Dwayne Roloson, Welcome to a Land of Positive Shot Differential
by Bettman's Nightmare on Jan 6, 2011 8:00 AM EST
Email
Print"It's always exciting when you're playing for a team in that is in first place." -- Dwayne Roloson, after his 1-0 blanking of the Washington Capitals on Tuesday.
Maybe it wasn't a big deal for a lot of people, but 41-year old Dwayne Roloson's trade to the Tampa Bay Lightning has given an old veteran perhaps one last chance at a championship. In the process, he has also joined a team with the number one shot differential in the league (as of the end of the day January 1st). The Lightning and their potent, Guy Boucher-led offense have averaged 32.41 shots per game (SPG) this year, a full five shots higher than the average of Roloson's teams in his career (27 SPG). It will also be the greatest volume of shots per game that Roloson has ever played behind; his next highest was with the Calgary Flames in 1996-97 (31.35 SPG) and the Islanders last year (30.4 SPG). In no other season have his teams averaged above 30.
But the amount of shots-for is only half of the upside.
The shot differentials of his teams have either been negative systematically (Ruff's Buffalo Sabres, Lemaire's Minnesota Wild), or just because they were pretty bad teams (Edmonton Oilers after '05-'06, New York Islanders).
Season --- Team --- SPG --- Shots against/game --- (Diff) --- NHL Rank
1996-97 --- Flames --- 31.35 --- 28.43 --- (2.93) --- 6th
1997-98 --- Flames --- 27.68 --- 27.68 --- (0.00) --- 13th
1998-99 --- Sabres --- 26.22 --- 29.96 --- (-3.74) --- 24th
1999-00 --- Sabres --- 27.18 --- 26.85 --- (0.33) --- 11th
2000-01 --- Wild ------- 24.63 --- 28.09 --- (-3.45) --- 23rd
2001-02 --- Wild ------- 23.83 --- 28.00 --- (-4.17) --- 26th
2002-03 --- Wild ------- 25.44 --- 28.46 --- (-3.02) --- 24th
2003-04 --- Wild ------- 24.17 --- 29.01 --- (-4.84) --- 27th
2005-06* - Wild/Oilers 28.49 --- 30.09 --- (-1.60) --- 23rd
2006-07 --- Oilers ----- 26.74 --- 29.59 --- (-2.84) --- 24th
2007-08 --- Oilers ----- 26.29 --- 31.26 --- (-4.96) --- 29th
2008-09 --- Oilers ----- 28.09 --- 32.48 --- (-4.39) --- 27th
2009-10 --- Islanders 30.40 --- 31.89 --- (-1.49) --- 20th
2010-11 --- Islanders 27.97 --- 31.58 --- (-3.61) --- 26th
In the 2005-06 season, Roloson experienced a similar boost to his fortunes, moving from the 26th-ranked shot differential in Minnesota to the 3rd-ranked shot differential in Edmonton. That Oilers team would reach the Cup Finals thanks to the solid play of Roloson and a fellow recent acquisition, Chris Pronger.
As you can see, Roloson's career will forever be marked by playing pretty darn well (.920 ES SV% since 2001-02, stdev 1.1) for some pretty questionable teams. His teams' shot differential over his career (-2.44) versus the rest of the league during that time (0.18) is a testament to that. Frankly, I don't know what he'll do behind a team like the Lightning; it would be foolish to compare the '05-'06 Oilers to the '10-'11 Lightning, and not just because they don't have Pronger. But I can confidently say that, unlike a number of quality veterans, Dwayne Roloson is going to truly get one more chance to win it all.
* The season was split between the Wild and the Oilers; I took the Wild shots-for/shot-against from all the games played to the day of his trade in the case of the Wild (March 8th) and added those to the 20 Oilers games after the trade. The rank is based on where the differential would be placed in the league for that year.
16 comments
|
0 recs |
Do you like this story?
Comments
Save Percentage and Shots Against Total
Has there been any analysis done on the correlation between the number of shots faced and the save percentage? I get the impression that goalies that face more shots per game typically get better save percentage. Is it true?
You know, I’ve been asking myself the same question recently. Snooping around a little, I found Phil Myrland’s post over at Brodeur is a Fraud on this question. Be sure to read the comments, as there are a couple of critiques that lead to further clarification by Phil and another reader (“overpass”).
Does he call it Luongo underwear?
Co-Manager at Behind the Net
by Bettman's Nightmare on Jan 6, 2011 3:26 PM EST up reply actions
To paraphrase Phil, it looks like the answer is no.
Does he call it Luongo underwear?
Co-Manager at Behind the Net
by Bettman's Nightmare on Jan 6, 2011 3:27 PM EST up reply actions
Incidentally, be sure to look at this post by Gabe on the number of shots-against before we can be pretty safe in considering an even-strength save percentage to be reflective of a goalie’s talent. You don’t usually see the 6,000 even-strength shots-against mark reached until a few seasons have passed. Roloson’s definitely there, even in the snippet I referenced from 2001-02 to the present (8,804 shots-against). Before that mark, there could be a bit more variation between a player’s numbers and their true talent.
Does he call it Luongo underwear?
