Stat suggestions for a high school team?

[PlayMakers]

Hi folks, I coach a high school team and I wanted to put together some advanced stats for the boys this year.

Players love feedback of any kind and I think some non-traditional stats could be good motivators. I also want my players to focus on the little things that make you successful, so I was thinking about tracking takeaways, giveaways, quality shots for/against... and I thought I'd reach out to the experts for some suggestions.

The players will be required to watch their shifts from the game and tag the video with their stats.

Hopefully they find it fun to re-watch and track their shifts and see how they rank on the team for the different categories. It gives me useful info as a coach as well. If a player is getting caved in Corsi-wise maybe I need to change that line or their matchup. If I have the sense that a young player is playing timid then I can use low takeaways as a concrete example of what he needs to get more involved physically.

Here's what I have so far...

All Skaters
G
A
+/-
TkA
GvA
ShQ (shot quality)
ShM (shots missed)
ShB (shots blocked)

Forwards
FO's won/lost by the C (not the wings)

Defense
Zone Exit Passes
Zone Exit Carries

Goalies
Shots
Quality Shots
Where they're getting scored on (5hole, low blocker, etc)

Click to read more...

 

[Michael Farkas]

That's a lot of work. Asking players to watch every one of their own shifts is a lot to ask at any level, but then tagging them for several items (that I'm assuming actually have some defined definitions) for the purpose - at least, in part - for competing against each other has the potential to be very problematic.

You don't have to worry about "advanced" stats unless there's a goal in mind. What problem are you trying to solve? What will you glean from the information that makes it actionable? I saw the takeaway:timid example given, but even that's too strong for me. I don't necessarily think that there's a connection there - and if there is, it should be obvious from the video without it being quantifiable.

Before I get too far into this, I think it would be helpful to learn what kind of resources you have available to you...sounds like not much, given that you're assigning a lot of manhours of work to high school students. And I'm not saying that pejoratively, I've been coaching college hockey for a decade now, I know what kind resources you get haha

 

[Doctor No]

Adding to what Mike said above, and especially at this level, you're going to end up with a significant difference opinion for players as to what constitutes a shot "on goal", let alone giveaways or takeaways (especially when they're going to be compared against their peers).

It's not even (necessarily) overt gaming, but I've had my goalies track shots on goal and they're consistently 20% higher than what an official scorer will count.

 

[PlayMakers]

That's a lot of work. Asking players to watch every one of their own shifts is a lot to ask at any level, but then tagging them for several items (that I'm assuming actually have some defined definitions) for the purpose - at least, in part - for competing against each other has the potential to be very problematic.

You don't have to worry about "advanced" stats unless there's a goal in mind. What problem are you trying to solve? What will you glean from the information that makes it actionable? I saw the takeaway:timid example given, but even that's too strong for me. I don't necessarily think that there's a connection there - and if there is, it should be obvious from the video without it being quantifiable.

Before I get too far into this, I think it would be helpful to learn what kind of resources you have available to you...sounds like not much, given that you're assigning a lot of manhours of work to high school students. And I'm not saying that pejoratively, I've been coaching college hockey for a decade now, I know what kind resources you get haha

Thanks for the reply Mike.

My resources are limited. The school buys a subscription to Hudl. I have a team manager who records the games, pausing at breaks so it's not a 2 hour video. Hudl Assist tags the goals, assists, shot attempts, shots on goal, saves and blocks as part of their package so some of this is done for us.

I wouldn't say I'm trying to solve a problem so much as I'm trying to enhance how I teach. I want to encourage players to focus on the process, the small things that lead to success (rather than focusing on the outcome). I think tracking stats that highlight those small wins will make it something they get excited about.

So, when I think about some of the little things that help you win, habits I want to encourage and promote, I'm talking about things like...

Winning races and loose puck battles.
Creating turnovers and limiting giveaways.
Taking quality shots, not just spamming pucks on net.
Tracking missed and blocked shots to show identify guys who need to hit the net or do more to create shooting lanes.
Tracking shot assists (guys who set up shots) to help understand who the facilitators are and celebrate them.

I want centers to track their performance at the dot. I want to know who's winning/losing and I want them reviewing those moments to learn from them.

