The NHL Draft And Goalies

With the exception of a handful of lucky franchises, just about every team in the NHL is looking for the same thing: a franchise goaltender. Teams are always in search of this player. Everyone would love to have the next Martin Brodeur or Patrick Roy, two of the greatest to play the game. Guys get billed as the next this or that all the time, but in a lot of cases, this is a mere mirage. Unfortunately for the teams that end up with these guys, they wasted three to four years building around the wrong player. Then, it's time to move into the next one, and that player may fail as well. It's a vicious cycle for a lot of teams.

In this weekend's 2008 NHL draft, two goalies were taken in the first round. Chet Pickard went 18th overall to Nashville, and Thomas McCollum went 30th overall to Detroit. According to the NHL's scouting service, these were the two best North American goalies available, and these teams have high hopes for both of them. Dan Ellis is a very good player and they have a couple other decent prospects, but the Predators hope they can get the fabled franchise goalie. They hope Pickard can blow Ellis out of the water and be the cornerstone of multiple Stanley Cup winners. They want to go into games knowing they don't need to bring their A offensive game every day because their goalie can carry them. That's what every team thinks when they select a goalie.

Detroit is taking an approach that kind of reminds me of an adage about baseball prospects. The rule with three good prospects is that one will reach his potential, one will become a decent player, and one will never make it. This isn't necessarily true in every case, and maybe it doesn't apply at all to hockey. However, Detroit now has three goaltending prospects that were taken within the first 92 picks. Jimmy Howard was taken 64th overall in 2003, and he's now the best prospect in Detroit's system, according to Hockey's Future. He's gotten a little taste of NHL action, and he will soon be joined by Daniel Larsson. Larsson was the 92nd pick in 2006, and he signed an entry level contract to come over to the states. He's their 11th best prospect, and now he'll adjust to the North American game in the AHL. I don't know when Chris Osgood is going to retire, but the Red Wings will probably be in good hands when he leaves.

Fans are familiar with this too. Names like Boucher, Oullet and Pelletier are enough to make any Flyers fan groan. They've seen it all too often. These players get billed as the goalie that will finally bring the cup back to the city, and they never meet expectations. Despite this, fans are clamoring for the organization to find that next great goalie year after year. Maybe the Flyers haven't had that player since Hextall, and it's been a long time since he played.

When a call for a team to draft a goalie with a high pick is made, it's always inevitably met with the same response: "It's too much of a crapshoot to spend a high pick on a goalie." This happens without fail. Is it true? In the eyes of fans who have seen the names Waite, Finley and Krahn come and go, teams should never spend a first round pick on a goalie because you never know what you're going to get. Coming in with an open mind, I set out to prove this.

I used data from the years 1983 to 2006 for this study. In other words, every goalie between Hasek and Bernier. These two are the oldest and youngest goalies to be active since the lockout, and even though it was a lot on my plate, it was the right way to approach things. In a Microsoft Excel spreadsheet, I listed all of their games played, wins, GAA, SV% and Stanley Cups won. I also came up with a couple formulas using these numbers. One judges success over along period of time, and one is a more normalized formula that can compare goalies who have played two games to 500.

First, we'll throw out the formulas and the stats and look at one simple thing: Who made the NHL? Who was good enough to put the pads on for even one appearance? The first round led the way with 42 out of 53 total picks making it, or 79.2%. Here's how the rest of the rounds shaped out. For comparison purposes, I made all of the drafts the "same". By this, I mean the first round is picks 1-30, the second round is 31-60, and so on. Expansion obviously shifted some picks from one round to another, such as the 29th overall pick being at the end of the first round instead of the beginning of the second like twenty years ago.

Second: 55.2%
Third: 48.5%
Fourth: 36.7%
Fifth: 23.7%
Sixth: 30.1%
Seventh: 30.3%
Undrafted: 25%

One observation: How sad is the 5th round? That's the lowest percentage out of anyone, with three rounds of players left behind them. I wonder why that is. Why is the first round leaps and bounds better than everyone else? There's probably a couple reasons. One, those players probably are better than other goalies. For a lot of them, their talent carries them from juniors or foreign leagues to the NHL. It's pretty obvious that a talented player is more likely to make it than a player with less talent. Second, teams will feel obligated to even give busts a chance. Even if it becomes clear that a particular goalie isn't the franchise player they hoped, their team might bring them up anyway to try and get something out of their investment. They spent a good draft pick on the player, and the fans were expecting him to be good, so they'll try to show him off before he fades into the sunset.

