By Jared Young (@jaredeyoung)

An expected goals model frames quite well what everyone knows to be true: that not all goals are created equal. Goals are created from tap-ins and bicycle kicks and all the shots in between, and expected goals allow analysts and fans to quantify the likelihood a shot will go in the net, and therefore the value of a shot. The next logical step in the discussion of goals is to analyze their value. Goals aren't created equal and neither are goals. The goal that makes a two goal lead a three goal lead does not nearly have the value as that stoppage time goal that captures all three points.

To measure the value of a goal I looked at game states in MLS from 2011 to 2015 and built a series of functions that estimate the expected points for home and away teams given the score of the game and the minute being played. Each of the functions fit tightly with the actual data and had an R squared greater than .85. The expected points functions look like this for games with a difference of two or less goals.

With this collection of functions we can now calculate the expected value of a goal in terms of the all-important table. Here’s an example: It’s the 75th minute and the score is tied. Based on MLS averages the home team’s expected points is 1.34 and the away team’s is 1.02. The home team scores a goal moments later and now they are expected to earn 2.63 points from this match while the away team’s points drop to 0.22. The goal scorer created 1.29 points above what was expected. It might seem strange to not credit the scorer with the full two points but this difference counts for the fact that a different goal may have been scored later to seal the win and also accounts for the fact that the game is not yet over and there is more work to do to earn the three points.

What does expected points tell us that expected goals does not? Essentially expected points helps measure how clutch a scorer's goals are as well as more closely measure the impact a player is having on a team's performance. We can also look for players who tend to take more important shots when the game is on the line and assess their performance in those times.

Let's rank some players in terms of expected points. First, here is a list of the top 25 players in terms of points created from their goals. Interestingly, the MLS leading goal scorer **David Villa** has not created the most points for his team. That crown goes to the Timbers' **Diego Valeri** who has created 6.8 points with his 7 goals.

Next, here are the top 25 goal scorers in terms of points created minus expected points. To calculate expected points the ASA expected goals model for each shot is multiplied by the points that would have been created if the goal was scored. This metric will correlate closely with goals minus expected goals with an edge given to the importance of the shots taken.

Here we have a very different list as **Giles Barnes** has created the most points relative to how many points he should have scored given his shots, the difficulty of those shots and the game state. Here you also see the odd goal scorer whose lone goal was a clutch goal. The Union's **Brian Carroll** scored a 90+ minute equalizer against the Colorado Rapids and that's good enough to put him at 23rd on the list.

Last let's look at who consistently scores clutch goals by looking at points per goal. Obviously the more goals a player scores the more this score will drift towards the mean (unless this player only scores clutch goals), so I've limited the list to just players with 5 goals scored or more.

There's a pretty big difference in the value per goal. Valeri, the leader in this category, has earned the Timbers nearly a point per goal scored while **Roland Alberg**'s goals have been worth just a third of the value for the Union.

Okay, so these metrics probably don't get rid of expected goals altogether, but translating goals into points does provide additional insight into the impact a player is having. As the season progresses ASA will track points created and which goal scorer has been the most clutch this season.

Special thanks to @MattyAnselmo and @KevinMinkus whose background work made these metrics possible.