I want the defensemen to see what passes and carries led to zone exits and which decisions led to turnovers.

I want the goalies to see where we're giving up shots consistently so they can be ready for shots from those locations.
I want to see if there are patterns forming in terms of where they're getting beat.

I'm open to suggestions of things that are worth tracking.

I agree that I don't want to give players a mountain of homework. I want to boil it down to maybe 5 or 6 things they're watching for during their shifts. The most anyone plays in a game is about 17-20 minutes and as I said the shifts are already parsed by the manager so they don't have to re-watch the whole game, just their 15 minutes of ice time. My hope is they each take 30 minutes on game days to revisit the last game and tag their video. I think it will get them thinking about those good habits and it will give me data I can package and present back to them in teaching moments.

 

[PlayMakers]

Adding to what Mike said above, and especially at this level, you're going to end up with a significant difference opinion for players as to what constitutes a shot "on goal", let alone giveaways or takeaways (especially when they're going to be compared against their peers).

It's not even (necessarily) overt gaming, but I've had my goalies track shots on goal and they're consistently 20% higher than what an official scorer will count.

I suppose it doesn't matter for the goalies. I'm not doing it so the goalies can compare their shots on goal vs the league. I'm doing it so they can see trends in our defense and their own game, specifically where the weak points are.

I agree, the skaters would need some definition on what constitutes a quality shot or a takeaway. I'm surprised to hear there is a difference of opinion on a shot on goal.

We do video sessions twice a week so I suppose the threat of calling up Tommy's takeaway is always there to keep them honest, but it's a fair point regarding how accurate their tracking will be.

 

[Michael Farkas]

Thanks for the reply Mike.

My resources are limited. The school buys a subscription to Hudl. I have a team manager who records the games, pausing at breaks so it's not a 2 hour video. Hudl Assist tags the goals, assists, shot attempts, shots on goal, saves and blocks as part of their package so some of this is done for us.

[/QUOTE]

I also use Hudl. That's a good start, sounds like you have Hudl Assist. That's a good value.

I wouldn't say I'm trying to solve a problem so much as I'm trying to enhance how I teach. I want to encourage players to focus on the process, the small things that lead to success (rather than focusing on the outcome). I think tracking stats that highlight those small wins will make it something they get excited about.

[/QUOTE]

Ok, this is good. In my experience, you're not gonna what you want out of this I don't think. This is too much, it's too broad, there's too much variance as to how these things can be interpreted, etc. What I'd recommend, is start small and address what you perceive a problem is.

For instance, I coached a team of small, skilled players years ago. We could score, we were a dynamite power play team. We were a dynamite rush offense. But against really good teams, we had a harder time. There aren't any power plays in the playoffs. Our problem was that we didn't really score near the net. I'm a high cycle guy these days, blah blah blah. So what I did was put a reward out there for the season for most POP. POP = Points off pads.

So, any goals off a rebound, any assist that led to a goal off of a rebound, and by my own judgment, a really good far-post drive that allowed for the shooting/passing lane would get an assist as well. Every 3 or 4 weeks, I'd go into the room, not say anything, and just start writing on the whiteboard...

Player | POP
#2 - 4
#4 - 1
#7...etc.

Pretty soon, you get everyone surrounding you, they're taking bets on what players might have, you throw a big number up there and brings the house down, you put a 0 up there, and guys give that player a hard time, etc.

I do the same with plus/minus. Actual GF and GA and then the plus/minus rating. You want plus/minus at this level. That's one thing every veteran coach told me - make sure you have that, and I do. It's valuable at this level.

I wouldn't overload because you can't solve all of these problems at once, you're not gonna get the data you think (players aren't going to assign themselves giveaways...my guys will watch tape and claim incorrectly that they deserve a real assist on a play...they can't get assists right, they aren't gonna figure out high-danger scoring chances or whatever haha). Even Doc said, you don't get shots on goal right. That's my experience as well. It can vary by ~20% on just regular old shots on goal...

