Entropy: shot differential, defense and goaltending (first-quarter report)
SCH NOTE: GMH put up this great post on Friday that deserves a look. I wanted to save it until today though so it wasn't lost in the weekend craziness - plus now we have a few more days until game time so it's the perfect situation to take a closer look.
I've been keeping this little spreadsheet for fun, but at the quarter-mark of the season, it might be worth taking a look at some of the trends and implications (if there are any to be found) through the first 20 games. Uh, keep in mind that I'm kind of a statistics retard; I read all the articles at Puck Prospectus but I probably couldn't sum them up for you in any coherent manner, and I'm probably missing some of the finer things in my analysis, so any contributions you guys could make in the comments would be A+ and much appreciated.
|
STATING THE OBVIOUS
- The 'Hawks are vastly outshooting their opponents. This is because the defense has generally played incredibly well, both in their own end (getting sticks and bodies in the way of shots) and in contributing in the offensive zone (logging 59 points so far on the season, tied with SJS for the league lead). I added the Blocked Shots column in as an afterthought, although only one number really stands out to me, and that's the lone blocked shot in the comeback against Calgary. But more on those two games later.
- Huet was shaky early in the season, and Niemi, despite the shutout against the Panthers, was only marginally better. Since the disaster against Dallas on 10/17, the goaltending has looked much more stable. Most goaltenders experience peaks and valleys throughout the course of a season, so it's likely we see a few dips along the way (like during the fathers' trip), but it's probably still too early to evaluate this team's goaltending.
- On a related note, does anyone know where I could find Player Contribution stats for Huet and Niemi? It seems like we'd gain a more honest appraisal of their work between the pipes if we could get a better idea of the quality of shots they're facing.
SIMPLE TRENDS
- Last night's rout of the Flames aside, the 'Hawks have lost both of the games in which they were outshot, despite getting great goaltending in those games. The Vancouver and Nashville games, unsurprisingly, were the only two games where the 'Hawks, goaltending aside, really looked bad as a team.
- Also unsurprising: The 'Hawks have a hard time winning--or at least putting teams away--when their power play malfunctions. They went 0-4 and 0-5 (respectively) on the power play against Vancouver and Nashville. Let's look at the other regulation losses: 1-5 in Detroit, 0-4 against Dallas (although the fluke goals also factored into the end result), and 0-6 in Phoenix.
THE POTENTIAL FOR DISORDER: OR, WHAT HAPPENS WHEN FLAMES ARE APPLIED
- The two games against Calgary this season have resulted in 1) the craziest comeback we're probably going to see this season, and 2) the biggest scoring margin so far. I don't know if anyone can argue that the numbers from these two games are highly atypical of some of the trends we've been seeing.
- The shock value of the first game, the comeback, came from how quickly the Flames put up a five-goal lead. It took only 12 minutes, or 1/5 of the game, which meant the 'Hawks still had 80% of the time left on the clock to work with. The two numbers that stand out, of course, are the single blocked shot, and the +20 shot differential, which maybe speaks to the chaotic nature of a game that essentially transplanted the Hare and the Tortoise fable onto a sheet of ice. In any case, one can see that terrible turnovers and bad goaltending led to the ghastly deficit, but it becomes equally obvious that the 'Hawks outplayed the Flames in every other area of the game, and deserved to win that one.
- And that brings us to Game #20, last night's ritual flaying of Kiprusoff and the Flames. The only reason why the 'Hawks ended up with a -2 shot differential was because they were able to put up a five-goal second period. The Flames were stagnant at 19 shots for about 10 minutes spanning the second and third periods, but weren't able to apply any pressure until it was too late, and anyway, Huet was on the top of his game last night. It's hard to say how the game may have swung had he not made that incredible dying-swan stop on Bouwmeester when the game was still in reach for the Flames at 2-1. Or had Hjalphabet not gotten all up in Bourque's grill. Or had the power play not woken up like Vesuvius with Pompeii in its sight line. It's a nice little reminder that small plays can matter a great deal over the course of 60 minutes.
