Soccer DFS Strategy: Stacking
**Originally posted on January 2nd, but still applicable** With no Premier League nor Champions League games this week, I want to take a look at a very important part of Daily Fantasy Soccer – stacking. Stacking is referred to having multiple players from one team in your fantasy lineup and I’ll look at whether or not that’s a viable strategy in soccer. To answer that, we can look at several different things, but let’s start with something basic – like how does one win a GPP?
The answer is goals. Ever since DFS Soccer started, I’ve looked at winning lineups in GPPs and they have always featured a very high amount of goals. Since I don’t have the data from DraftKings from last season, I’ll take a look at the last six big slates and examine the top 10 lineups in the biggest tournaments of those slates. Amongst those 60 lineups, 55 of them had at least five goals within them.
That’s a lot of goals, and I think we can safely conclude that winning a GPP without at least five goals in the lineup is highly unlikely. But finding goals is hard, especially on just one team on a given weekend. So, to answer whether we should stack, we should next answer the question of how often teams score lots of goals in one match.
How Often do Teams Score Lots of Goals?
This is data from every single Premier League match last season. Looking at this, we see that most of the time, teams don’t score a ton of goals. If you want to see that in percentages:
Unfortunately, 62% of the time teams only scored 0 or 1 goals. 85% of the time, teams scored 2 or less goals per match. But what about those 15% of the time when teams score three or more goals? Well, let’s look at how often players from the same team did well in terms of DraftKings points.
Let’s take a closer look at how often players from the same team do well, maybe that will tell us more. The following charts show how often players from the same team achieved at least 15 or 20 points in one match based off of attacking stats only. This ignores win and clean sheet bonuses, as I’ll discuss defensive stacks later on. This is purely offensive statistics and the data is from all Premier League matches last season (except five or six due to double game week data issues).
To win a GPP, you generally need at least 15 to 20 points from most or all of your players. Based on the numbers above, the verdict on stacking is pretty negative. There are obviously times when you can stack, as evidenced by the 6.5% of the time when three or more players from a team exceeded 15 points, or 5.7% of the time when two or more players exceeded 20 points, but these are hard numbers to go after. The question then becomes, WHEN should we stack?
The first issue here is with figuring out which teams will score three or more goals in a given game. That’s generally the easy part – we can look at sports book odds, form, table position of both teams, expected goals, overall squad quality, home or away, and a number of other factors and use them to determine who the heavy favorites are. The second step is figuring out who will score multiple goals on that team, as often times we see a lone striker getting the majority of chances – such as Sergio Aguero this past weekend when he scored five goals in 20 minutes. That’s a bit harder, but we can again predict who will get the most chances and thus has the highest chance at a goal. The question now becomes whether we can predict times when multiple players from the same team will have a great gate, and whether those chances will be higher than playing players from different teams.
The answer, then, is quite situational – you should stack only when the slate allows you to. Let’s take a look at two different slates that we recently had to drive home the point. Below are winning percentages for each team in the last Premier League slate.
Looking at this slate, we only had one team that was a heavy favorite to win. In a slate such as this one, it’s absolutely fine to stack players from the heavily favored team– in both cash and GPPs. In cash, the reasoning is that Manchester City will have the most goals in the slate, so you have the highest chance at securing goals in your lineup by having multiple Manchester City players in your lineup (generally two). In GPPs, the reasoning is that in the case Manchester City go off for a three or four or five goal match, your two or three players will have enough goals and assists between them to give you an extremely high ceiling. Looking at the top lineups in this slate, almost all teams had a Sergio Aguero/Kevin De Bruyne stack.
Let’s compare this to another slate we recently saw:
In this slate, the matchups are a bit more even, and there are a number of teams with a chance greater than 50% to win. Here, stacking would be unwise in cash games as you want to diversify your shares amongst different teams that can all go for multiple goals. What you want to do is find likely goal scorers in all matches, while also finding players with lots of crosses and shots from the favored teams, as they give you a good floor in cash games.
In GPPs, stacking is a bit trickier. One could argue that stacking two players from one team give you a chance at picking up an assist and a goal at the same time, but that strategy is generally not recommended. If we look at last year’s stats, the most common assist-to-goal scorer combination was Cesc Fabregas assisting Diego Costa, which happened six times. The second highest combination only occurred four times. Four times in a season is not an event you want to chase in a week-to-week fantasy game. If you’re going after anything, go after multiple goal games where your players will have a high overall chance at picking up goals and assists, not necessarily assisting each other. Having two attacking players from a team expected to score two goals is almost never a good idea, especially in GPPs.
Looking at the winning lineup in this slate, the winner had one forward and one midfielder from Real Madrid, while the rest of his M/F spots featured three different teams. His Real Madrid midfielder didn’t even score, and he won primarily because his other three M/F players all scored. Second, third, and fourth place featured five players from five different teams in the M/F spots.
The verdict here is pretty straightforward. Only stack when there is one or two extremely heavy favorites in the slate. An extremely heavy favorite would be one where the winning percentage given by the sports books is above 75%.
The Defensive Stack
A common strategy on DraftKings is to stack your goalkeeper with two of his defenders. A win and a clean sheet will earn you 20 bonus points amongst those three positions, which is a very significant amount. The strategy also carries a lot of risk, as two goals conceded will likely hurt your chances at winning a GPP and absolutely destroy your cash game chances. The answer for cash games is simple, avoid stacking the full defense at all costs. In slates where there is only one big favorite for a clean sheet and a win, you can stack a goalkeeper with one defender, but I would almost always recommend having at least one defender from another team.
In GPPs, stacking a defense is a viable strategy, but I wouldn’t stack one defense in more than 20% of your lineups on a given slate. What I like to do is stack the top three defenses in a slate in 15% of my lineups. That way, 45% of my lineups consist of three different defensive stacks. The other 55% of my lineups are mixed and matched.
Hope you all found this helpful. Please don’t hesitate to ask any questions in the comments below or on Twitter.