In brief, the draft lottery is essentially a lottery of the worst teams in the league to see who will win the first draft pick in the June draft. The odds of winning the first pick is weighted based on the team’s position in the standings as the end of the season, with the team that finished with the least points receiving the best odds to win the first pick. There is only one pick (first pick each round), while the rest of the draft order is based on league standing.
The problem here is late in the season, bad teams may deliberately try to lose games in order to increase their odds of getting the first draft pick. However, I have two ways of fixing this:
The first option is the easiest. The lottery would remain, but the odds of winning the first pick would be increased for the best team not to make the playoffs (the team that finished in 17th place overall.) Using the current odds, that means the 17th placed team in the league would have a 20% chance of winning the lottery, the 18th place team would have a 13.5% chance of getting the first draft pick, while the 30th placed team (i.e. the ‘worst’ team in the league) would get a 1% chance of getting the first draft pick.
This first option would have two obvious advantages: late in the season, there would be extra encouragement for teams who are in the hunt for a playoff berth to keep winning. And if they miss the playoffs, at least they have a better chance to get the first draft pick. The other advantage is that even lower placed teams would want to win to gain a better chance to receive the first draft pick. That means that when a top placed team places a lower ranked team, there is incentive for the lower placed team to try to upset the higher ranked team.
The only disadvantage of this proposal is that lower placed teams may take years to improve through the draft. If they cannot benefit from the better odds of winning the first draft pick, the team may not find as much success.
The other option is more complex, but my preferred option: The draft order (i.e. the order in which teams make their selections) would depend on the just completed season, but rather their overall points from the previous five years. The NHL would add up the final points standings for each of the past five seasons and the team that compiled the least points over five years would the first draft pick, and the order would continue in with the team with the second least point total over the last five years would receive the second draft pick, and so forth.
The main advantages of this option are that it takes out one season wonders. If a team has an unexpectedly good (or bad) year, then this would not have a large impact on the team’s selection in the draft. It also means that team that continuously perform badly are given better draft picks every year, despite one year where may have performed slightly better. Any new teams added to the league would automatically receive the better draft picks for the first five years, as their overall points total will be lower (until they have completed their fifth season, they would not have points for a number of season, which means their points total will naturally be lower.)
One final advantage of all this is that it would not be as easy for a team to deliberately tank at the end of a season in order to get a better draft pick. The previous four seasons would still have to be taken into account when determining the draft order. As such, the obvious advantage of tanking would not exist – and most teams cannot continue to continue losing every year and still attract sufficient attendance at home games to continue operations. Therefore, losing would simply not be an easy option.
The main disadvantages of the second option would be that teams in the last two seasons who have underperformed may not get better draft options. Also, teams that have started to become more successful may still benefit from their previous poorer performances.
Once the current 2014-2015 season is completed, I will provide an example of how my suggested ranking differs from the current system.