Keeper vs Keeper: Perkins’ Last Game Edition

Honestly, I felt funny doing this analysis of Perkins’ last game as a Timber. It was a bit tough to watch now that Perkins isn’t a Timber anymore. I was critical of his distribution but I definitely liked him and loved his dedication to the team. This post isn’t meant to support or argue against trading him; I just wanted to get some data to back up what I think about the importance of keeper distribution – no matter who’s playing. I’ll do the same for Ricketts, probably against RSL as that game got the most votes when I posted asking which one I should do. So here goes…

If you look at the basic stats, Hartmann won the showdown. His 6 saves and 1 goal against beats out Perkins’ 2 saves and 1 goal against. But, of course, that isn’t the whole story. A keeper stops balls from going in his net (hopefully), but a keeper also starts off a lot of possessions. How they do it matters. Dig a little deeper into the chalkboard data. Hartmann had 22 successful passes and 18 unsuccessful passes, while Perkins had 23 successful passes and 7 unsuccessful passes. Perkins wins that one with a better percentage of success, but so what? What did any of those passes lead to? That’s what I set out to find.

I watched the game and counted every time a keeper had the ball in their possession and sent it out to the field, whether a goal kick, a punt, a kick or a throw. I didn’t count punches or blocks, which I hope accounts for the fact that my totals don’t quite match the pass totals from MLS.

Here’s what I found….

Hartmann got the ball more often, with 35 distributions compared to 29 for Perkins.

But Perkins wins the battle when it comes to helping create more possession. Perkins’ distributions led to 8:21 of possession by the Timbers for an average of 17 seconds each. That helps account for almost 10% of the game. Clearly if keeper distribution choice has an impact on possession, it becomes an important part of the game. Hartmann’s only led to 6:15 of possession, an average of 11 seconds per.

Two of the distributions led to shots by Dallas. Both were from short distributions, one a throw and one a short pass.

Perkins’ distributions led to two shots and one corner kick, with the corner kick and one shot stemming from short distributions.

So the winner is……. Perkins.

There’s more info for data geeks like me after the jump.

Perkins also killed Hartmann in successful possession by his team. 79% of Perkins’ distributions led to Timbers possession compared to only 51% for Hartmann. This means Hartmann might as well have been flipping a coin where Perkins was largely successful. To add to Perkins’ domination, only three of the times Timbers ended with possession could really be called luck while 8 of Hartmann’s were. So Hartmann was actually worse than flipping a coin.

Hartmann (12 goal kicks all of them long, 5 long punts, 7 long kicks, 6 short kicks, and 5 throws). This means that Hartmann only sent out 11 balls he had enough control over to be resasonably sure Dallas would get first touch and retain possession, which is only 31% of distributions). Ten of these (all the short kicks and all but one long throw) stayed in Dallas possession resulting in a total of 4:03 of possession or 22 seconds each on average. Yes, somehow he flubbed a throw – I remember it being one of the long ones we somehow intercepted. These more controlled distributions were only 31% of total distributions but accounted for almost 2/3 of total possession starting with Hartmann.

For Hartmann’s long kicks he didn’t fare well at all. They resulted in a total of only 2:12 of possession with an average of just 5 seconds per distribution. Worse yet for him and better for the Timbers, the Timbers just edged out Dallas on the possession battle, getting first touch on 61% and winning possession on 52% of Hartmann’s long distributions. Of the times Dallas retained possession, just over half were essentially luck because either the Timbers knocked it out of bounds or let Dallas settle it after getting first touch.

Perkins (7 goal kicks – 4 long and 3 short, 0 punts, 8 long kicks, 6 short kicks, and 8 throws). This tells me Perkins had really been working on his distribution (and now Montreal gets to benefit, darn it). This means that almost half (14 out of 29) of his distributions were short enough that he had reasonable surety it would end in Timbers possession. I’m happy to say that every single one of those did end in Timbers possession. They yielded a total of 4:42, which is more than half of the total 8:21, and an average of 20 seconds per.

It’s good that we’ve seen fewer long balls from Perkins because this is the one place where I didn’t see a significant improvement from previous games. Of the long balls, Timbers maintained possession only 50% of the time and several of those were due to luck such as being knocked out of bounds by Dallas.

Bottom line, short balls are the only way to make possession likely. For both keepers, long balls were at best around 50/50. Possession was telling too. For Perkins, short balls yielded an average of 20 seconds of possession plus a shot and a corner kick, while long balls yielded only an average of 9 seconds of possession and one (blocked) shot. For Hartmann, short balls yielded an average of 22 seconds of possession plus two shots, while long balls yielded only an average of 5 seconds. When you consider that the ball spends 3 seconds in the air, that’s pretty dismal.

I couldn't figure out how to paste the MLS chalkboard distribution graphics into this post, but it's worth a look at the chalkboard site. Look at the difference on the graphics. Hartmann’s long balls were pretty much all over. Perkins had far more short passes overall and his long balls were more targeted.

Finally, let’s look at how Perkins has changed over time. I’ve done a similar analysis three times this season. Previous games I didn’t count every distribution, just times when he had time rather than quick passes that I included this time. Still, it’s worth looking at the trends.

	vs Houston 5/15	vs Vancouver 5/26	Vs Dallas 8/5
Total distributions	33	17	29
Long kicks	26	11	12
Short kicks	0	5	9
Throws	7	1	8
% Resulting in Timbers possession	48%	65%	79%

Perkins’ choice of distribution got tremendously better. Like night-and-day better. From pretty much a coin toss vs Houston and zero truly short kicks, to short distributions dominating and 79% possession. Kudos to Perkins or Toshack or both for addressing something that was worrisome earlier in the season.

If you read this far you must have gotten something from this. I know I did. Let me know what you think.