It’s DFS in its simplest form. But what is the best way to play DraftKings NHL pick ’em game? Ryan Clifford takes you through some lineup trends to help assist your Tiers building process.
DFS couldn’t be made any easier. Pick the player that will score the most points out of these groups. No salaries, no goalies, no position limits – this is DK Tiers.
Although the rules are quite simple, many of us have trouble actually figuring out how to win playing the tiers contests on DraftKings. Well, I took a deep dive into the format. I broke down the optimal lineup for all 107 tiers slates in 2019-20 to help identify trends and patterns to help us in our build process.
How Important Is Stacking?
The importance of stacking in GPP’s might have been diminished a bit by DraftKings new scoring system this season, but how important is it in the tiers contests?
Well, consider this number. In 107 slates in 2019-20, the optimal lineup – that is, the perfect lineup – contained a full 3-man forward stack just four times. That’s a 3.7% rate. No, full stacking should not be considered a viable strategy here. In the optimizer, you can set a rule to use no more than 2 forwards from the same line.
But, can we stack 2 players together? The short answer: yes. Mini-stacks have appeared in 51 of 107 (47.7%) optimal lineups this season. I don’t consider it to be a “rule” I follow, but I don’t mind my lineups containing a mini-stack.
Players Per Team
Just because we shouldn’t be stacking full lines together, that doesn’t mean we need to force 8 players from 8 different teams into lineups. In fact, the optimal lineup contained at least two players from one team 86% of the time (92 out of 107 lineups). Therefore, I’d say it’s actually quite the contrary – I personally add a rule that forces at least two players from one team.
Is it wise to pile up 3 or 4 players from the same team, though? Have a look:
No, I don’t think you want to force any more than two players from the same team, and I would set team maxes to three players. As you can see above, one team had four players in just 3 out of 107 (2.8%) optimal lineups.
Defenders definitely don’t have the same DFS upside as the forwards, but that doesn’t mean you shouldn’t be rostering them in this format.
One or two defensemen is definitely the sweet spot – 61.7% of optimal lineups – but I don’t think you should really write anything off at three or less. Four defenders appeared in just 3.7% of optimal lineups – and there was one lineup with five – so I think anything over three is probably overdoing it.
Trust Your Projections
Whether you’re using the projections from our optimizer on EliteFantasy.com or you make your own numbers, you must trust your projections for this format. Ownership should be a major factor in GPP’s – just like any other sport – but you are simply trying to find the highest scoring players from each group…use your projections. Here’s why:
Besides the second tier, you can see the highest projected player appears far more often in the optimal lineups than the lowest. But it’s not just the highest versus lowest that tells us to trust the projections…here is a similar chart using the three highest projected players in each tier.
Again, for the most part, the highest projected players will appear more than the lower projected players.
…But Don’t Rely On Projections Too Much
Just because the higher projected players are appearing more often, that doesn’t mean everyone should immediately start tossing in lineups with all of the highest projected players – that’s not going to work either. In fact, only 3 times out of 107 optimal lineups were there five players that topped their respective tiers. Only 24 times were there five players that were one of the top-two projected players in their tiers.
Remember, I generally will set rules for anything that is occurring in 85-90% of optimal lineups. Using the chart above, that means I am setting rules to include at least one player that has the highest projection in his tier (87% of optimals) – but will have no more than three of those players, something that occurs in just 7.5% of optimal lineups.
Let’s have a look at the inverse of the chart above. This shows how many times the lowest projected players appear in optimal lineups:
As you can see, there is a much sharper decline in the lower projected players, but there is still generally one or two of the low-end guys in optimal lineups.
So, using the 85% threshold, I am setting a rule that says “use at least one player that is projected as one of the two lowest in his tier (88.8% of optimals), but no more than three of these players”. Also adding a secondary rule that says not to use more than two of the absolute lowest projected players – something that is only occurring in 3.7% of perfect lineups.
Putting It All Together
Just like the Showdown format, I highly recommend MME’ing these Tiers contests. You want to be building at least 20 lineups – this is going to give you a much better shot of landing on the right combination of players while using the rules below.
So, using all of the data above, here is a list of 12 rules I am currently using to build my Tiers lineups – the 12 Commandments of NHL Tiers, if you will – as well as a short video showing you how to set up the rules and groups on our optimizer.
Rule #1: Use no more than two forwards from the same line. (96.3% of optimals)
Rule #2: Use at least two players from at least one team. (86.0% of optimals)
Rule #3: Use no more than 3 players from any one team. (96.3% of optimals)
Rule #4: Use no more than one team with 3 players. (99.1% of optimals)
Rule #5: Use no more than 3 defensemen. (95.3% of optimals)
Rule #6: Use at least 1 player that has the highest projection in his respective tier. (87% of optimals)
Rule #7: Use no more than 3 players with the highest projections in their tiers. (92.5% of optimals)
Rule #8: Use at least 2 players that are projected in the top-two of their respective tiers. (87.9% of optimals)
Rule #9: Use no more than 5 players that are projected in the top-two of their respective tiers. (92.5% of optimals)
Rule #10: Use at least 1 player that is one of the two lowest-projected players in his tier. (88.8% of optimals)
Rule #11: Use no more than 3 players that are one of the two lowest-projected players in their tiers. (82.2% of optimals)
Rule #12: Use no more than 2 players that have the lowest projections in their respective tiers. (96.3% of optimals)
Here is a short video in which I show you how to set up all of the above rules:
NHL 20 Tiers Contest Tonight!
Todd Cordell and myself will be putting on another slate of NHL 20 sims and holding a DK-style Tiers GPP. Here’s all the info on how to get in on this!
*Lines and rosters at www.fanlinksports.com
*Games begin streaming at 9:35 EDT at www.twitch.com/TCordell65