I think later round picks have been making it more and more often over the years. One reason for this might be the NHL having better relations with leagues around the world. Agreements with the IIHF have been put in place to get NHL draft picks over and out of their contracts overseas. In early years of this study, players such as Vladislav Tretiak were unwilling or unable to join an NHL club even if they were drafted. These days, it's easier for foreign players to come to the states to play.

Before moving on, kudos to Jordan Sigalet and Rob McVicar. They both made one appearance in the NHL, and neither of them had to make a save. Their Moonlight Graham stories must be tough on them, and I'm sure they would've liked one puck coming at them, one win, or even the red light going on behind them. Just once.

Is that really the best measure though? Teams want to win a championship. They're not satisfied with a player taking the ice once. Sorry Rob McVicar, but teams want to draft players that stick around longer than you. Let's look at the number of Cups won by goalies drafted in this period. There were 17 total, not counting Ed Belfour's win as an undrafted player. The first round checks in first again with 8 of the 17 Cups.

Second: 7
Third: 0
Fourth: 0
Fifth: 0
Sixth: 0
Seventh: 2
Undrafted: 0 (not counting Belfour, again).

Wow. This is definitely skewed towards the first 60 picks of the draft in a major way. It's pretty unbelievable that goalies taken between the 61st and 180th pick did not reach hockey's peak in a 23 year period. These results are skewed in another way, and I'm talking about players such as Martin Brodeur and Patrick Roy. They each won multiple Cups, and they run up the tally for the first and second rounds respectively. Who are the two renegades? Dominik Hasek and Nikolai Khabibulin. Hear that, NHL? If you're taking a goalie late, look for an eastern European.

Now is that the best measure? I'm going to take a look at one last thing for now, and that's goalies "good enough" to win the Stanley Cup. I'm talking about the Dan Marinos of the NHL, and according to my spreadsheet, more specifically, Marty Turco and Olaf Kolzig. These are the goalies that haven't necessarily hoisted the greatest trophy in sports, but goalies that probably should have or soon will. I used the two formulas I mentioned previously to figure this out. To qualify as good enough, goalies had to have a career success number greater than Cam Ward and a greater normalized number than Bill Ranford. Why Ward and Ranford? They each had the lowest number in those categories among all goalies that have won the Cup.

Before I post the numbers, I'm going to admit this isn't a perfect measuring stick. It's a statistical formula that quite frankly is pretty arbitrary. Goalies that played with strong defensive systems in front of them, such as Chris Osgood and Robert Esche a few years ago, benefit. Goalies on great teams benefit. Goalies that played recently benefit because the style of the game is different. There are some goalies that qualified for this that many believe aren't good players. This isn't about that though. It's about finding some way to quantify who should be able to win. The first number is the percentage of total goalies drafted that are good enough, and the second number is the percentage of goalies that made the NHL that are good enough.

First: 37.7%, 47.6%
Second: 17.9%, 32.4%
Third: 12.1%, 25%
Fourth: 10.1%, 27.6%
Fifth: 7.9%, 33.3%
Sixth: 5.5%, 18.2%
Seventh: 6.6%, 21.7%
Undrafted: 5.8%, 23.3%

As you can see, the first round again sets the pace. This should be obvious though. Once again, these are the more talented players, and more talented goalies tend to do better. However, when looking at these measures, you have to consider something. Even people who think goalies are too risky to draft high will acknowledge that they're probably more successful earlier in the draft. Teams have to weigh this risk. Of course first round picks are supposed to be better, but would it make more sense to take a skater in the first and take a chance on a player in maybe the second round? By the way, big front office guys. If you take a goalie in the 5th and get him to the NHL, expect good things.

That's a preliminary look at what I've found. I'm going on a bit of a trip this weekend, so no more updates until maybe Monday night. No Phillies preview tomorrow, but they're off Monday, so I'll only miss one game. I might make two more posts about this, since I want to reach a definite conclusion, and I'll post some more in-depth stats for each draft over the years for you to take a look.