Wednesday, July 1, 2009

A Look at UFA Goalies: Is Craig Anderson a Starter?

One of the most interesting free agent goaltenders this year is Craig Anderson, the Florida understudy who has put up some eye-popping stats as a backup over the last 3 seasons (24-14-7, 2.52, .928).

I don't think Anderson will continue to put up those numbers if he was given a shot at a #1 job, because they are likely at least to some degree a result of a smaller sample size. However, the evidence does suggest that he is a legitimate NHL goalie.

Even if you don't trust his most recent couple of years, his .920 save percentage over his last 3,000 shots against at the AHL level is evidence of a goalie that is very good at the minor-league level. Florida must also hold Anderson in fairly high regard, since he managed to take 27 starts away from one of the league's best goalies in Tomas Vokoun.

Two areas we have to usually be concerned about when evaluating backup goalies are whether their numbers have been impacted by their performance in relief appearances, and whether they are playing easier opponents. The first one is not an issue for Anderson as his numbers are very similar both in games he starts and games he enters. Opposition hasn't been a concern either, over the last 2 years 21 of Anderson's 40 starts have come against playoff teams, which is a decent ratio considering that only 3 of Florida's 8 divisional opponents over that time span ended up making the playoffs.

Last year Anderson actually did much better against playoff teams (4-2-0, 1.82, .949). This year he fattened his stats a bit on the weaker teams (8-1-2, 2.08, .940), and had a few rough outings against some elite opponents (7-6-2, 3.18, .913 against playoff teams). I think a team that signed him to a bargain price would be pretty happy even if he could just hold onto that .913.

Anderson just turned 28, so age is not an issue. I think I'd place my bets on him if I had a team that was up against the cap and needed goaltending help. I think Anderson should be at least capable of putting up league average numbers over an entire season. It's difficult to fully assess his ability at this point until we get more information, but what is there looks pretty promising.

Here are some of the other key UFA goalies and their salaries from last year, as well as their even-strength save percentages and number of even-strength shots faced since 1998-99:

Wade Dubielewicz ($0.500): .929, 817 SA
Ty Conklin ($0.750): .924, 2930 SA
Manny Fernandez ($4.333): .923, 6724
Martin Biron ($3.500): .921, 9184 SA
Craig Anderson ($0.575): .921, 2290 SA
Nikolai Khabibulin ($6.750): .920, 9734
Dwayne Roloson ($3.000): .920, 8506 SA
Martin Gerber ($3.700): .920, 4726 SA
Scott Clemmensen ($0.500): .920, 1364 SA
Manny Legace ($2.500): .919, 6343 SA
Mathieu Garon ($1.000): .918, 4192 SA
Kevin Weekes ($0.700): .916, 6763 SA
Brent Johnson ($0.825): .916, 4619 SA
Curtis Sanford ($0.650): .916, 1895 SA
Andrew Raycroft ($0.800): .914, 4644 SA
Jason Labarbera ($0.800): .912, 2104 SA
Brian Boucher ($0.650): .910, 4703 SA
Antero Niittymaki ($1.375): .910, 3315 SA
Joey MacDonald ($0.500): .908, 1592 SA

(Source: UFAs and salaries from HFBoards, ES SV% stats from NHL.com)

League average over the period has been .917, so none of the guys are close to the elite level, but a few of them are serviceable starters. I wouldn't mind too much having Roloson, Khabibulin, or Biron in my net next season, although I wouldn't sign any of them to long-term deals or pay them anything more than $2-3 million. Fernandez should also be able to provide somebody with average play.

The goalies teams might want to roll the dice on would be Dubielewicz, Conklin, Anderson, and Clemmensen (or Jonas Gustafsson, of course, but that's another discussion). Of those 4, I'd probably rate Anderson as the best bet at being able to fill a starter's role, although I wouldn't hesitate at all to sign Conklin as a backup. The other two likely got somewhat lucky over a small sample size. Dubiewelicz is intriguing, he has a .920 career AHL save percentage and has been pretty decent in some NHL spot duty, although I'd like to see more. Clemmensen was pretty ordinary in the AHL last year and doesn't have the supporting minor league numbers; I'd be very leery of him as a small sample size fluke last year playing on a team that has typically been a potential shot quality outlier.

