Is the draft lottery working?

  1. TheOtherOne
    Motivation: As a Red Wings fan, I've experienced the entitlement of a team capable of staying competitive for over a decade, followed by the unfamiliar frustration of a team that's not used to being bad. My fanbase is split between those [rightfully] wanting to win games because of pro sports' competitive spirit and pride, and those [rightfully] wanting to tank for the chance at a better future. Notice I called both sides right: I don't blame the fans for their divide, but the system creates it nonetheless.

    Purpose of the lottery: I think the draft and lottery have a couple purposes and I want to analyze whether it's accomplishing them.
    1: Distribute top talent fairly to all teams
    2: Create parity by making the worst teams better
    3: Discourage tanking

    Odds: What I've been hearing a lot lately is "you NEED a top 3 drafted player to compete for a Cup." I'm going to accept that premise without questioning it to simplify things. But it's also important to remember that a top 3 drafted player is obviously not sufficient by itself- you need quality depth and support.

    Lottery odds are, for the last place team for example, 18.5%, 16.5%, 14.4% for 1st, 2nd, 3rd pick. Added up, that's 49.4% chance at a lottery pick.

    The key is that since it's a probability, it doesn't really make sense to look at a single year. It's 50/50, you might win it or you might not. But probabilities work for averaging over time. So the important number is 1/49.4%, which means the last place team will win a top 3 lottery pick on average once in 2.02 years.

    Do this math for all positions, and we see it will take, on average,
    years for a team in a given position to get a lottery pick.
    ** This is the main point of the thread. I think this is a much more useful number to look at than anything else, because it breaks the lottery down and simplifies, with the "prize" being a lottery pick, what it takes to win in terms of time spent as a losing team.

    Additional analysis with some subjectivity...
    I'm going to split non-playoff teams into "bad" 1-7 and "bubble" 8-15.

    - The lottery has been this way since 2016, and we've seen wins by #13, #8, and #12. So out of 9 potential lottery picks, 3 were won by bubble teams. We can see it in the odds and we've seen it happen in reality. When we talk about teams refusing to tank and staying a bubble team forever- such a team will get a lottery pick, on average, once in 14 years, while staying in the better half of non-playoff teams.

    - For bad teams, tanking, or losing a game on purpose, has the potential of changing the expected value by 1. In a given year, it's a crapshoot (out of those 9 potential lottery picks, only 2 were won by bottom 3 teams). But as far as the year-to-year average goes, it doesn't matter much if you're dead last or 4th to last. If you're a bad team for 5 years you should get a lottery pick regardless of your specific standing.

    - The way I see it you can rebuild one of two ways. 1: You can have a bad team with no depth, finish in the bottom half of non-playoff standings, and take on average 3 years to win a lottery pick. Then take several(?) more years to build the surrounding depth necessary to compete for a cup. 2: You can have a bubble team with decent depth but lacking high end talent, finish in the top half of non-playoff standings, and take on average 14 years to win the lottery. From this it seems like method 1 is a bit faster, but there is a lot of fuzziness because no matter which way you choose, everything has to go right. I'd like to encourage method 2 (i.e. discourage "tanking" even more), but maybe the lottery odds need to be adjusted some more for that to happen.

    Ok I already spent too much time on this so go ahead hfboards tear it apart if you want.

    Share This Article

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice
monitoring_string = "358c248ada348a047a4b9bb27a146148"