Co-Manager at Behind the Net
by Bettman's Nightmare on Jan 6, 2011 3:41 PM EST up reply actions
1. Theoretically, there should be a slight positive correlation because of the existence of recording bias.
2. Back in late 2008, I looked at data for each season from 1997-98 to 2007-08 and found no correlation between shots faced/minute and save percentage for league goaltenders that had reached the minimum shots against threshold. I attribute the absence of a correlation to the fact that there was less parity in the league during this time period as compared to now, and that the league’s better defensive teams allowed both fewer shots and fewer quality shots, thus counteracting the effect of recording bias. In fact, this is supported by the shot quality data generated by Alan Ryder – in the pre-lockout period, there was a positive correlation between shot quality against and shots against at the team level.
3. In the last three seasons, the amount of parity has increased in the league to the extent that the above effect has likely disappeared. While I haven’t looked at the relationship between shots/minute and save percentage over this more recent period, there does appear to be a negative relationship between a team’s shot differential and its PDO (shooting and save percentage) at EV, but only when the score margin is other than zero. When the score is tied, there is no such tradeoff. Similarly, there does not appear to be a relationship between the number of shots faced on the penalty skill and save percentage.
So one would likely find a slight positive correlation between shots/minute and save percentage over the last three seasons, given the existence of both recording bias and increasing parity among teams.
JLikens is correct. There is a small positive correlation owing to recording bias and score effects. Obviously, in a less balanced league, there would be a negative correlation, as better goalies play for better teams (think Ken Dryden or Bernie Parent in the 70s, or whoever plays for Team Canada in the Olympics).
Thank You!
for enlightening me. So it’s probably not that much of an effect.
I went with sub 91% and 25-29 wins. Call me crazy, but I don’t think it’s 100% the goalies fault they have been so leaky. I know that Tampas defense has been limiting shots and have an excellent PK, but I don’t see any reason why both Ellis and Smith should have such appalling numbers in the same season, especially compared to their averages.
You could argue that Ellis played well in NSH because he had an excellent defense in front of him, and Smith hasn’t been the same since his concussion last season, but I’m prepared to be a goalie apologist. Roloson is a better goalie than both of them, but I don’t think his numbers will be great.
Mourning Gagne forever.
The only thing I can say about Ellis and Smith is that we never really got a good taste of what they could do; they have had the starting job before and, during that time, were league-average to below league-average, but that was only for less than a year. Just taking a glance at their numbers outside of the NHL, and I’d wager a guess that they are better than the first-half numbers, but still below league-average to league-average goalies. Roloson, to me, is a league-average floor with above-average upside.
Does he call it Luongo underwear?
Co-Manager at Behind the Net
by Bettman's Nightmare on Jan 6, 2011 5:13 PM EST up reply actions
Going into this season, even-strength save percentage since the lock-out
Dan Ellis .9229
Mike Smith .9152
Dwayne Roloson .9152
Is that including this year? Because I’m not getting the same numbers…at all.
Does he call it Luongo underwear?
Co-Manager at Behind the Net
by Bettman's Nightmare on Jan 7, 2011 2:33 PM EST up reply actions
(Forehead slap) I see, it isn’t, duh. Including this year (why would you cut that out? Smith and Ellis suffer from criminally small sample sizes…)
Ellis
- ESSV% 91.778; PKSV% 91.509
- combined 91.769% across 2,940 shots
Smith
- ESSV% 91.121; PKSV% 91.566
- combined 91.143% across 3,421 shots
Roloson
- ESSV% 91.608; PKSV% 92.982
- combined 91.656% across 6,520 shots
Does he call it Luongo underwear?
Co-Manager at Behind the Net
by Bettman's Nightmare on Jan 7, 2011 2:46 PM EST up reply actions
Wow, outside Roli’s PKSV%, those are remarkably similar numbers across the board. Almost spooky. But again, he wasn’t worth the two-year risk while Khabibulin was worth the four-year risk at what, three times the money? /eyeroll
I went with 91-92% (closer to his career average), <25W. Tampa’s halfway through their year, and Roloson (now 7-14-1) would have to win at least 18 games in that last 41 games to do it. If he played all of them, sure, no problem, but he’s not going to. Even if he plays 30, which would be quite a bit, especially at his age, 18W would be a hell of a mark. It’s not impossible, but I think he comes up just short; I’ll register 22W as a guess, just for shits and giggles.
SNN Sports - A theoretical Oilers blog (i.e. theoretically, I write stuff there). Link now 100% less broken.
Honestly, the only difference I see is sample size, which makes me literally and statistically more confident in Roloson’s floor versus Smith and Ellis.
Does he call it Luongo underwear?
Co-Manager at Behind the Net
by Bettman's Nightmare on Jan 8, 2011 2:48 PM EST up reply actions
Nashville overcounts too, I think.
My blog and Twitter, featuring coverage of the most frustrating team in the NHL
If you don't know how to use Timeonice, read this.
Behindthenet quick link to QoC/QoT/Corsi/PDO/Zonestarts
"Numbers don't lie, they just don't agree with you"--George E. Ays
by red army line on Jan 9, 2011 10:33 AM EST up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Arctic Ice Hockey
FanPosts
Community blog posts and discussion.
