As a fan of the Canes, I'm delighted that Khudobin is playing well. As a guy interested in Hockey Economics and Hockey Analytics, particularly goaltender performance, he's like the Holy Grail. Figuring out what he is worth requires figuring out (1) How good is he? (2) How many shots will he see? (3) How many goals against he would give up relative to an average goalie? and (4) How much is a goal worth? It's interesting because there really is no reasonable comparison goaltender. The closest is probably Viktor Fasth, who signed a 2 year $5.8m ($2.9 a year AAV) deal after being 0.926 on 565 ES shots last season.
How good is Khudobin?
Khudobin came into this season having made 413 saves on 444 Even Strength shots, a 0.930 ESS%. Add in another 393 saves on 420 ES shots and he's now at 0.933 ESS%.
There's two basic ways to estimate his talent: Frequentist and Bayesian. A Frequentist would say with 806 saves on 864 shots, out best estimate of his ability is the observed value 0.933. However, his true ability has a 95% probability of being between 0.914 and 0.949. So he could be a 0.914 goalie who has been getting good bounces or a 0.949 goalie getting bad bounces. Or somewhere in between.
Gabe Desjardins, the Father of Analytics, had an article (Fooled by Randomness, it's one of the classics) that looked at how many shots a goalie needs to see to have a decent handle on his ability. He came up with a minimum of 2000. So even if Khudobin plays 90% of the rest of the season, he's still going to be below Gabe's threshold.
A Bayesian would look at his 0.933 ESS% so far but would also incorporate the knowledge that most goalies are average (or around 0.920 ES). As a result, a Bayesian would say it is more likely that Khudobin is somewhere in between the observed 0.933 and the average 0.920. The Bayesian estimate for his ability is around 0.926.
How many shots will he see?
More shots means higher potential value. The Canes give up a lot of shots. Last year they gave up 1250 ES shots in 48 games, which works out to 2135 shots in a normal 82 game season. This year they have yielded 1381 in 54 games, which projects to 2100 shots.
So if Khudobin is the starter and he stays healthy, he very reasonably will see 1500 ES shots in a season.
Note that I haven't worried about PK shots. Goalies don't differ on the PK link, so there is no expected difference between Khudobin and an average goalie on the 300 or so PK shots he would face.
How many goals against he would give up relative to an average goalie?
An average (0.920) goalie facing 1500 shots would be expected to give up 120 goals. If Khudobin is really 0.933, he would be expected to give up 101. If he is really 0.926 he should give up 111.
How much is a goal worth?
About $120k
So how much is Khudobin worth?
Average salary $2.74m If he is 0.926, he is +9 EGVA. At $120k per goal, that's another $1.1m. Total $3.8m.
If he is 0.933, +19 EGVA Another $2.3m. Total $5.0m
Will he get it?
I don't know! On the one hand, he signed at a huge discount this year due to his lack of a track record. He still won't have enough shots faced to really have a great handle on his true talent. He will probably sign at a discount again, but not as big a discount. A Fasth type deal would make a lot of sense.
On the other hand, GM's have been throwing money at goalies like crazy. If Steve Mason is worth $4.1m a year and Lundqvist $8.5m a year, Khodobin's talent is closer to Lundqvist than to Mason.
Connect with Canes Country