Over the past few months, I’ve come to realize that one of the greatest advantages you can create over your peers in DFS contests is to roster players whose status for an upcoming game is uncertain (injury, scratch), but end up playing anyways.
A few weeks back, our own Tomjk321 won a large NHL GPP because he rostered the Winnipeg Jets second line on a day when winger Jack Roslovic was listed as Day to Day (he ended up with a hat-trick). I asked Tom about the thought process that led him to roster WPG2 when we didn’t have that line graded out particularly well. His response was that more often than not, an entire line will have reduced ownership when one player is injured, so in reality, you’re getting ownership leverage, not just on a player, but on an entire line combination.
Last night’s example of injury leverage won i_slewfoot_u $10,000 in the $5, 14.2k entry, 150-max contest on DraftKings. I_slewfoot_u rostered Nashville Predators defenceman PK Subban, who was a Game Time Decision, and ended in just 1.5% of lineups. I’m not sure if this was a late swap, or if i_slewfoot_u had Subban in his lineups and was ready to switch him out for another defecenman if he didn’t play, but really, it doesn’t matter.
Subban ended up with 1 G, 1 A, 4 SOG and 2 BS for a total of 8 DK points. Subban’s counterpart, the slightly higher rated Roman Josi, was 18% owned in the contest. Josi finished with no points, two shots and two blocked shots for a total of 2 DK points. I compare these two because at this stage of the contest, Josi was the most logical defencemen you could global player swap out for Subban, (though Tyson Barrie and Keith Yandle were also options).
So, knowing that Subban would be virtually unowned, and Josi was one of the chalkiest defecneman on the slate, should you have swapped out Josi for Subban? Or, the flip side, should you have rostered Subban at a higher than normal rate while building, and swapped him out if he didn’t play?
In both instances, I’d argue yes, you should have rostered Subban – as much as possible (Alex and Tomjk321 have some more nuanced opinions below).
In the 46 games that Subban and Josi have played together this year, Josi has markedly outscored Subban in 17 games (by 1.5 or more DK points), Subban has markedly outscored Josi in 11 games, and 18 have been under the 1.5 point threshold on either side. So, put another way, Josi wins the H2H battle 37% of the time, Subban wins it 23% of the time and it’s relatively close 40% of the time.
Josi wins at just below a 2-to-1 rate, yet was owned at a rate 12x higher than Subban. And this, of course, doesn’t factor in the roughly $1000 in cost savings you get by rostering Subban in the first place.
Any way you slice it, taking Subban at 1.5% ownership is the sharper move.
This is an example of an extreme version of a strategy that Alex preaches on his NFL shows and in his Sunday morning column, where he tallies up the value for each offensive stack relative to the stack’s ownership. The thinking being that if two stacks have similar projections, you take the less owned stack.
One question remains though: What did the pros do with the Subban situation last night? Well, a few rostered Subban at high rates, a few got overweight and a few ignored him entirely – opting to trust their projections. Here are the top-ten NHL players on RotoGrinders and how many times they rostered Subban last night:
RotoGrinders NHL ranks (150 lineups, unless otherwise specified):
1: Testosterown – 0 Subban shares
2) LowKey – not in contest
3)LouMister35 – 14 shares (41 lineups)
4) Awesemo – 5 shares
5) DollarBillW – 15 shares
6) Bric75 – 4 shares
7) TomJK321 – not in contest
8) GiantSquid: 0 shares
9) 1800Eddie – 14 shares (125 lineups)
10) Joda: not in contest.
Additional pros:
TommieNation: 0 shares
i_slewfoot_u: 9 shares
Mirage88: 9 shares
All in all, most of these pros weren’t fussed by the idea of stuffing Subban into their lineups through the late swap, or by simply rostering him and then swapping for a Tyson Barrie later on. It’s hard to tell if this was a conscious decision, or if it was more of a time constraint issue. Because, the thing is, the pros, are often playing multiple sports at once, and won’t prioritize adjusting their lineups – especially for relatively small-time stuff like a $5 NHL contests. Earlier this year, for example, Mirage88 finished second in a GPP instead of first, because he didn’t swap out Jacob Middleton for the Sharks, who was benched in pre-game rushes. It was 10 pm Eastern Time, so for all we know, he was in bed.
The point is that the sharpest people in the business aren’t always sharp when it comes to late swaps, and you can gain an advantage by being on top of things.
Advice from Tom and Alex
I asked our two site founders for their input on late changes, late swaps and injuries as they pertained to the Subban situation.
Alex said that he had put Subban in at a normal rate – relatively to his projections, and would have been ready to swap him out, had he been unable to play. I spoke to Bric75 in our Premium Slack as well, and he said he had the same strategy as Awesemo.
Tom agreed with my assessment, that more often than not, given these circumstances, it would be better to roster Subban. However, he added a caveat: you need to think of Subban as a leverage play – not in a vacuum – but as part of larger roster construction. If, for example, you had a lineup with the .2% owned Maple Leafs fourth line last night, you probably didn’t need to worry about getting more contrarian by swapping out Josi and adding Subban to your lineups.
My guess is, Tom’s take is how a lot of pros approach a decision like this.
There is a lot to consider here, from a macro standpoint. If, for example, you have the nuts lineup heading into the Nashville game, you’d probably want to keep Josi (assuming he is on said lineup), if a few of your lineups are under-performing, it probably makes sense to go with Subban.
Line Changes
A smaller, less extreme example of the Subban pivot is simply staying on top of what’s happening during NHL line rushes. Because line point totals are so closely correlated, being aware of a line change a half hour before game time can give you a huge edge on the field. We saw it last night, when Miles Wood was moved up to the Devils top line, after he was projected on the second line for most of the day. For NBA or MLB, it’s probably as simple as making sure you’re on top of starting lineups and injuries as the night progresses.
As always, thanks for reading!
