Underdog
Network

;

Best Ball ADPs - Is It Okay To Reach On Players?

May 26th 2022

Hayden Winks

This column will not be “don’t have player takes”, which I 100% completely disagree with. This column instead will encourage drafting “your guys” at good prices and will find when it’s reasonable to draft your targeted player ahead of ADP.

Best Ball ADP

I think there are inefficiencies in the ADP market, even on Underdog Fantasy where it only pulls from paid drafts and updates daily on a 48-hour rolling window. Some of those inefficiencies are centered around player archetypes, positional value as a whole, and then just team/player projections.

But even with that, we shouldn’t constantly be reaching to draft these undervalued players if we can project that they’ll be available at our next pick. We’ll calculate the odds that these players will be available now:

The data I’m pulling from is Underdog Fantasy’s Best Ball Mania II, specifically looking at Rounds 4-14 because I needed to eliminate early and late picks to get a true distribution. To no surprise, it’s a completely normal bell curve with each player's most likely draft pick being the draft pick at their current ADP, followed by one pick before ADP and one pick after ADP. In fact, 84.2% of draft picks in these middle rounds are made within one round of the current ADP, meaning 12 spots before or 12 spots after ADP. 

What this chart shows is that for every time we draft a player 12 spots before ADP, there are just as good of odds as someone drafting this same player 12 spots after ADP in the next draft. I think we constantly compare our picks to ADP when we should be thinking about the times when someone is scooping up this same player sometimes 12+ spots after ADP. That’s the team that has a higher ceiling and could haunt us later in best ball playoffs.

At the same time, I’m definitely not arguing to be robotic and draft the top-available player in the queue every single time. That’s dumb. There is some wiggle room here, as the chart shows. The teams that performed best last year were the teams that drafted players within 10 picks of ADP just about every single time. These “ADP Loiterers” seem like party poopers, but they are consistently getting great value and letting the board fall to them, unlike the “I Know Better Bros":

Reaching To Complete Stacks

Here’s where this science turns into art, however.

The goal of our drafts is not to have the most value compared to ADP (though this is correlated to success). Instead, the goal of our drafts is to be good enough to have a top-2 team in Weeks 1-14, then have a good enough team to finish 1st in a 10-person group in Week 15, then have a good enough team to finish 1st in a 16-person group in Week 16, and then score the most points in the Best Ball Mania III finals in Week 17. To do that, we must apply Contingent Based Drafting in order to add correlation at ceiling outcomes, in addition to scooping up values and picking the best players. (Read that column if you haven’t already.)

To add correlation at ceiling outcomes, we will occasionally have to make slight reaches to complete stacks, to grab the last remaining player in a positional tier, and to finish the desired roster construction based on how we drafted early in our drafts. The questions then become, how far ahead of ADP should I reach to add this correlation? And what are the odds this player is still available if I try to wait until my next selection?

Well, here we go:

The primarily green column answers this question: “What are the odds a player whose ADP is ___ spots after my next draft pick is available?” The primarily red column answers: “What are the odds a player whose ADP is ___ spots before my next draft pick is available?”

Example 1: I drafted CeeDee Lamb with the 18th overall pick. I’m in the middle of Round 6 (66th overall) now without a quarterback, and Dak Prescott certainly makes a lot of sense as a stacking partner to Lamb. It’s too early to draft Prescott now with his current ADP sitting at 84.2 overall, but will he be available when I pick in Round 7 at 79th overall? I’d be drafting Prescott 5 spots ahead of ADP if I did, so the table projects a 72.3% chance of him being available. To me, it’s worth drafting another player in Round 6 and then hoping (with reasonable odds) that Prescott is available in Round 7. 

Example 2: I drafted Cooper Kupp with the 2nd overall pick. I’m in early Round 7 (74th overall), and I don’t have a quarterback yet. Matthew Stafford would add correlation to my Kupp team, but his current ADP is at 91.1 overall, so it’s too early to draft him now. My next pick in late Round 8 is at 95th overall, so if Stafford is available then, it’d mean he’s fallen 4 picks after his current ADP. Based on this table, the odds of this happening are 30.8%. To me, those are just high-enough odds to grab another player in Round 7 and hope Stafford is available in Round 8. Others will disagree.

Consider this table a loose projection for a few reasons, however. This general data is pulled from Rounds 4-14, and earlier picks have a tighter distribution of where they’re picked than later picks. This general data also doesn’t include the changed odds of waiting for your stacked QB to fall, meaning if the other drafters are also trying to stack, then they won’t be going out of their way to draft Prescott/Stafford like we were in these examples. Finally, this general data doesn’t know if the room is hammering TEs early or punting RBs or whatever, so odds can shift based on which positions are being valued more or less in your particular draft.

Conclusion

In general, it’s good to draft players who've fallen after ADP, it’s fine to draft players a few spots ahead of ADP, and it's bad to reach on players 10+ spots ahead of ADP. This doesn’t mean we can’t have player takes and must be robots to ADPs, but it does mean we should aim to get good prices on “our guys”. Sometimes we need to make slight reaches to complete stacks and add ceiling correlation, but we can’t do that every single time. With the table shown above, we can generally calculate the odds our targeted player will be available at our next pick. Hopefully that makes the galaxy-braining of if we should reach on a stack or hope this player falls to our next pick easier to comprehend.