Playoff Best Ball Thoughts

The Ultimate game for Galaxy Brainers

Hey Folks,

I wanted to get some thoughts up on playoff best ball tournaments: it’s something I’ve dabbled in content for in past years, and have played, but admittedly not at the same level of volume as regular season drafts.

If you are looking for elite content on the subject: I could not recommend enough Pat Kerrane’s work on Legendary Upside, and specifically his collaborations with Daniel Racz and Chris Gee. I think everyone who subscribes to me knows how highly I think of Pat in this space, but I consider Daniel one of the sharper guys in the space, especially for playoff best ball; something he puts a ton of brain power and time towards. So to be clear: I do not hold myself out as a one stop shop for elite playoff best ball advice, but if you consider me a part of your process for your fantasy football endeavours, I’m more than happy to let you in on some of my thoughts for a format I find quite fun. I’m certainly happy to share my perspective and how I’ve been approaching things so far this year.

I’m hopeful that if you like this stuff I’ll have the time to write up some more specific builds later on, but consider this a more generalized take on some of the constant FAQs regarding this game and giving you an overview of what I’m emphasizing. This first piece should be especially useful if you’re not massively plugged in to the playoff best ball world but are interested in diving in further.

---

Given this is a best ball post, it seems only fair I update you with some general accountability on how my best ball season is going overall.

I played a total of 350 teams this year, 150 of which were in BBM and the rest of which were almost entirely in low-stakes tournaments between the Big Board, Puppies, Poodles, and Pomeranians. I advanced 24/150 BBM teams which was 1 short of the “neutral” advance rate. It’s obviously not a particularly high number, but considering my most-drafted player was injured in the first half of game 1, it’s hardly shocking.

I will say also that I simply don’t emphasize advance rate in how I draft my teams. In fact, I pretty much do the opposite, trying to emphasize younger players, suspended players, or injured players who I hope will have lower tournament-wide advance rates due to slow starts, allowing me to get the chance of holding them at lower rostership going into the playoff weeks. There’s a balance to this of course, but I make a concerted effort to forgo some level of regular season advance rate for hopefully improving my playoff advance rate.

To be clear, not all these types of bets work out as I laid out. Players like Rashee Rice, Alvin Kamara, and Jamhyr Gibbs stand out as assets I was overweight on in drafts who hit in exactly the way I envisioned - pairing slower starts with standing out as elite late-season options. For players like Breece Hall and Jonathan Taylor; obviously this didn’t come to pass. Other major stands like Jordan Addison and Sam LaPorta turned out to be fantastic picks but were immediately so from the jump: there's only so much you can control. Meanwhile, while Adam Thielen stands out as an example of a veteran I often talk about in the off-season who helps your advance rate, but is likely to be a duplicated piece who projects worse come playoff time than in-season, Mike Evans was a massive boon to your advance rate, a player you’d love to have on any possible roster at this moment, and it was a clear miss by me in draft season to not make him more of a focus.

These are just a few examples to illustrate my point that (a) there are legitimate levers you can pull in draft season to optimize your portfolio for different stages of the best ball season, (b) there is a lot of guess work in how to action those levers, and (c) they are not always mutually exclusive endeavours.

All that being said, for this week at least my thesis came true, advancing 3 of 24 teams to the BBM semi-finals (vs. an expected rate of 1.56), two of which came on the back of a Monday Night Miracle from a textbook “late-season play” in Jaxon Smith-Njigba. Should any of those three make the finals, or any of my nine semi-finalist teams across the board, I’ll do a little write up so you can sweat alongside me.

---

Anyhow, on to today’s main event: the playoffs!!

Overview

Each tournament - and I’ll focus largely on the two main underdog tournaments open right now - works like a condensed version of the season-long tournaments. You draft a team and face off against those you draft with. The top 2 teams advance, followed by another two rounds, each of which is successively more difficult to advance from, followed by a small-field DFS-like final.

Underdog has in the past had just one team advance from the initial pods, which necessitates a major difference in strategy, but for the moment here is the breakdown:

Gauntlet ($25): 2/6 → 2/14 → 1/8 → 400 person final

Mitten ($5): 2/6 → 1/6 → 1/10 →188 person final

Each draft is 10 rounds, and you need to start a total of 1 QB, 1 RB, 2 WR/TEs, and 1 Flex (WR/RB/TE).

FAQs

1. What is the aim of your roster construction?

This is they key issue, and there is no perfect or universally applicable answer, but there are some simple mathematical edges you can implement to avoid losing EV for no reason.