8 comments:

sunnymehta.com said...

Great analysis.

It's interesting to see Khabi go for 3.75M to Edmonton, while Conklin and Anderson go for 1.3 and 1.8 to STL and COL, respectively. I mean, on the one hand, our confidence in Khabi's SV% is much higher than the other two because he's done it for over four times as many shots. On the other hand, how does age factor in? Khabi is past his prime (though I've heard goaltenders seem to defy or at least delay the age curve somewhat). If you're Edmonton, why not just re-sign Roloson and save yourself 1.25M?

I was really hoping the Devils would've taken a flier on Conklin or Anderson. Would've been a perfect situation to take a cheap low-risk gamble (as they wouldn't be relied upon for big minute duty) for possible big payoff (if they retain their form and/or Brodeur declines).

Anonymous said...

Khabby has pedigree and perhaps they think they may need all four years of his contract as a safety net to transition Next Young Goalie to starter status, and clearly weren't willing to give four years to Roli since they wouldn't even give two. If you're wondering about age decline, the three-year difference between goalies might be the difference.

I was really hoping the Devils would've taken a flier on Conklin or Anderson. Would've been a perfect situation to take a cheap low-risk gamble (as they wouldn't be relied upon for big minute duty) for possible big payoff (if they retain their form and/or Brodeur declines).

I don't think Conklin or Anderson would've wanted to take a flier on the Devils. Much like Clemmensen, I believe they were all looking for the opportunity to play more than once a month, something that also helps "retain your form".

believe it or not "Marty Brodeur's backup" is not exactly what most goalies with a shred of talent want to be.

eyebleaf said...

Great post. I'm surprised Raycroft's save percentage is that high. I hate him. With all of my being.

The Contrarian Goaltender said...

Eyebleaf: Those numbers are since 1998-99, so they include Raycroft's excellent rookie season. We can break it down further like this:

Pre-lockout: .932
Post-lockout: .905

The latter Raycroft (.013 below league average) would be the one you know and love.

Bruce said...

Two areas we have to usually be concerned about when evaluating backup goalies are whether their numbers have been impacted by their performance in relief appearances, and whether they are playing easier opponents.

I'm interested to know this info for Josh Harding, widely rumoured to be on the block. Harding posted a great Sv% of .929 this year and a truly crappy W-L record of 3-9-1. Given 13 decisions in 19 GP suggests he got a fair amount of mop-up duty.

There is a pretty extreme disconnect between Harding and Backstrom 2007-09:

Backstrom 70-37-16, .634
Harding 14-24-3, .378

... which is not easily explained by this:

Backstrom .921
Harding .916

... or this:

Backstrom 29.5 SA/60
Harding 31.7 SA/60

... or even the combination of those two:

Backstrom 2.32
Harding 2.68

I'm interested in your take, CG. Is it goal support? Failure to make the Big save? Small sample size for Harding? Hard to blame Lemaire for not giving him more time given Backstrom was earning an extra standings point for every 2 decisions.

The Contrarian Goaltender said...

Bruce: That's an interesting question. I looked into a bit and I think I have an answer: It's a combination of factors. I'll put it up in a post shortly.

Anonymous said...

Contrarian, what do you believe explains the enigma that is Andrew Raycrap? He had an outstanding first year and was execrable every year after that. His dropoff was even worse than Jose Theodore's or Kipper's.

Do you think Raycrap just had a very lucky extended unnatural "hot streak" his rookie year, has been playing with an undisclosed and unremedied injury for a long time, or what?

eyebleaf said...

Anon, Raycroft got lucky. That's it, that's all. This world's a crazy place.

And Contrarian, thanks for showing me the Raycrap I know and love. You, sir, are good people.