LineStar NFL Takeaway - Week 9: HOW are people stacking?

📝 @zeroinDenver 🎯

Welcome back, everyone. Again, it's your dude, ZeroInDenver, here to try and bring some insight into NFL DFS contests. If you're new, this is where I do some analysis on NFL DFS contests and do analysis on what might work and what might not in terms of strategy and lineup construction.

This week is I want to look at HOW people stacked in lineups. We know that stacking is a thing, but what positions are people stacking? What do these lineups look like? Is this something that we should be doing?

Let's dive in.

Strategy

The Contest

For data, I'm looking at the NFL $3.5M Fantasy Football Millionaire [$1M to 1st + ToC Entry] contest on DraftKings. This contest has 207K entries, but we're only going to focus on the top 5000 places (5041 lineups to be exact - roughly the top 2% of this contest). The winning lineup had a score of 205.34 and the score needed to min-cash was 127.78.

Let's talk about stacks in this lineup. Right away, I look at Herbert and Allen. QB+WR is a popular stack combination. Look closer and you'll see we have DST + RB, another popular combination. However, this person also stacked WR and TE with no QB correlation.

For me, I'll say that I aim for all of my lineups to have a QB + WR stack, but I don't focus on other stack types. Let's take a deeper look and see how popular other stack types were.

Overwhelmingly, 2-man stacks were the most used, almost 3 times as much as 2-2 stacks and 3-man stacks. This table shows how often each stack type was used, what the top score was for each type, and what the average score was.

It shouldn't be a surprise, but stacking the QB with another player from the same team is a popular strategy. How popular? 90% of all stacked lineups did this:

This falls in line with how I've been building lineups. I aim to have no more than 3 players from a team. Consider this - Baltimore was a popular stack. If Lamar Jackson goes off and Mark Andrews goes off, what's the likelihood that Marquise Brown goes off too? Let's imagine that all 3 do well. At that point, I think it's really unlikely that a 4th player from the Ravens does well. You can also look at it this way - Let's say you add Rashod Bateman to your Jackson, Brown, and Andrews stack. Do you think Bateman outscores anyone that you could put instead of him? Or, let's say Bateman does really well. What's the opportunity cost to Andrews? Looking at how Baltimore players scored, the answer is clear. While Bateman did OK for his price, Andrews was terrible in comparison.

Other Stack Types

Of the most popular stack types outside of QB-RB/TE/WR stacks are DST-RB stacks. Here are the numbers on how often those were used in the top 5041 lineups.

But do we need to stack QB-RB/TE/WR?

Absolutely not. Here's the stack types and scores for lineups without a QB stack:

Here comes the fun part - trying to figure out why someone built their lineup this way.

The easy, and usually most reasonable answer, is that a lineup optimizer spit this out. Not only was Lamar Jackson not stacked with anyone else from Baltimore, but the opposing TE was rostered. Was that intentional? My guess is no. Conklin was played just because he was a really cheap option.

Let's talk about the Arizona stack of Conner and Kirk. Was this intentional? I can't figure out why Conner was played. There weren't a lot of great options at that price. Maybe you can make the argument for Devanta Freeman, but I'm not sure about that. On the other hand, Kirk makes some sense to me. Sure, Murray was out, but so were Hopkins and AJ Green.

Again, I'm thinking someone built this using an optimizer instead of hand-selecting players, but if I'm missing something, let me know!

Let's look at one more:

This one is fun. 2-2 stack with 2 players from Cleveland and 2 players from Las Vegas. At the same time, this lineup has three players from the CIN-CLE game, but Matt Ryan in QB.

Burrow is too expensive for this lineup, but Baker Mayfield was $200 less than Matt Ryan. To me, Mayfield makes more sense in this lineup, if you're already playing 3 players from that game.

Again, I think this was an optimizer lineup. Unless I'm missing something, I can't imagine how a rational person would pick Matt Ryan.

Honestly, the only thing that makes sense in this lineup is THE GREAT HUNTER RENFROW.

So what do I take away from this?

• QB-RB/WR/TE stacked lineups are still good.

• DST-RB stacks don't seem to be necessarily bad or good.

• Non-QB stacks can do well, but they don't always make a lot of sense to me.

• I wouldn't avoid playing other stack types (2-2, 3-2, 2-2-2), but I wouldn't force them through optimizer rules.

• If I insist on building multiple stack types, I would build half as 2-man stacks (QB+WR) and the other half as a mix between other stack types.

As always, if you find this useful, please let me know.

Also, I don't have all the DFS answers. I'm trying to learn, just like you. So if something looks off, doesn't make sense, or you have a better idea, please let me know.

Thanks again for reading and I'll be back next week!