Can The Capitals Win The Southeast Division?

Alex Ovechkin celebrates

Photo credit: Don Heupel

The Washington Capitals are in second place in the Southeast Division, only one point behind the leading Tampa Bay Lightning with 21 games remaining this season. Can the Caps catch them?

As usual, we will ask the Monte Carlo Machine to simulate the rest of the season. To determine the chance of winning any individual game we will use the Log5 method. Log5 is a method of estimating the winning percentage when two teams play each other. The formula is:

W%(A v. B) = W%(A)*(1 – W%(B))/(W%(A)*(1 – W%(B)) + (1 – W%(A))*W%(B))

Since we know what the win percentage is for both Washington and Tampa Bay, and their remaining schedule, we can pop these results in, run the Machine 10,000 times and see how often the Caps come out ahead of the Lightning in the division.

SEWins

The chart above shows how often each team is projected to finish the season with a certain number of points. Thirty-two percent of the time we expect the Capitals to end the campaign with 103 points while the Lightning would finish with the same number of points 17% of the time. But since Tampa Bay has more upside, we would expect them to win the division more often than not. According to the Monte Carlo Machine, the Caps have a 43.94% chance at winning the Southeast Division.

In contrast, I ran the same simulation for the New Jersey Devils — everyone’s favorite underdog. If we go on Mirtle’s premise of needing 90 points in the Eastern Conference to make the playoffs, the Devils would have to win 17 out of their next 22 contests. In other words, they would make the playoffs in 6 out of every 10,000 seasons. Now that is a long shot.

Editing by Chris Gordon and Ian Oland.

  • Tim

    I like the Caps chances of winning the SE, especially with how Tampa has been playing recently. Even if they don’t win the SE, they should take either the 4 or 5 seed. I’d be happy with either result – especially if we get a weakened Pens team in the 1st round.

    Just curious – why is there the enormous spike on 103 points for the Caps?

  • Chris

    Just curious, but isn’t the log5 for “when two teams play each other?” We don’t play each other the entire rest of the season… we play lots of other teams too – would a “better” approach be to do this vs each opponent/each game and see the points results? Not saying we should do that, but just curious about the math.

  • http://www.russianmachineneverbreaks.com Russian Machine Never Breaks

    @Chris: I probably should have been more clear. I ran 10,000 seasons for the Caps, against *their* remaining opponents and 10,000 for TBL, against *their* remaining opponents. Not WSH/TBL head to head.

    What you are seeing is the total points based on the team’s respective remaining schedules.

    @Tim That really isn’t games where they end up with 103 points, but games where they end up with 101.01 to 103.

    -Neil

  • Paul R

    Sorry, Neil. It is a ridiculous question. One point behind with 21games to play. Can they win? They could win with one game to play. I like stats and data, but not only could we win, we could win by a lot. We could lose, perhaps by a lot, maybe by not much. But could we win–of course. The answer is there is a 100percent chance we could.

    Love your stuff. Just a clumsy question.

  • http://www.russianmachineneverbreaks.com Russian Machine Never Breaks

    @Paul I wouldn’t say it is “ridiculous.” I am in the minority thinking they have a chance. I originally told Sky Kerstein a week ago they had a 20% chance and got laughed at on Twitter — which hurts more than you think.

    You’re right tho, there is a 100% chance they win the division or don’t.

    High five for you at RMNB 2: Electric Boogaloo.

    -N

  • Paul R

    Sorry about ‘ridiculous’. You do good work and deserve more respect. While we are talking Caps, two things–do not trade Semin, and I think there is a 100 percent chance we will win the SE. Just me…

    Paul R

  • Bradford

    Are you taking win percentage or points percentage into account here? If you’re going straight off of two points for a win, no points for a loss, the analysis gets confounded somewhat – especially with Tampa taking so many games into OT lately.

  • Tim

    Neil – thanks for the explanation. That actually makes a bit more sense. So it pretty much means the Caps have a 32+% chance of getting 26-28 more points then?

    Paul – as of right now (which is 2 days after your post, to be fair), the Caps are 3 points behind, and have played 1 more game than Tampa. So Tampa could easily be 5 points ahead, if the games played were even.

  • Neil, RMNB

    @Bradford I am using win percentage to figure out wins, then estimating OT losses using GF and GA for total points.

    @tim that sounds about right.

  • PainKiller

    Neil, thanks for giving us additional ways to look at situations. I wonder, how long have you been providing statistical analysis? Have you kept track of how many times your projections and predictions have been correct (or, close to correct.)

    Again, thanks!