9 recs |
45 comments
|
Comments
It’s strange how the two games against Calgary unfolded when you consider how the Hawks have performed against everyone else so far. Lucky for the Hawks it was in their favour. One can only imagine what the third game between these two will be like.
It's never about the eventual destination, but rather the long journey and its challenging obstacles that are presented and what it takes to overcome them, that makes the taste of success all the more worthwhile!!!
Re: Entropy
After our dreaded loss at home to the Canuckle Heads, the goaltending has been top notch.
I find it amazing how many people still cry for Khabby, ignoring what he did the first 3 years of his 4 year contract. (Edmonton can expect great goaltending in a few more years at the end of his contract)
Huet was the candle under his arse last year. (just as Niemi is for Huet this year) Competition is a great thing. After the job Niemi did Sunday, look for Huet to stand on his head Wednesday.
Khabby was also oft injured, and last year Huet was huge when Khabby went down in February. Foley commented that he may have contracted the “Chicago flu” on Saturday, and many wondered if he got the Detroit flu in the playoffs last year. (and more likely, he has a serious back problem, oh, I mean “lower body injury”)
Another point is that Khabby was notoriously weak in Shootouts. No way he was going 9 rounds with anybody. Huet and Niemi have been awesome in OT and shootouts this year. I wounder if they have stats to compare them to the rest of the league in that area.
Also, it’s no secret that the Hawks essentially swapped Havlat for Hossa in free agency. Most likely because Havlat was oft injured (as he is now) except for the final year of his contract. Obviously Hossa has been out, but once he returns, which guy will be more durable?
It would be nice if everyone would get behind Huet and forget about his rough start. He is doing a great job!
Actually
according to Puck Prospectus, Huet has the worst GVT (-1.3) in the league in the shootout, although I don’t even pretend to know how that’s calculated. He’s 2-1 in the coin toss this season, with his only loss being the opening game in Helsinki. He was great in the nine rounds against Colorado, and pretty good in the second go-round with Colorado; Niemi, on the other hand, is 0-1 after losing in Colorado, but he kept the ‘Hawks in it for eight rounds, so in my judgment, there’s not much more you could ask of them.
GVT
GVT: Goals Versus Threshold. The value of a player, in goals, above what a replacement level player would contribute. GVT aggregates offensive, defensive, goaltending and shootout contributions in a single metric. The sum of player GVT on a team equals that team’s goal differential plus the replacement level, which is 50% of the average team scoring level.
nGVT: GVT, normalized for league scoring level of 3 goals per game and a schedule length of 82 games. Typical values: 0 is a replacement player, 5 is a third-liner or #4 defenseman, 10 is a second/first liner or #2 defenseman, 20 is a top-10 league scorer or league top-3 defenseman or goaltender, 30 to 35 is MVP-caliber.
so overall Huet’s GVT is still a minus at -0.9 largely based on the early struggles…but when your team currently has the #2 and #4 current GVT top rated D-men in Seabrook and Keith respectively…that tends to help.
by Crease Monkey on Nov 23, 2009 5:20 PM CST up reply actions
I guess shootout GVT just refers to the number of shots he lets in versus the number he faces, but I still don’t know how Huet is the worst in the NHL this season when he went through (and won) that 9-rounder and is 2-1 overall.
sdflkk like I said, I’m a little bit retarded when it comes to sabremetrics.
Shootout GVT isn’t much different. It’s still the value of a player, in goals, above what a replacement level player would contribute in the shootout. It just requires you to eliminate the strength of defense metric that would normally factor into the basic GVT.
Since COL and FLA are the only two teams the Hawks have had shootouts with you just factor their strength on offense… or lack thereof. FLA is currently 24th in the NHL in GPG leading to a -6.8 GVT offensive GVT. FLA also has the league’s worst D rating at a -8.3 GVT. I couldn’t find PP’s rating for COL’s offense, but their success has largely come from their top ranked G at an inflated +10.7 GVT in conjunction with their mid-tier D only contributing +1.9 GVT so you’d have to assume COL’s offensive GVT rating isn’t very high.
by Crease Monkey on Nov 23, 2009 6:23 PM CST up reply actions
I’m sure you’ve got a handle on it. It will be much easier when PP evolves into what hopefully will be a BaseballProspectus like site, and you have a search-able database and more readily available stats. Of course, then you’ll be another $40 lighter per year to read it.
by Crease Monkey on Nov 23, 2009 7:01 PM CST up reply actions
Sopel rumors
“the Maple Leafs are rumored to be willing to take Sopel’s contract if Chicago includes a second round draft pick.”
