Total Player Charts: Conceptualizing Time On-Ice Percentages

Ben Wendorf
December 10 2012 08:48AM

 

 

I've always felt pretty strongly about Tom Awad's now-two-year-old assertion that, absent something better, even-strength ice time is a pretty good proxy for overall player value.

His numbers bore that out. However, I suggested that we could better refine and increase predictability of even-strength ice time by expressing it as a percentage of the team's even-strength time in games the player played (a wordy way of describing 5v5%). Though it will probably be a post in-and-of-itself, 5v4% and 4v5% are both testing to be similarly positive refinements on those TOI metrics.

Rather than handle that today, I wanted to unveil a little idea that had been bouncing around my head for a while. I'll be damned if I can recall the post from Beyond the Box Score, the excellent baseball statistics web site, but they had a season preview a while back where they used radar plots of a number of metrics like OBP, Fangraphs' Baserunner rating, UZR/150 (something to that effect), and for each player on a given team their talent in those metrics filled a certain portion of the radar plot. The more you filled the radar plot, the better you were. Author Note: Having heard back from BtB's Justin Bopp, the creator of the Diamondview Composite Player Evaluation that inspired the Total Player Charts, I can put my agitation and accreditation worries to rest).

There's certainly room for doing this when the statistical hockey gods agree on a metric already (spoiler alert: they never will) - till that time I'm content to build a radar plot of three metrics, 5v5%, 5v4%, and 4v5%. In part, this is to represent player value, as well as where that value gets allocated (just even-strength, or powerplay and penalty kill as well). I'll call them Total Player Charts, or TPCs (which, it turns out, is also an acronym for all sorts of important ish)...you can do a lot of neat stuff with 'em.  

Let's have a look.

TPC Charts - What They Look LIke

I’ll never forgive him for taking that shot at Niedermayer in the Cup Finals, but I have to respect one of the most important forwards to their team of the last few years, Daniel Alfredsson.

The light blue triangle represents the average of the top 10 league performances for the category. I've split defensemen and forwards for these charts. The dark blue represents the player, and the dotted triangle is the league average (in this case, for forwards). In the case above, you are looking at the TPC of Daniel Alfredsson over the last five years, who happens to be one of the most valuable forwards (by TPC) in the NHL for that time.

Top defensemen look a bit different; here's one of the most valuable defensemen of the last five years by TPC, Jay Bouwmeester:

TOI fiend Jay Bouwmeester. He even puts a guy like Zdeno Chara to shame.

Top defensemen frequently cover the bulk of the triangle, while top forwards are all over the map of deployment. Take Alex Ovechkin:

Alex Ovechkin’s TPC.

Or Sidney Crosby:

Sidney Crosby’s TPC.

And now see Pavel Datsyuk:

Now see a guy who’s often in the discussion for most-complete player, Pavel Datsyuk. In some ways, though, the difference between Datsyuk and Alfredsson’s TPCs is a testament to the difference in team depth.

Or David Koci:

That little sliver of insignificance would be David Koci’s playing time.

Haha, just kidding, Koci is terrible. You know you're bad when they've given up trying to figure out if you're a defenseman or a forward.

Uses and Discussion

So what else can I do with this? Well, for starters I can look at particular player populations, and with my newfound GIF powers I can do some pretty neat things - like show the difference between 1st, 2nd, 3rd, and 4th line forwards:

TPC 1st Thru 4th Line Forwards

Or between 1st, 2nd, and 3rd pair defensemen:

TPC 1st Thru 3rd Pair Defensemen

Maybe you'd like to see the TPCs of 50, 40, 30, and 20-goal scoring forwards:

Hey, think your favorite player is going to score 50 goals? And he's 2nd unit on the powerplay? Good luck with that.

Maybe I want to see how a player's TPC evolves as they age - first with defensemen:

As you can see, penalty kill work is definitely not wasted on the youngsters. However, this is a tricky graph, because of survivorship bias. What if, instead of just going off the player population, we used year-to-year change trends and created a hypothetical defenseman who played from ages 18 through 40 (keeping in mind this would be a pretty exceptional defenseman)?

It's interesting to see this one, as it's typically the offensive defenseman that hangs around the league. How about forward age progression?

And now one built not directly on the forward population, but on year-to-year age progression trends:

Once again, you see the offensive-prone forwards hanging around, although unlike with the defensemen penalty kill time seems to sag right along with powerplay time for forwards.