I'd focus on what you perceive to be an issue and hammer on that. No sane person goes to the gym and works out every muscle group, right? It's Monday, it's chest day, I'm doing chest stuff. Make your goals clear, concise, and achievable. Some stuff will cost you other stuff...like, if you go "hey, our controlled zone entry game stinks...we gotta be better..." Well, all right, you need a way to do that. You need your breakout, you need your panic-out, you need your NZ routes, right? You need all of these components. But you have to be ready for giveaways to go up until you can figure out how to improve controlled entries. Because if guys are sitting there going "ok, I have to be more creative with the puck...but I can't have my giveaways go up..." it's paralysis by analysis almost there. At least the opportunity for that is there. You don't want that. I mean, maybe you do. I don't know what your style is, but I'm a heavy offense guy, I'm a pace pusher. I want it to be comfortable for players to make mistakes, etc.

So, when I think about some of the little things that help you win, habits I want to encourage and promote, I'm talking about things like...

Winning races and loose puck battles.

[/QUOTE]
Don't we all? haha. Who isn't winning races? Are we talking about backchecking forwards? Are we talking about forecheckers? Are we talking about d-men on puck retrievals? Ideally, every player wins every race, but that's hard to do. This is a complex one. There's a team tactics angle here that needs to be considered in my opinion.

Creating turnovers and limiting giveaways.
Taking quality shots, not just spamming pucks on net.
Tracking missed and blocked shots to show identify guys who need to hit the net or do more to create shooting lanes.
Tracking shot assists (guys who set up shots) to help understand who the facilitators are and celebrate them.

[/QUOTE]
I kinda touched on the turnovers/giveaways part already.
Shot quality isn't too tough for your backup goalie or a scratch to keep track of. Instead of a dot, write a number. One sheet, with half the rink or just one zone, per period. Then you simply enter it into Excel - like slot shots or house shots or whatever. You'll quickly be able to generate ratios per players.

Missed/blocked. Hudl Assist tracks these or no? Regardless, this is one of those ones were you might have to "feel" if it's a problem before you put too many resources on it.

Shot assists. That's something you can probably get from Hudl Assist, right? I like the thought here, guys will be interested in that. But again, these are stacked/blocked - if you suspect you're not getting shot quality, do you really want to get shot assists fired up already? Maybe. I think it's a valuable stat regardless, but I think you gotta bring it to them.

I want centers to track their performance at the dot. I want to know who's winning/losing and I want them reviewing those moments to learn from them.

[/QUOTE]
Hudl Assist tracks this.

I want the defensemen to see what passes and carries led to zone exits and which decisions led to turnovers.

[/QUOTE]
That's purely a video situation for me. That's not numbers, that's qualitative. I'd put a couple of shifts together and make them part of a video session.

I want the goalies to see where we're giving up shots consistently so they can be ready for shots from those locations.

[/QUOTE]
One thing I have done for my goalies at times is I put someone behind the glass (I didn't tell them, so that they didn't overthink the game) and recorded over their shoulder so that they could re-live the shots. It's hard for goalies to view the game from the "TV angle" in my opinion. So, even if you go, "hey, you've given up 42% of your goals high glove" that isn't necessarily actionable. What if six of them were unstoppable deflections that went bar down? If you want to focus on the process, I might go a little different way with goalies...but there are a lot more goalies on this board (than I or anyone would like haha) and they'll probably be able to provide some thoughts...I don't think like a goalie, I'm just a regular, sane hockeyman.

I want to see if there are patterns forming in terms of where they're getting beat.

I'm open to suggestions of things that are worth tracking.

I agree that I don't want to give players a mountain of homework. I want to boil it down to maybe 5 or 6 things they're watching for during their shifts. The most anyone plays in a game is about 17-20 minutes and as I said the shifts are already parsed by the manager so they don't have to re-watch the whole game, just their 15 minutes of ice time. My hope is they each take 30 minutes on game days to revisit the last game and tag their video. I think it will get them thinking about those good habits and it will give me data I can package and present back to them in teaching moments.

Like I had suggested up top, I think 5 or 6 is a metric ton for what you're doing. And believe me, when I was getting going, I absolutely thought the same stuff. And I'm not saying you're wrong now and I'm right, none of that, there isn't right/wrong here generally...but there's information overload at a certain point and I think you're there based on my experiences. I don't think there's a chance in hell that you're getting half of your team to tag their own videos. And I don't think you'll get consistently reliable data even if you do.