I thought I was tripping
when I saw this pop up in my Google Reader. Thanks for the bump; hopefully people can shed some more light on the data.
Last night’s game kind of shattered the shot differential trend a little, but at least we can point to the outstanding goaltending performance as the explanation. It’ll be interesting to see, also, how these things play out at home vs. on the road; maybe I’ll add that contrast at the halfway point when there’s more data to look at.
Interesting Stuff
In its most basic form (which is where my simple mind usually works) when the Hawks outshoot their opponents and do OK on the PP, they win. Sounds like a pretty simple winning formula.
EXCEPT, that our boys have a rather incredible differential in shots taken and allowed, which seems to be mostly a product of their great defense, as you mentioned. But I wonder how much is also because of pure puck possession time. The other team can’t shoot if we’re controlling the puck. After watching our top line keep the puck in the offensive zone for a full minute against the Nucks last night, I’m wondering where our team ranks in possession time. Surely, someone’s got the stats on that, don’t they?
Is it maybe coming up in parts 2-4?
Thanks for doing the work on this. Great stuff!
puck possession time
would be a great addition to this data; unfortunately, I’m not sure how to find those stats without actually going and reviewing each and every game with a stopwatch.
I do agree with your overall assessment, though — the ability of this team to not only keep the puck, but also to recover it when they don’t have it, is largely contributing to their dominance in shot differential.
Crap, I meant to take out the (part 1 of 4) because it was actually referring to a first-quarter report, not … a report in four concurrent parts.
Apparently, it's NOT available
Just doing a simple Google search of puck possession reveals nothing. I’m surprised—I thought they kept a stat line like that along with the players’ time on the ice. In fact, one article I read (from last season) talked about the LACK of that particular statistic, and how it SHOULD be kept.
And my mistake—it was in the Oilers game that the Hawks had a full minute in the offensive zone on one shift. The Hockey Night in Canada guys were pretty impressed.
Thanks for this first quarter report. It will be interesting to see how it continues in the next 20 games.
by Preacher000 on Nov 23, 2009 10:00 PM CST up reply actions
a puck possession stat would be tits.. but yeah it seems impossible to track.
Come on NHL – help the American (and Canadian) economies and pay someone to keep track of this.
by Matthew Dirt on Nov 24, 2009 8:47 AM CST up reply actions
Even if its not in the game...
its still in the game.
twitter.com/BlackhawksDL
by Original Six on Nov 24, 2009 2:38 PM CST up reply actions
Awesome post
The Defense is playing damn solid hockey where in fact these stats show as well as the point total being put up. Also, Huet has been solid lately and I think he excels when it comes to positioning and that has been evident lately.
jt
by HolyBlackhawksBatman on Nov 23, 2009 6:26 PM CST reply actions
speaking of shot differential
TOR put 61 SOG tonight (including 12 from Kessel), and lost to the Islanders in OT. 61-21 SOG totals.
I watched that game
Roloson was a fucking ninja. Leafs did well to come back and tie it, but that overtime goal was a sloppy one to give up.
I wouldn’t be surprised if those numbers were flipped – 61 shots.. only 12 not from Kessel
by Matthew Dirt on Nov 24, 2009 8:49 AM CST up reply actions
Just for the record
Sportsline Power Rankings have the Hawks at 2 and ESPN has them at 1. That’s all I have seen so far.
Big shot differential is nice
…but I kinda like the more conservative style used on this trip too.
There is a balance to be struck between controlling the puck in your own end and getting it the hell out of your zone.
The first strategy produces some of the sick time of posession and shot differential stats but some scary turnovers and great chances. Goal differential did not track the lopsided shot differential from earlier this year.
