Welcome and Best Ball Strategy

Welcome to my first blog post! This blog will be for me to write about any of my sports analytics endeavors that don't make it into a full project. I have no schedule for posts, but hope to get at least one out a month.

Best Ball Fantasy Overview

Lately I have been messing around with a lot of best ball data for underdog fantasy's Best Ball Data Bowl. Best Ball is a style of fantasy football where each week a teams lineup is automatically set after every game has played; maximizing points scored. This is opposed to standard fantasy, where players set their lineups before kickoff attempting to predict the optimal lineup. Best Ball also removes trades and waiver wires, making it a season-long fantasy game completely dependent on the draft. In preparation for their upcoming Best Ball Mania IV competition, Underdog Sports partnered with Fantasy Data Pro's and released Best Ball draft data from the past 3 years. This comes as part of a data competition that ended on August 1st. I first began looking at the Best Ball data back around April when the contest was announced, but with finals and job applications I burnt myself out and only recently started putting a submission together. While this is not my submission I thought I'd share some of what I looked into and found interesting.

I got inspiration to dive back into the competition from this tweet by Hayden Winks. It is a clever graph showing the average total points of teams who drafted x amount of RB's over an entire draft.

I am a casual when it comes to fantasy football and all of my friends are as well. While much I have seen online regarding draft strategy discusses 0 RB, I don't recall coming across it in my leagues and have never played Best Ball before. Below I will share what I found interesting as I sought for more insight into the strategies of Best Ball.

Findings

Position Totals

One of the first things I looked at was distribution of points between positions from last year. Total position points is slightly different in best ball data than it would be in standard fantasy. It refers to the total points a player contributed to their team that season, not the total fantasy points their stats correlate to, meaning if a player had the top 2 QB's in a certain week only the top QB would be credited with points that week.

Outside of QB's, contributed points seem to be relatively normally distributed (although TE does have a slightly raised right tail, likely from elite talents such as Travis Kelce being consistent top scorers). These distributions tell us that WR's score the most points (the best ball rules award .5 points for every reception), a fact many fantasy players likely know. But this doesn't necessarily imply the value for a WR over a RB is great enough for a 0 RB strategy, as there are many other factors outside of points contribute, such as draft value and opponent draft strategy.

Drafting Strategies

To get more insight into the different draft strategies, I separated each drafters first 5 picks into one or more of the categories in the graph below. The strategy names are descriptive and are defined based on their prefix and the position that follows it.

• "No" means that a team didn't draft the given position in the first 5 rounds.

• "Heavy" means they drafted 3 or more in the first 5 rounds.

• "Reg" means regular and indicates drafting 1 or 2 in the first 5 rounds.

• "Good" exists only for the TE position and means 1 was drafted

As you can see from the table we can be pretty confident WRs were more valuable than RBs early in the draft last year - so much so that 0 RB had the highest Average Total Points of any strategy. What stands out to me is a 2 QB strategy was not only more successful than other strategies, but that it performed even better when also using 0 RB. We can reason that this might be caused by the unique rules of Best Ball. While a great QB is very valuable, there is no reason to be surprised if they put up some duds over the course of a season. A second elite QB can make you more likely to have a high scoring total at the QB position. It also dilutes the QB position among the other 11 teams in your draft.

But what rosters can help take home the top prize of a Best Ball Mania and similar competitions?

This video by Peter Overzet is about how week 17 is all that should matter when drafting in Best Ball Mania. He discusses how once in the playoffs, many teams have the same players (this is because the playoffs are made up of each drafts league winner). Since your opponents very likely have the same players as you, victory - and prize money - will come down to the differences in rosters.

0 RB and 2 QB aren't just more successful relative to the league (last years league wide average total points was 1521), they are also rare. 0 RB was used by 6.72% of drafters, 2 QB by 1.56%, and by 0.29% when combined. None of this guarantees success, but I'd suggest one of these early round strategies if you want to maximizes your chance of winning.

Pick Position

One last thing I found interesting was this chart.

Oddly enough, the first overall pick is the only slot that appears to have a significant deviation. I was expecting the distribution to be relatively uniform, as each draft has a random pick order. While this could likely be due to chance, there is a potential that certain aspects of the first pick contribute to a disadvantage, such as pressure to draft the "consensus number 1" or the large amount of time between the first pick (1 overall) and 2nd pick (24 overall). Take this with a grain of salt - it's trivial - but I found it too interesting not to share.

If interested, you can find the code I used for the visualizations in this blog along, with code for everything else I looked into with this data but didn't share in my GitHub repo. My apologies for the lack of code comments, something I will fix in future posts.

My submission to the Best Ball Data Bowl can also be viewed on their public GitHub repo here. I'll be the first to say it's not my best work, but I'm proud of what I put together with the time I gave myself. It includes some of the points in the blog, as well as some other methods of analysis.

I hope you've enjoyed my first blog! Comments, questions or constructive criticism is always welcome, feel free to reach out to me on socials. Thanks!