View Single Post
05-21-2013, 12:40 PM
Student Of The Game
seventieslord's Avatar
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 30,969
vCash: 500
Originally Posted by quoipourquoi View Post
I'm not ignoring team influence. In fact, I was discussing it with you in the Top 40 thread until you stopped responding to legitimate points raised by both TCG and myself.
Thanks for pointing out that there were replies there. I hadn’t realized. I was watching for bolded threads and for whatever reason that thread never showed as bold all weekend. I’ll answer those points over there.

Only four goaltenders have led the league in save percentage as a starter multiple times since the statistic became official: Roy, Hasek, Belfour, and Thomas. Thomas did it with leads of .007 and .008. Belfour did it with leads of .004 and less than .001, meaning we haven't seen a goaltender do what Thomas did in those two seasons since Hasek in 95/98/99 and Roy in 89/90. And Thomas' save percentage jumps up in the playoffs both in terms of cumulative numbers and game-to-game consistency, with one of the runs being a .940 Conn Smythe when the league average was .911.
We’re talking about two different leagues though. I think that, for the most part, save percentage was closely tied to actual goaltender performance. With the exception of Martin Brodeur, from 1987 to 2004, it seemed that the goalies considered the best were also the ones who had the highest save percentages most frequently. But team style of play and coaching have distorted that lately. People should be somewhat skeptical of goalies such as Thomas/Rask, Rinne, Backstrom, and Hitchcock and Dave Tippett’s recent creations.

It’s funny, save percentage just seems to have finally won the stats battle as the best simple stat that exists for goalies, but just as it did, team effects became more pronounced and now we’re in more danger of overusing the stat than ever before.

Lundqvist has consistency, but Thomas has an incredibly high peak (I just compared it to Hasek and Roy) that makes up 25% of the eight year sample. You realize Thomas' leads over his own backups during his Vezina years were by .023 and .020, yes? The fact that Rask is better than the New York backups doesn't make Lundqvist better at stopping the puck; Rask could've come into the league as a #1 goaltender and been successful.
– I readily acknowledge that Rask as a comparable is better than what the other goalies have typically had to contend with. That said, he still hasn’t been tested in a variety of situations. In other words, his greatness as an NHL goaltender has not been definitively proven.

-********* No one said Thomas was a bad goalie. In his Vezina years, he did badly outplay the other goalies on his teams. And I’m glad you see that as a useful stat, because Lundqvist outshines the entire NHL in that regard.

So for Lundqvist to be higher than Thomas in an eight-year sample, he should either have two save percentage titles (his highest finish is fourth), a higher cumulative save percentage (he doesn't), a better playoff resume (he has a penchant for playing as many below average games as above average games), or more trophies (he collected fewer than 10/30 Vezina votes every season from 2007-2011 for having high-GP stats, only winning one Vezina to Thomas' two and a Conn Smythe)

I mean, the difference in GP is practically the same as the difference in GP between Alexander Ovechkin and Sidney Crosby. It's not that big of a deal, considering Boston was trying to play Rask as much as possible since he was the goalie of the future. If you think a concentrated number of GP would have negatively impacted Tim Thomas, may I suggest the 2005-06 season (38 games in 95 days; 7th in save percentage) and the 2011 playoffs (25 games in 62 days; Conn Smythe)?

-********* If we’re trying to distill their eight season performance into one number, then let’s stop trophy counting and look at cumulative save percentage. Because if trophy counting matters, it must be acknowledged that for him to have an extra vezina with numbers that aren’t any more impressive, his lows had to be lower… right?

-********* While it’s correct that Thomas has him by two points, the GP difference is a massive one. Averaging numbers tend to regress to the mean in a larger sample size. If you could replay the last 8 seasons and find 137 more games for Thomas, would you place a bet, with even odds, that he maintains a .922 save percentage over that larger sample? Everything that we know about statistics indicates that this is not a smart bet to place.

-********* I disagree that this is similar to Crosby and Ovechkin in terms of the magnitudes involved. Crosby over the last 8 years has cumulatively been a 15.7% better producer than Ovechkin. Ovechkin has played 27.9% more games. In the case of Lundqvist/Thomas, Thomas has an edge of 1.3% (his error rate of 7.849% is 1.3% better than Lundqvist’s 7.953%; what I also noticed is that their sv% gap of “two points” is a rounding anomaly, because they’re actually just a hair over a point apart, .92151 to .92047), and Lundqvist has played 36.7% more games.

-********* People confidently state that Crosby is the best post-lockout forward, and for good reason. His production edge is far enough ahead that the difference in GP is not significant enough to change our confidence in that. Looking at Ovechkin’s 1.22 PPG in 601 games, if Crosby was just 1.3% better in points per game and played 36.7% fewer games, it would not be fair to call him a better player. (Crosby would have 1.236 PPG in 440 games – 543 points).

Originally Posted by MadArcand View Post
Remind me again how close did Lundqvist get to a Conn Smythe?
seriously, that’s all you have to say to that?

seventieslord is online now   Reply With Quote