Is the LA Galaxy offense better than last year's? A look at the numbers.

When the curtain fell on the 2014 season, it was a little bittersweet. Yes the Galaxy won the cup, sending Landon Donovan out the way he deserved, however, his departure raised a specter of doubt concerning the Galaxy going forward.

Would they still be good? Most agreed they would be. After all, they still had the league's MVP in Robbie Keane leading the charge. There was also no doubt that the Galaxy were going to replace Landon Donovan with a big name star who would no doubt put up strong numbers.

But would they be as good? That was the real question. From both an aesthetic and analytics standpoint, the 2014 LA Galaxy were the best team this league has ever seen, and it was entirely unclear how long it would be until the league saw another like it.

That was then. Since that time, the LA Galaxy have had some ups and downs, but ended up assembling a roster sporting the likes of Keane, dos Santos, Gerrard and Lletget, which has people wondering—could this team be better than 2014?

From a season by season breakdown, the answer is probably no as the 2014 team was together the entire year. From a current roster standpoint, the question is a lot more interesting. Here is a position by position comparison.

Note:

G= goals,  xG= expected goals, CC=chances created (key passes + assists), xA= expected assists, UnAst%= percentage of unassisted shots, Touch%= percentage of team's touches adjusted to time on the field, p/90 = per 90 minutes

Robbie Keane vs. Robbie Keane

Player	Shot p/90	UnAst%	G p/90	xG p/90	CC p/90	xA p/90	xG + xA p/90	Touch%
2014	4.0727	15.6	0.634273	0.584199	2.03635	0.282418	0.866951	7.3
2015	3.668306	15.5	0.9487	0.628672	1.960647	0.211244	0.839916	6.3

In the analytics world, a player like Robbie Keane is a bit of a marvel because you can pretty much set your watch to his consistency. There are a number of boring analytics reasons that explain why he's so consistent, but that's a different article entirely.

While the differences between years is minimal, they are intriguing. For instance, Robbie has been touching the ball less and shooting less, however the average positioning of his shots has improved, resulting in a slight rise in goal scoring and expected goals. Perhaps this can be attributed to him dropping back less due to Gio now occupying that space.

As his xG has risen, his expected assists have decreased, which is why his XG+xA remains roughly the same from last year. In other words, while Robbie's role has slightly changed within the offense, his impact on the offense per 90 minutes, has remained the same.

Gyasi Zardes vs Giovani dos Santos

Player	Shot p/90	UnAst%	G p/90	xG p/90	CC p/90	xA p/90	xG + xA p/90	Touch%
Gyasi Zardes	2.525597	10.8	0.546075	0.377816	1.023891	0.11843	0.49624573	6.5
Giovani dos Santos	3.47181	30.8	0.534125	0.373887	2.670623	0.437982	0.81186944	7.5

As good as Gyasi Zardes was last year for the Galaxy, Giovani dos Santos, in his admittedly short time with the team, has been better. Gio has not only been shooting more, but creating about 20% more shots for himself (a mark of sustainability). Their goals per 90 and expected goals per 90 are pretty dead even, however, Gio really pulls ahead of Gyasi in the chance creation department, creating chances at over double the rate and for an average shot quality about 4 times greater. Giovani dos Santos' 0.811 xG+xA p/90 is truly at a DP level falling just short of Keane's  0.84 mark.

It should also be noted that since a forward's expected goals is also a function of the offense's chance creation as a whole, the near identical xG rate between Gyasi  Zardes and Giovani dos Santos can be viewed as a testament to the Galaxy's offense and what it is able to consistently create for its forwards.

Landon Donovan vs Sebastian Lletget

Player	Shot p/90	UnAst%	G p/90	xG p/90	CC p/90	xA p/90	xG + xA p/90	Touch%
Landon Donovan	2.008114	31.8	0.30426	0.2857	3.772819	0.335294	0.62129817	10.6
Sebastian Lletget	2.331081	26.1	0.709459	0.247297	1.317568	0.12973	0.37601351	8.8

Sebastian Lletget has been an absolute revelation, and his contributions to this team can not be heralded enough. That being said, when matched up against the greatest American player of all time, he doesn't compare so favorable. The big difference between the two is that Landon, late in his career, was primarily a creator while Lletget in the early stages of his own, is primarily a shooter. This is not to say that Lletget doesn't excel at opening up the play, creating space for others and passing as a whole, however, when he gets into goal scoring positions, he is definitely more in the Robbie Keane mindset of sniffing for goal.