On this trip, the d have been more inclined to clear the zone. We lose time of posession and give up more shots, but they are of the conventional “carry or dump it in” variety rather than occasional “opposition guy in front of the net alone with the puck”-type.
In San Jose anyway, I vote we continue clearing the zone. At home, posession is better, because we can preserve favorable match-ups due to the last shift.
Agree
And the visiting team centerman has to have his stick down first on faceoffs. I think there’s a much different approach to playing a road game vs. home game and it’s not just the hostile environment. I thought the Hawks played smart in Vancouver and our puck possession game is going to be tested in SJ with their ability on the face off and their very, very talented top line. The return of Hossa will be great but it will be up to Madden/Ladd/Brouwer to decide that game.
The Return of Hossa
Please be respectful and use all caps when typing that phrase.
I be on my green like Irish Spring.
JUST PUT THE WHOLE THING IN CAPS. PEOPLE GIVE YOU MORE RESPECT. AND ADD SOME ’S.
twitter.com/BlackhawksDL
by Original Six on Nov 24, 2009 2:41 PM CST up reply actions
I feel like a little kid on christmas eve...
waiting for tomorrows game. It’s like knowing that red rider BB gun is under the tree, but you just can touch it yet. Tomorrow though…you will be shooting the shit outta everything.
I am also not very patient and give into my need of instant gratification almost always. So this is killing me.
Pain is a good reminder you are alive
by Bad Ass Burish on Nov 24, 2009 12:53 PM CST reply actions
I know the feeling.
Compound that by realizing that you mistakenly bought tickets to fuggin cirque du soleil for the lady friend, not realizing this would be The Return of Hossa in the midst of a 7 game win streak against the only team better in the standings.
Thank god its a west coast game, hopefully I won’t miss too much.
I be on my green like Irish Spring.
This is the one time it is acceptable to buy a VCR. Just get a blank tape and record it. Desperate times…
twitter.com/BlackhawksDL
by Original Six on Nov 24, 2009 2:42 PM CST up reply actions
Chicago, Chicago
Only here would getting ranked #1 bring this kind of reaction out.
by chiblackhawks on Nov 24, 2009 2:29 PM CST up reply actions
been meaning to reply here for a while now
because I love the numbers of the game, though they don’t always make sense to me.
Stuff I’ve extrapolated:
SV% in loses (5) – 0.889
SV% in SO loses (2) – 0.869
SV% in OT wins (2) – 0.81*
SV% in SO wins (2) – 0.884
SV% in wins (9) – 0.941
- Since both Niemi and Huet played in one game, I just took the combined 15 saves out of 20 shots (75% SV) as the SV% of that game
The only times we’ve lost in regulation when goaltending was above the 0.900 was when we were outshot. All losses otherwise recorded below-0.900 save percentages.
The only time we’ve won in regulation with the SV% below 0.900 was that home game against Edmonton. We’ve won twice in the OT (SJS, CGY) and once in the SO (COL) with below-0.900 save percentages— I’m not touching Calgary but I’m guessing limiting San Jose to 23 shots factored a lot in winning against them, as they had deadlier scorers on their line.
We’d probably need more data later on in the series, and of course there will be outliers and what-have-you, of course. I feel stupid pointing the obvious out, but there’s definitely a positive correlation between the performance of our goalies to the results of our games, so if either of them hit a cold streak, we’re likely fucked.
Speaking of power rankings
Much props to Scott Cullen from TSN for giving mathematical quantification of the TSN power rankings. Unlike The Hockey News, who still has The Scum ranking in at #8.
We need to rec this post.
Even though it’s been frontpaged, it’s still going to get pushed down on the right rail by onslaught of dimwittery that is all too common lately.
Preserve the valuable Fanposts: give them a rec.
NOW STOP IT RIGHT HERE
I rec'd it (belatedly)
also, the Cycle Like the Sedins blog linked to this piece…
I think a certain recent piece of “dimwittery” did vanish from here, didn’t it? of course I wish it hadn’t appeared in the first place…
a little chippy: hockey, mostly
"if you don't watch the violence, you'll never get desensitized to it!"

by 

















