NHL DFS Strategy: How I Qualified for the Fantasy Hockey World Championship
It hasn’t been my best NHL DFS season. Transparently, in terms of “classic” GPP play it has been a losing one. Part of it is has been my own personal choices — Monday’s and Thursday’s I focus on Showdown, and Saturday’s are typically family days for me, while Sunday is a lot of work for DR and main slate NFL. In previous seasons, I was playing NHL six nights a week. That has left the majority of my action on either the large Tuesday slates, short slate Wednesday, or the occasional Friday if it is a light news slate.
Part of it has been the change in the DraftKings scoring system. We talked in some depth about how player correlations changed and therefore impacted what “optimal” GPP strategy might look like and how players are correlated. It took me a while to come up with an approach, but I think I am close. This has ultimately lead to me deploying more 2-man stacks and PP stacks than ever before, while still having a healthy amount of 3-man stacks in my mix.
While the “classic” GPP play hasn’t clicked on the right night yet, I have been lucky enough to win two seats to the Fantasy Hockey World Championship in Nashville. Having been to the 2017 Toronto FHWC as a contestant, and to the 2018 Washington DC FHWC as a guest, I know it will be a great event and opportunity to connect with other DFSers and sweat a six-figure first-place prize. One of the biggest benefits of qualifying for this year’s event is that the trip is fully compensated by DraftKings, meaning that 100% of the ticket value goes directly into the prize pool. On top of the trip/fun equity component, this event now has a legitimate chance to realize a return on investment as it functions as a high stakes NHL GPP with zero rake (besides the rake paid on any qualifiers). Below is a look at how I qualified for the fantasy hockey world championship, twice.
FHWC Seat 1 – November 14th
GPP Strategy: TB3 – EDM PP Stack with a correlated goalie
When I studied the player correlation data one of the things that popped out was that there was a lot stronger correlation between players on the same PP unit but different even-strength lines than I had realized. This was especially true for some players who played with “bad” even-strength linemates like Ryan Nugent-Hopkins. Tampa Bay was playing the New York Rangers on this slate, and Brayden Point and Nikita Kucherov were both 25-30 percent owned while Victor Hedman was 20 percent owned. Similarly, Florida 1 (vs Winnipeg) was 25-30 percent owned. This set up a position where I could pivot in two ways — off of Tampa Bay 1 down to Tampa Bay 3, and off of TB1/FLA1 onto EDM as a PP stack. With a contrarian build, I was still able to get Tampa exposure (they won 9-3), while also getting exposure to the best player in the world (McJesus) in a contrarian fashion by stacking him with Leon Draisaitl and instead of somebody dusty like Zack Kassian I was able to pivot to Ryan Nugent Hopkins. Despite Hedman and Kucherov putting 30+ points up each, I was able to win with 60 from McJesus.
FHWC Seat 2 – December 12th
GPP Strategy: MIN 2-Man, Johnny Gaudreau + Gio mini stack, and SJ1 + Correlated D
Looking at this lineup and the first thing that probably comes to mind is “Whew, what a flaming pile of sh*t” and the second thing is “that is not a SJ1 + Correlated D stack.” This is one of my favorite DFS lineups I have made, however, since it was a perfect example of how in DFS you are playing a game to beat other people, and not to score the most points.
If you ran our optimizer last night, San Jose 1 was legitimately everywhere. They were in a great spot (at home vs NYR who we already stacked against above) and featured full correlation, high ceiling metrics, and very cheap prices. They could also be correlated with high upside dmen in either Erik Karlsson (high upside, or a bagel it turns out) or Brent Burns. At a full-stack cost just above $17000 it was pretty obvious that they were going to be chalk, but I thought it was still the perfect play.
This was like the advantage of playing with position in poker. If I hit the nuts early, I could still with the line that easily was the best line on the night in my opinion. If I was behind, I could throw a hail mary onto a high ceiling line, versus an awful goalie (Martin Jones), that would be virtually unowned. I now had a position where I could either play the best line on the slate, or I could play contrarian if I had to. At around 10:25PM it was pretty clear that I was drawing dead — many contestants ahead of me were going to be on San Jose — and I had to swap. I went to bed (after tilting NFL Showdown) and woke up to a 2nd live final seat.
Playing chalk in an extremely volatile sport like NHL is generally suboptimal for a Q, but if you are able to set up a swap you are playing from a strong position, and this strategy can be applied to any sport with games that lock at different intervals. The rest of my lineup was “OK” — a cheap Gaudreau playing “3rd line” but still getting 18+ minutes a game and Top PP time with Mark Giordano went virtually unowned because everybody wants to run 3-man stacks and his linemates were terrible. Because of these contrarian pieces, I was able to survive zeroes from both Zach Parise and Erik Karlsson.
There was a lot of luck in both of these wins (I was so far behind I went to bed and dad sweated them while catching some zzz’s) but they are also representative of important DFS concepts. We aren’t playing to score the most points, we are playing to beat our opponents. We can still get exposure to GOOD plays in contrarian ways. And late swap can be used tactically to outplay opponents who “set it and forget it.”