Toronto FC: Expectation Versus Reality all Adds up the Same

Aug 14, 2016; Houston, TX, USA; Toronto FC forward Jozy Altidore (17) celebrates with teammates after scoring a goal against the Houston Dynamo during the first half at BBVA Compass Stadium. Mandatory Credit: Erik Williams-USA TODAY Sports
Aug 14, 2016; Houston, TX, USA; Toronto FC forward Jozy Altidore (17) celebrates with teammates after scoring a goal against the Houston Dynamo during the first half at BBVA Compass Stadium. Mandatory Credit: Erik Williams-USA TODAY Sports /
facebooktwitterreddit

We use americansocceranalysis.com’s formula for expected goals, to see how lucky (or unlucky) Toronto FC have been so far this season.

As someone who grew up in England, it wasn’t until the age of 13 that I started watching the NFL (or as we called it, American football). This was soon followed by the NBA and so on, and eventually, my passion for North American sports actually convinced me to forge a career in the media industry.

One of the first things I noted — especially with the NFL and even more so MLB — was how detailed the statistical data was. Seriously, it seemed like they had everything covered, although if anything, it has become even more expansive over time.

More from Toronto FC

Regardless, I’ll gladly admit I embraced the analytical side of sports, as I didn’t know any other way when it came to the North American sports scene. However, then they started to apply this to soccer (or as we call it, football) — on both sides of the Atlantic mind — and it just didn’t seem right.

Maybe it was because I had grown up with football not being such a slave to in-depth statistics? In any event, I’ve never really put as much stock into the attempts to offer more detailed data for the original version of football.

This doesn’t mean I don’t at least attempt to keep an open mind. In that respect, I recently found myself intrigued by an excellent site called americansocceranalysis.com.

What particularly caught my attention, was their methodology behind expected goals, which is apparently a craze sweeping the world of soccer metrics. Admittedly I was very dubious about it all, but decided I should persevere for the aforementioned sake of objectively.

Just as it sounds, americansocceranalysis.com has come up with a formula which attempts to tell you have many goals a team should have scored in any given game. They use factors such as a shot’s distance, angle, height and location from the centre of the goal mouth.

Toronto FC
May 14, 2016; Toronto, Ontario, CAN; Toronto FC midfielder Sebastian Giovinco (10) watches as his shot goes past Vancouver Whitecaps defender Pa-Modou Kah (44) to beat Whitecaps goalie David Ousted (1) for a goal in the first half at BMO Field. Mandatory Credit: Dan Hamilton-USA TODAY Sports /

They also include aspects such as a goalkeeper’s position at the time of the shot and how far he has to dive. Finally, they consider how a shot was assisted, with a through ball seen as being more likely to result in a goal than a cross.

Of course, it goes into a lot more detail than I’ve provided above. I just wanted to give you all a rough outline, but feel free to click here for a more thorough explanation.

I thought it would be fun to put the formula to the test, using Toronto FC. Basically, as per the table below, I’ve compiled all their results to date, looking at goals scored versus expected goals, to see how many points they should theoretically be on:

 Date Home HG HxG Away AG AxG Points xPoints
 03/06/2016 New York 00.97 Toronto 21.45 3 3
 03/13/2016 New York City 22.59 Toronto 20.91 1 0
 03/20/2016 Kansas City 10.5 Toronto 01.21 0 3
 04/02/2016 Colorado 11.76 Toronto 00.64 0 0
 04/09/2016 New England 11.24 Toronto 10.93 1 0
 04/16/2016 DC United 02.05 Toronto 11.22 3 0
 04/23/2016 Montreal 00.87 Toronto 21.55 3 3
 05/01/2016 Portland 21.29 Toronto 11.14 0 1
 05/07/2016 Toronto 11.94 Dallas 00.79 3 3
 05/14/2016 Toronto 32.71 Vancouver 40.58 0 3
 05/18/2016 Toronto 11.62 New York City 11.01 1 1
 05/21/2016 Toronto 00.7 Columbus 00.88 1 1
 05/28/2016 New York 30.54 Toronto 02.26 0 3
 06/18/2016 Toronto 10.67 Los Angeles 00.5 3 1
 06/25/2016 Orlando City 32.1 Toronto 21.56 0 0
 07/02/2016 Toronto 11.53 Seattle 11.5 1 1
 07/09/2016 Toronto 11.53 Chicago 00.66 3 3
 07/13/2016 Columbus 10.6 Toronto 10.81 1 1
 07/16/2016 San Jose 20.74 Toronto 11.84 0 3
 07/23/2016 Toronto 43.06 DC United 11.16 3 3
 07/31/2016 Toronto 31.7 Columbus 01.21 3 1
 08/03/2016 Toronto 11.21 Salt Lake 01.1 3 1
 08/06/2016 Toronto 41.53 New England 10.94 3 3
 08/14/2016 Houston 11.26 Toronto 11.65 1 1
 08/20/2016 Philadelphia 11.2 Toronto 30.79 3 0
 08/24/2016 Orlando City 10.83 Toronto 21.51 3 3
 08/27/2016 Toronto 01.26 Montreal 10.31 0 3
 09/10/2016 Chicago 11.31 Toronto 21.02 3 1
 Totals 46 46

Now of course, we’re only talking about one team out of 20. However, I was amused to find the expected goals would have resulted in the same amount of points TFC currently find themselves on.

It certainly supports those who say any fortune balances out over the course of a season. In effect, Toronto FC are where they deserve to be.

More from Tip of the Tower

The issue is, as detailed and interesting as americansocceranalysis.com’s methodology is, it’s — just as with any similar formula — open to criticism. For example, New York’s 0.97 expected goals for the season opener may as well be 1.00, and what if a cross is more accurate than a through ball?

Further, if you’re going to take into account a goalkeeper’s position and distance they have to travel on a shot, is this fair or unfair against them? Some are more athletic or just plain taller than others, thus increasing their chances of reaching said shot.

Finally, the formula takes into account penalty shots, which have a higher degree of success. However, what if a team is awarded a penalty, because a player conned a referee by diving?

Overall, this is not an attempt to dump on americansocceranalysis.com (as much as it may sound otherwise). Their website is extremely well put together and has a lot of useful information.

Sometimes though, you get the feeling the soccer metrics crowd are trying too hard to come up with different ways of analyzing and breaking down statistics. And if that sounds like the words of someone stuck in the past when it comes to football, I’ll happily let someone out there explain to me why I’m wrong.

Next: 5 important TFC storylines for the remainder of the season

What’s your take on the whole soccer metrics scene? Have you embraced it the same way you do statistics in other sports, or do you think there’s an element of overkill taking place? Let us know in the comments section below.