Sideout by Pass Quality Pt1
The application of expected sideout
Other articles in this series:
Sideout by Pass Quality Pt2
Sideout by Pass Quality Pt3
As volleyball coaches, we generally understand that pass quality influences sideout quality. I’ve done into that topic a few times this summer and introduced the idea of Expected Sideout as a metric that quantifies exactly how that passing quality contributes to sideout ability.
Expected Sideout is more of an evaluation method. You can use it to see you most effective servers, or in a comparison to actual sideout to help quickly determine training priorities.
The layer beyond just looking at an Expected Sideout number is to examine your sideout % by pass quality. I find myself looking at these numbers more than the xSO or xFBSO numbers now. xSO and xFBSO tell you a lot about your serve/pass game, but I find that the best information about offense and defense is found by parsing the Sideout By Pass Quality statistics.
I’ll show what this looks like in practice, a few different ways.
Let’s parse some things out. Let’s talk Earned Sideout. You can see the Earned Sideout for Canada was 64% and for China it was 53%. We can also see how it was earned by each pass quality. The sideout on Bad Pass was essentially a wash, but we earned a small-to-moderate advantage in each of the other pass qualities.
Many teams, including me, like to combine Perfect and Good to become “In-System” Sideout.
You can also combine Medium and Bad to be “Out-of-System” Sideout:
It all just depends on what you want to look at, the more you expand the data, the more accurately you can parse the situation, but you add nosie. When you compres the data, it’s more reliable, but less specific.
I do think it’s worth separating your Medium pass actions from your Bad pass actions. Even though the total data is quite a bit smaller (teams at many levels tend to be at or over 50% In-System, so the Medium and Bad Pass buckets are often less than half the size), I think they are fundamentally different from each other in a way that Perfect and Good Pass are not. Medium Pass situations are often still about scoring, while, at most levels, Bad Pass situations are more about managing the game and setting up the chance to defend in transition.
Here is a screenshot of a simple diagnostic that I use both in-match (hence the bare-bones formatting) and post-match to put all this into a package.
If I had to use one snippet that showed as much relevant data as I want in the sideout phase without adding too much noise, it would be that one. I always want to see total Sideout %, and then I like breaking it down by InSys, Medium, and Bad Pass situations. I also want to see the total Good Pass % as well as the Dead Balls- aces and overpasses. This helps me see the effect of quality of pass, losing opportunities to sideout at all (aces and overpasses) and how they all fit together.
I’ll expand on this more in upcoming posts, but if the Triangle is my first level of post-match diagnostic, this is my second level.