There are effectively two key numbers to keep in mind that you’re looking to maximize. First, how many combinations of five players can you make from the players you have remaining in the final (the super-bowl?) Second, how many viable super-bowl combinations can you create?

Naturally those figures are in direct competition. If you draft all 10 players from just two teams, you will have the maximum amount of combinations of players come the super-bowl, but only one combination of teams to make your team live.

The maximum number of combinations of teams you could put together with at least five total players is technically six. And the maximum number of combinations of you could put together with multiple options on both sides of the bracket is four.

However, the more you attempt to optimize your team to balance out the risk of a team losing along the way, the more fragile your team becomes both during and after the draft. Not only would one injury or under-performance in the super-bowl drastically reduce the odds of a spread-out team, but the more ornate your structure is the more likely it is that essential players in your build will be sniped and you will either need to pick sub-optimal players or wind up with largely wasted picks.

Let me illustrate with an example:

The specific build that lets you hedge bets in each conference requires you to select 3 players from two teams in one conference, including both QBs from that team, and then select any two players from two more teams in the other conference, ensuring that each combination as at least one RB and one WR.

An example of such a team would be:

Patrick Mahomes - Josh Allen

James Cook - David Montgomery - Rico Dowdle

Ceedee Lamb - Stefon Diggs - Brandin Cooks - Justin Watson

This team is a cool experiment in theory, but it necessitates almost everything in your draft to fall perfectly such that you can hit on the right combination of players to make the math work. It also very plausibly relies on the likes of Rico Dowdle, Justin Watson, or others to count in your lineup in the Super Bowl and win you a final. The plausibility of this last part largely depends on what happens in the playoffs (see the next question).

Generally speaking, my approach for *most* teams is to take the path that is easiest to execute in my draft lobby and provides the most optionality I can from there provided my first assumption hits. For me, that means stacking up one team heavily, and then either stacking up one team on the other side to match, or taking shots on several players from teams in the opposite conference to give myself plausible combinations if I make the right bet in the first conference.

Unlike seasonal best ball, I just can’t give you a generalized answer and say this means the optimal strategy is to draft “6-2-2” or “5-5” or anything of the sort, because (as I’ll touch on below) each team you choose to make your primary stack invites its own context for how you have to build. In what I plan to write for next week, I will go team by team among the top contenders and outline how I would build under the assumption I’m playing them to make the Super Bowl.

In any case, there are a few musts no matter how or who you draft:

• If you are playing two teams in the same conference you may as well take both QBs

  • Otherwise you are still only live in the final with one of the teams making it but have depleted your live scenario by allocating resources to a team you can’t win with in the super bowl

  • Generally I would make one team my “primary” and play out the rest of my draft based on the assumption they make the super bowl. Your primary team has to be the team you draft a QB from. Then you can choose 1-3 “secondary” teams depending on the depth and choice of your primary stack to pair RBs/WRs with

• Don’t draft a combination of teams that doesn’t have all positional slots covered

  • For instance, if you play Dallas as your primary but don’t have Pollard (or Dowdle) as a part of your stack, make sure that each of your AFC pairing options include a RB. Ideally, you can cover each positional minimum in your primary stack and then just need to add options at any position with your other pairings

• Avoid likely playoff matchups or other negatively correlated scenarios

  • While the ultimate goal is your teams meeting in the Super Bowl, your odds to get there are better advancing as many of your players for as long as possible. We know some playoff matchups are much more or less likely than others: for instance, Dallas will likely play the NFC South winner in Round 1, and San Francisco in Round 2.

  • If I’m building an AFC primary stack with Dallas as one of my secondary teams, I would be more inclined to use the Eagles or Lions as my second NFC team. Alternatively, if you want to be especially correlated, you could potentially play a team like the Rams, in hopes they spring an upset in Round 1 over the Eagles or Lions, and thus push Dallas out of a 49ers matchup in Round 2, improving your odds in a correlated fashion.

  • This especially extends to regular season scenarios: if you are playing anyone from a bubble-team, even as a one-off, don’t take players from another bubble-team reducing your odds to get all 10 of your players through to Round 1

  • Lastly, the potential combinations of players you will be up against in the finals should you pick the correct teams is so much smaller than in Best Ball Mania that you should be far more mindful of inter-team correlation: if you are taking the 49ers and don’t have Christian McCaffrey on your team, I would argue it’s almost mandatory to take one or both of Elijah Mitchell and Jordan Mason to play for that injury scenario.