I used to do longer video sessions where I'd sit and talk to guys. And I tried to keep it light, I keep jokes in there, I keep them engaged...but even so, these days, it's no more than 20 minutes and it's not often. They lead it. I set it up and they knock it down, so that it sticks better.

"Ok, so here we have a breakout that goes **** up...we have the slash support from the weak side, that's good. But what goes wrong here, Player 10?" - It's not even necessarily true that Player 10 is at fault or on the ice at all...but he ought to know what's going wrong...so he describes it to the group and we talk about it. And they'll point things out, and I'll point things out, and we work through it and we move on. But the notion that I go, "hey, all you guys...review this game...find yourself, tag these instances of what you're doing, and..." no chance will that work out. You'll also end up with players realizing that maybe they're only playing X time, and this other guy is playing X+6 amount of time...

I don't know, I just don't think you can do that...if you can, god bless ya...but I can't imagine...I think you focus, 1 or 2 things, achievable, clear, and you bring it down to them, and you make it a competitive thing that you track...

 

[Hockey Stathead]

Really cool idea to go beyond goals/assists/points at that level.

A few thoughts:

-instead of regular +/-, maybe do specifically for 5v5 otherwise it can get distorted as guys who play powerplay (can't get plus but can get minus) get penalized while guys who play shorthanded (can't get minus but can get plus) get benefit. 6v5 and 5v6 at end of games can also create distortions with empty netters. Usually more informative to express as a percentage too rather than a whole number.

-would focus more on shot attempt +/- (Corsi as you say) anyways since players have very little control over their goalies' save percentage. And defensemen have little control over teammates shooting percentages when they're out there. And the season is shorter than 82 games so things don't necessarily have time to "even out." Luck will play a role in the much smaller sample of "goals" compared to "shot attempts."

-giveaways can be deceiving as the top of the list is mostly defensemen who handle the puck a lot and make plays. Erik Karlsson was at/towards the top for some of his best seasons. Doesn't mean giveaways are good but need to be considered with what else a player does. Making the "safe" play like chipping glass and out might appear better but often the player trying to make stickhandling/passing plays with the occasional gaffe will help the team more on the whole.

-shots missed/blocked should be taken as percentage of all shots taken so that players who shoot a lot aren't penalized.

-maybe track percentage of controlled OZ entries by forwards compared to when they're forced to dump or the play gets broken up.

-flipside, can track successful DZ entry denials and forced dumps by defensemen compared to giving up blueline with control.

-for your own benefit, maybe track zone start percentages and total ice time for each player broken down by ES, PP, and SH. Can use those to get rate stats like goals per 60 minutes so everyone is on the same scale.

 

[KaraLupin]

I think at lower levels, where you aren't getting paid to do just hockey, and your own goals are split between education and hockey, it's going to be hard for every player on the team to do the homework (even if light) you give them after EVERY Game. And even if it's not after every game, asking a player to go through their shifts of MULTIPLE games is worse.

The focus should be on you bringing them specific situations, that you found for them, and helping them realize the solution quickly without them having to dedicate additional creative brainpower.

It does nothing beneficial for the player to discover their own faults through hard labor and problem solving, stat watching, etc. If there are glaring team faults or individual faults, that's your job to figure for them, through practice drills, video session. You can even do a mailing list where you track every player and pick one thing you want them to focus on specifically, then make a tailored video session for each player to watch in their free time. Emphasize you did the work for them to nitpick, so they will want to spend time to watch.

I just think asking players at that level to go and nitpick their own game, and discover their own "Faults" through the stats might be confidence crushing, stimulation overload in sense of still wanting to focus on their strengths alot, and sometimes misleading of their skillset depending on the stat.

It's good coaching in theory, but maybe scale it down to the even more basic if you plan to try to introduce this to HS kids.

 

[Primary Assist]

To answer the question, some subjective stats help as well. I used to track controlled zone exits/entries in addition to uncontrolled zone exits/entries. That way you can really track what players really lead the rush vs the ones that are glass and out every time.

Puck retrievals, especially in the o zone, are a helpful stat too. That's an interesting one because the best forechecker may not actually lead the team in the stat. Similar to how a net front guy may not get an assist on a goal, but be integral to it, a puck hound forechecker may be able to angle + steer a defensive player into giving up the puck, but may not retrieve it himself