With the Leafs having fallen out of the playoff picture once again, fans are faced with an unfortunate dilemma. On the one hand, we want the team to lose so that they have a better draft pick and an increased chance at winning the lottery to draft Connor McDavid. On the other hand, it's not a lot of fun to watch your team lose, and cheering for them to lose is even worse.
A number of solutions to this problem have been suggested. In the past, my preferred option has been a weighted draft of non-playoff teams not just for the 1st overall pick, but for all of the first 14 selections. A fair number of people object to this approach, however, on the grounds that even though it eliminates the guarantee of a top pick for teams that lose a lot, there's still some incentive to tank.
One alternate proposal is to award the top picks on the basis of "wins after elimination from the playoff race" or something along those lines. This approach would see teams begin to accumulate "draft points" as soon as they're mathematically eliminated from the possibility of making the playoffs, so that fans of teams that are out of the playoff race have a reason to want to see their team win. I have problems with this approach, however. My primary objection is that it essentially doubles the pain of fans of particularly bad teams. First your team fails to make the playoffs, then they fail to win enough to secure a high draft pick. As a fan, this is worse, not better.
Another suggestion is that the NHL could have some kind of "loser's tournament" among teams that miss the playoffs to determine draft order. This proposal also has the problem that it doubles up on pain for fans of bad teams. It's also just fundamentally weird to reward teams for mediocrity, which is what such a tournament would do. Also, I just can't see NHL owners or GMs agreeing to a system that would see some teams stuck in a perpetual losing cycle, with no real ability to wind up with a marketable star.
One other suggestion I've heard is that the NHL could simply assign draft order randomly for one year, and then cycle each team through the draft order in subsequent years, such that the team that has pick #1 in the first year then gets pick #30 the following year, and so forth. This solves the problem of fans wanting their team to lose for a better pick, and it also solves the other concerns I've raised above.
There's a pretty significant problem with this approach though: because it would take 30 years to cycle through the league, no one would ever agree to it. GMs simply wouldn't agree to a system that might see 29 other teams guaranteed a 1st overall pick before they get it.
This problem is compounded by the fact that there's simply no way to know what the NHL will look like in 30 years. 30 years ago the league was only 2/3 the size it is today and 4 of the teams that did exist have since changed cities. Who's to say what the NHL will look like in 30 years? Or that they won't abandon this approach to the draft after 8 or 10 years? No GM wants to be the one that winds up having been screwed by a rotating draft order that is abandoned before they're able to take advantage of it.
But fear not! I think I've got a solution that addresses everyone's concerns.
My approach, which I'm calling Draft Blocks is a modified version of the rotating draft order, and it works like this:
The 30 teams are divided into 6 groups of 5. It could be done by division, alphabetically, randomly, or some other way. In a given year, each block is given a group of 5 picks. One group gets picks 1 through 5, another gets 26 through 30, and so on. Within each group, draft order is determined by random draw. Draft rounds 2 through 7 would keep the draft order from round 1. The following season, each "draft block" is given to another group of 5 teams. Ideally, the order in which draft blocks were rotated wouldn't be linear; rather than giving block 1-5 to the teams that drafted 6-10 the previous year, perhaps it would go to the group that drafted 16-20, and so on, so that teams were rotating into and out of top pick locations every 2-3 seasons.
This approach would guarantee that every team in the league got a top 5 pick exactly once every six years (unless they decided to trade it). The blocks would cycle quickly enough that the system doesn't need to be kept around for 30 years for everyone to have had a shot at the best picks. It would facilitate long-term planning and pick trading, because teams would know ahead of each season what the approximate value of their pick was, as well as the values of their picks in upcoming years.
I feel like this approach solves all of the problems that people raise with the draft (aside from the fact that it exists at all): there's no benefit to tanking, there's no double-punishment to fans of bad teams, and it doesn't subject some teams to the possibility that the system will change before they get their chance at top picks.
This general idea could still be used if the NHL expands to 32 teams by changing to 8 blocks of 4 (or vice versa, although I think 4 blocks of 8 puts too many picks in each block, which would reduce the parity of picks between teams). This approach runs into problems if the NHL wants to expand to 34 teams, as that number doesn't divide nicely, but hopefully the NHL doesn't expand to 34 teams in the near future.