The Ottawa Senators have two critical decisions to make heading into the 2018 NHL Entry Draft.

At the top of the list is the future of defenceman Erik Karlsson. With one year left on his contract and a growing concern about losing the two-time Norris Trophy winner to free agency for nothing next season, Ottawa is once again listening to trade offers from around the league.

The team has said publicly that they will make an attempt to re-sign Karlsson on July 1, but it’s still unclear if the team will let the situation get to that point – especially if they can claw back valuable 2018 draft picks to use in Dallas.

The second most interesting decision concerns what Ottawa will do with their fourth-overall pick. The Senators, by way of the Matt Duchene trade with Colorado, have to give up either a 2018 or 2019 first-round selection to the Avalanche. The team has been adamant that they will use this year’s pick and relieve the conditions of the Duchene trade next year.

It’s an interesting dilemma for the organization. A fourth-overall pick has considerable valuable. But there’s skepticism that Ottawa will be competitive next year. They’ve already moved on from winger Mike Hoffman, are considering moving on from the aforementioned Karlsson and don’t have a clear fix in net. That’s a recipe for another lottery ball and a shot at a high pick next season.

Compounding the issue is the fact that the Sens may need to make the draft pick decision before they know if Karlsson is going to stay. It’s difficult to create a forecast when your best player is on the fence about returning and there’s no one in the system to replace him.

Ask 10 people to forecast what Ottawa’s 2018-19 season looks like and you will get 10 different answers – and that’s true if they’re throwing darts or running complex statistical models. We do know a couple of things. One: A lottery team tends to remain a lottery team in the subsequent year. The data is a bit choppy, but it states what should be obvious – it takes a bit more than one off-season to turn a good team into a bad one, and vice versa.

Embedded Image

Two: Results tend to regress towards league average. For example, teams finishing dead last in one season finished, on average, sixth worst in the following season. That type of regression manifests at every level for a variety of reasons. If you only knew how a team finished in a given year, you would need to regress their results about 80 per cent towards league average for a guess at where they would finish in the following year. This is indicative of what the NHL has become. It’s hard to remain exceptionally bad for long and it’s also hard to remain exceptionally good for long.

Maybe you look at this and conclude that Ottawa will improve next season. Or maybe you think the rough off-season – and what looks like a clear missive to get to the cap floor – will cloud their 2018-19 forecast.

Regardless of where you stand, the most important consideration is the changing of the draft lottery rules. The NHL, concerned about tanking, flattened the lottery odds. They effectively punished the incentive for teams trying to race to dead-last by worsening the odds, then took the odds they garnished and layered them back onto teams in the middle of the pack.

Ottawa learned this better than anyone in April. The Senators finished with the league’s second-worst record last year, but ended up picking fourth because of the slide rules. The league has gamed the math to make this more common going forward. So when you evaluate the merit of giving up this year’s fourth-overall pick, you have to wonder about the probability of ending up with a worse pick next season.

Consider the table below. It shows the draft odds for all 15 lottery teams going forward. The far right columns are the notable ones. The “Match+” column indicates, by where they finish, the probability of ending up with a pick that matches or exceeds fourth overall in value.

Embedded Image

This is where the math gets cruel. If Ottawa finishes as the third-worst team in the league next year, they actually have a better than coin-flip chance of selecting fifth or worse in the draft. Those odds get progressively worse as you work down the table. Another way of saying this: If Ottawa were to give Colorado the 2018 pick in order to preserve the 2019 pick they only have one option – an 82-game tank with the goal of finishing dead last.

There are so many factors that go into this, but I can’t imagine a scenario where Ottawa has an interest in outright tanking a season. The math simply makes it more desirable to hold on to the fourth-overall pick and hope things improve next year. (Such a strategy would also incentivize the front office to build a more competitive team, which may or may not include additional efforts to retain Karlsson.)

Either way, it’s an interesting time in the nation’s capital and certainly not an enviable position for general manager Pierre Dorion. We’ll soon find out what his decision is.