This can also be useful for looking at how player types are matched. For instance, look at Anaheim's top forward line of the last few years:

The top line of Anaheim, with real data this time. Getzlaf did less than I expected in 4v5 time, but you can see why Ryan’s point totals have been a bit more sporadic in recent years (what’s your excuse, Getzlaf?).

You start to get the sense of how they are different players (unless you want me to GIF it, I could GIF everything if...no? It's killing your browser? Okay, nevermind.) In some cases, you start to get a sense of which are the stronger players, too. Take this duo of 2011-12:

Here are the TPCs of one of the more-common pairings last year, Carlo Colaiacovo (now on the Red Wings) and Alex Pietrangelo of the St. Louis Blues. 

I don't think anybody would argue that Corey Perry isn't better than Bobby Ryan, though that's closer than the gap in the argument between Colaiacovo and Pietrangelo.

Long story short, TPCs are a pretty snappy way to look at player value at a glance, or even the differences in whole lines or depth charts. If you have any TPCs you want to see, let me know in the comments and I can do a follow-up including some pretty incredible individual season TPCs.

As usual, my research is indebted to Gabe Desjardins and his invaluable website, behindthenet.ca.

 

PREVIOUSLY BY BENJAMIN WENDORF 

 

417ea714751a4580f06253a09a79bab0
Benjamin Wendorf was co-manager of the SB Nation Winnipeg Jets blog Arctic Ice Hockey (formerly Behind the Net); he is currently in the PhD program in Africology at the University of Wisconsin at Milwaukee. You can follow his graph work on Tumblr or his tweets @BenjaminWendorf. He can also be reached at bwendorf84 AT gmail DOT com.
Avatar
#1 Clay
December 10 2012, 08:54AM
Trash it!
0
trashes
Props
0
props

Thats awesome

Avatar
#2 Steve Burtch
December 10 2012, 09:02AM
Trash it!
0
trashes
Props
0
props

Very interesting - I'm reminded of one of the iterations of FIFA (I forget the year) where they had radar plots for player attributes as a visual representation.

This is purely usage, which goes back to the idea that coaches use their players effectively (which may or may not be a solid assumption).

What this settles for me is what coaches think of players, not necessarily the best usage of players. Still interesting nonetheless.

Thanks for the write up and intro to the idea.

Avatar
#3 draglikepull
December 10 2012, 09:09AM
Trash it!
0
trashes
Props
0
props

Some really great stuff in this post. Here are a couple things I picked up on that you didn't mention:

1. PK time seems to be split pretty evenly among 1st/2nd/3rd line forwards with 4th liners not actually seeing much PK time. The standard idea is that 3rd and 4th liners are your penalty killers and that you need to have some guys who can PK in the bottom 6 but from what you've presented here it's not clear that's the case.

Relatedly, something I haven't looked at in any detail but that I've sort of noticed generally is that teams seem to play different players on the PK during the regular season and the playoffs. For example, during the regular season last year Pavel Datsyuk was 7th among Detroit forwards in SH TOI at 1:13 per game. During the playoffs he was 2nd at 2:47, just 3 seconds behind the team leader. That's not an accident. Mike Babcock knows that Pavel Datsyuk is his best penalty killer but doesn't want to risk him getting hit with a slapshot or something during the regular season. Anyway, I'm getting away from your charts here, but I think what you've got here should make us rethink who kills penalties a bit.

2. Teams don't seem to understand the aging curve of forwards. Guys in their late 20s and early 30s seem to get as much or more TOI than guys in their mid-20s, which doesn't make sense given what we know about peak scoring. You could argue that forwards get better at defence as they age, which might justify their increased 5v5 time, but 5v4 TOI for these players remains high too. They're clearly not having their responsibilties shifted from offence to defence as we might expect if it were true that increased defensive acumen was the reason they still got big TOI.

Avatar
#6 Mackeneasy
December 10 2012, 02:46PM
Trash it!
0
trashes
Props
0
props

I am wondering if it would be possible to overlay all of the players from one team in some format in order to give you an idea of where the weaknesses lie on a team.

Other stats may need to be added, or perhaps there are other stats that this info-graphical format would apply to.

Thanks

Avatar
#8 V
December 10 2012, 10:22PM
Trash it!
0
trashes
Props
0
props

I think it would be interesting to see Sedins + Burrows or other lines with players of clearly different mindsets and roles.

Comments are closed for this article.