with the height of Djokovic´s 2015 (which I guess is the highest bar out there, right?). Maybe that´s too high, and will make most player´s skyscrapers look too tiny... so in this case a more average reference would be better.

 

[El Dude]

Rod Laver actually has three years--1961, 1967, and 1969--that are all higher than Novak's 2015, but they're all before ATP era rankings and largely based upon TennisBase's tournament type, so it is hard to judge. But after Laver's 1969, Novak's 2015 is the highest.

 

[El Dude]

Up next, is Jimmy and crew.

Not really any surprises here. This is where I had to start using Tennis Base, as Wikipedia and the ATP site don't really differentiate beyond Grand Prix (Masters) and other, lower level tournaments.