Best Ball Win Rates & What They Tell Us (Fantasy Football)
Winning is fun. Actually, it’s quite addictive that even if you’ve rarely tasted victory in fantasy football or any other contest, you remember that feeling of that one time… And boy was it good.
I’ve been regularly getting roasted in CandyLand to my four-year-old to the point where I ran the mathematical probabilities after losing for the sixth time in a row. My son is an outlier and there’s no way he should be winning at this type of rate on a game that is purely based on DRAWING CARDS FROM A DECK. (I’m not mad about it though.)
Speaking of win rates… I hope you enjoyed that professional transition. On the most recent Fantasy Footballers DFS podcast (Best Ball Strategy & Win Rates + Quoting Dolly Parton), Betz and I discussed Best Ball and win rates diving into one of the great metrics for the format but it probably is worth outlining even more. Let me define the term and the key factors used to give context to a player’s win rate.
Note: Win rate data in this article is credited to the RotoViz BestBall Explorer App.
Defining Win Rate
“There Can Only Be One” -Highlander
We care about win rates when looking at Best Ball for one simple reason: it’s all that matters. Ok, so it’s not everything but in a format where you must go full Ricky Bobby “if you’re not first you’re last”, it’s valuable to see which players produced outcomes that won BestBall leagues. Simply put, win rates show us how often this player ended up on a winning roster. It’s calculated by taking the total number of times a player was drafted and dividing that by the number of winning rosters they ended up on. For reference, on an individual player basis, anything above 8% is really good. 10-15 percent and you’re cooking with gas. 20+ is otherworldly and you need to frame that bad boy.
Let’s use Alvin Kamara as the example because he paced all RBs with a 23.6 percent win rate according to RotoViz’s BestBall Explorer App. In 5,305 recorded BestBall10 drafts last year, Kamara ended up on 1,254 winning rosters. It’s impressive considering he was a first-round RB albeit drafted behind guys like Christian McCaffrey, Saquon Barkley, and Ezekiel Elliott.
Over the last six years, Kamara’s 2020 campaign ranks 7th best among RBs. Here is the top-10 list:
Maybe you look at this list and at first glance, it just seems to be a who’s who of top fantasy performers. CMC’s dominant 2019, Todd Gurley‘s ridiculous 2000+ scrimmage yard 2017 season, Devonta Freeman came out of nowhere, and David Johnson‘s RB1 season. But then you get down to #8 and you see J.D. McKissic‘s name…
Wait, so how did a player who finished as RB24 in 0.5-point leagues and RB17 in PPR end up being so valuable in 2020? How did he have a historically great win rate despite never finishing above RB8 on a given week? Heck, McKissic would’ve been chilling on the bench of your BestBall rosters the first month of the season when he was the RB50!
If you want to zoom in on 2020 specifically, here is the top-15 and this can help give a clearer picture of where win rates stem from…
At the top, it’s easy to identify RBs like Alvin Kamara, Dalvin Cook, and Derrick Henry as true difference makers. Those dudes were unbelievable in 2020. McKissic’s win rate now has some context when we see that he was drafted only 130 times out of the 5,305 sampled. Based on that win rate, he graced 30 winning rosters… that’s it. But we need to dive in a bit further…
The Key Components of Win Rate
McKissic ending up on 30 winning rosters is only part of the equation. For example, James Robinson‘s win rate looks elite at 20.43 percent but because he was only drafted in seven percent of leagues, his win rate isn’t as impressive. Only 76 teams enjoyed the benefits of Robinson and he likely would’ve been even higher if not for flaming out at the end of the year. Keep in mind that of big tournaments like the BestBall Mania II on Underdog, the final three weeks of the season aren’t just important… they are everything. Robinson missed two games and his volume slowed down at the very end.
Maybe you can admit that seeing a PPR maven like McKissic and UDFA wunderkind James Robinson isn’t crazy when you jog your memory of the two players and you can recall that they were both super fantasy relevant. But as we go a little deeper, why are guys like Rex Burkhead and Jerick McKinnon on that list? Do you remember anything that Rex Burkhead did in 2020?
Burkhead finished as the RB44 in a season where the Patriots sputtered. He was in a clear RB timeshare. He didn’t play another snap after Week 11. But he did have three top-7 performances including a Week 3 versus Las Vegas where he finished with 31.3 fantasy points. Oh, I’ll throw in that he was one of our dart throws mentioned that week in DFS coming in at under one percent rostered. Cash money!
Win rates tell us more of a story than we realize and far more than an end-of-season fantasy finish. If win rates are calculated based on how often a player ended up on a winning roster, we also make take into account how often they were drafted and where they were drafted.
Perhaps the most important distinction for win rates is opportunity cost. What did you have to give up in order to select that player for your roster? Obviously, spending early draft capital on high-end RBs like Christian McCaffrey or Saquon Barkley did not work out so well in 2020 due to the injury bug. There’s no way you could forecast that but you especially felt the pain in BestBall leagues. You couldn’t replace that player. You couldn’t pick up a backup like Mike Davis (the RB6! from Weeks 3-14) or Wayne Gallman (the RB6! from Weeks 7-13) like you could in redraft.
In other words, misfiring on an 18th round pick does not cost you nearly as much as a Round 1 guy. McKissic’s impact wasn’t as great due to only being on 30 winning rosters but if he turns into a nobody that never scores a point for your squad it’s ok. Win rates reflect just a percentage based on the number of times a player was selected, not necessarily the true impact of their fantasy performance. We need more context to illustrate this metric.
The other key factor for painting a fuller picture with win rates is where a player was taken in terms of their average draft position (ADP). Players can have different win rates in different rounds because taking a player in the 1st versus the 2nd matters. It comes back to our old friend: opportunity cost.
Let me give you an example with the most valuable player in Best Ball in 2020: Travis Kelce. Yes, if you were among the smartest people to draft Kelce early in 2020, you felt like a true mastermind. He only put together the best fantasy season of his career as the far-and-away TE1 with 260.3 fantasy points.
But when compare Kelce in the 1st and Kelce in the 2nd it was vastly different. Kelce in the 1st round had a 0.5 percent win rate. However, if we look at Kelce as a whole, he paced the league at 24.4 percent. That’s a wide of a gap as you are going to find. Reaching for a player (no matter how great of a season they might have) is not optimal because of what you are passing up in that position. Opportunity cost FTW!
If you did take Kelce too early in the 1st round, you likely were passing on RBs like the aforementioned Henry or Kamara. Even at the back of the 1st round, you could’ve easily passed on Tyreek Hill or Davante Adams. There is a cost. Maybe you got lucky and snagged Adams in the 1st and went Kelce in the 2nd. If you did, good Lord. Those teams were almost automatic with 93 percent of them finishing in the top six percent of leagues.
2020’s Win Rates By Round & Position
I want to pique your interest one more time with win rates and tease an upcoming series where I discuss how to approach each of the four main positions in BestBall by looking at historical data. Here is a visual look at win rates from 2020 based on the four major positions and what round those players were taken in.
While this chart does not give you player-specific insights, it does allow you to sit and think back to where certain players were taken in 2020. Who was good for fantasy in 2020 and where were they taken? As always, information like this is descriptive of last year, not prescriptive for 2021.
I’ll follow this up by going position by position highlighting historical trends and how to approach them in 2021.