Lletget shoots more, has been scoring more (although that goals per 90 number is likely to drop significantly) but hasn't been the chance creator Donovan was. Of course, no American ever has. Lletget's goal scoring rate has outpaced his xG to such a degree that it is likely the Galaxy offense is reproducing the Gyasi Zardes effect from last year, which I explained in this article.

This should not take anything away from him, however, as he is still the most dynamic American player in his age group in the pool.

Stefan Ishizaki vs Gyasi Zardes

Player	Shot p/90	UnAst%	G p/90	xG p/90	CC p/90	xA p/90	xG + xA p/90	Touch%
Stefan Ishizaki	1.579739	45.7	0.225677	0.103811	2.527583	0.258175	0.3615346	9.6
Gyasi Zardes	1.654949	17.6	0.29205	0.243862	1.411574	0.147485	0.39183342	6.9

Gyasi's numbers are a tad skewed by his time at forward earlier in the year, however, his numbers at forward were so low that it hasn't actually skewed the goal scoring side of the numbers as one would expect.  While Zardes and Ishizaki are completely different types of players, their overall contributions to the attack have been roughly the same.

But again, since Gyasi has seen so many minutes at forward, all of these numbers should be taken with a grain of salt as they are not yet a reflection of Gyasi the winger.

Marcelo Sarvas vs Steven Gerrard

Player	Shots p/90	UnAst%	G p/90	xG p/90	CC p/90	xA p/90	xG + xA p/90	Touch%
Marcelo Sarvas	1.442071	12.8	0.110929	0.125349	1.331142	0.073952	0.19930156	11.7
Steven Gerrard	1.929556	42.9	0.137825	0.147473	1.929556	0.184686	0.33215927	10.1

As sad as it was to see Marcelo Sarvas go, Steven Gerrard has proven to be more than capable of filling in and then some. Gerrard has proven that he still has the legs to be a box to box player, and on the offensive side, he's done an incredible job. You would also have to expect that Gerrard's goal scoring will increase as he gets more comfortable with the team.

Juninho vs Juninho

Player	Shots p/90	UnAst%	G p/90	xG p/90	CC p/90	xA p/90	xG + xA p/90	Touch%
2014	1.64945	29.1	0	0.069577	0.92969	0.077074	0.14665112	13
2015	0.977411	28	0.156386	0.106733	1.133797	0.069592	0.17632494	12.4

Juninho's withdrawn role makes him hard to evaluate using the metrics I'm using, however, I think we can all agree that Juninho is a class player and one of the most underrated in the league. He's shooting less as Gerrard is clearly the one tasked with going forward. Other than that, his numbers are pretty consistent with last year, pointing to his reliability.

Attacking 6 totals

Of course, you can't really use the number of "upgrades" and "downgrades" as an indicator of which team was better. Soccer is a fluid sport, and, often times when systems change, different positions begin to take on different roles from previous seasons.

For instance, although the Galaxy have lost chance creation at the lm position, they've also gained a lot of that back in from the forward position. They've also gained more unassisted shooting as a whole (thanks to Dos Santos and Lletget), which has resulted in a rise in xG despite chance creation dropping.

As you can see below, the Galaxy front six of Keane, Gio, Lletget, Gerrard, Juninho and Zardes, have been much deadlier than last year's front 6 of Keane, Zardes, Donovan, Juninho, Sarvas and Ishizaki.

Year	Shot p/90	G p/90	xG p/90	CC p/90	xA p/90	xG + xA p/90
2014	13.27767	1.821213	1.546452	11.62148	1.145344	2.691982
2015	14.03311	2.778545	1.747924	10.42376	1.180719	2.928116

Of course, the sample size is still relatively small at the moment, but it certainly appears like the LA Galaxy have managed to top their impressive and analytically record breaking 2014 offense.