This will be part 3 of my 3-year review series. Each Saturday between now and the NFL season I will be releasing a new section.
Week 1: Snake Draft Strategy
Week 3: Auction Draft Strategy: Nomination Strategy
Week 4: Are you smarter than ESPN: Lineup Decisions
While last week I focused on where to spend your money during an auction draft, today I want to try to understand how to navigate the draft. Specifically, the one thing you can really control, who do you nominate for bidding when it's your turn. This is your chance to control which player everyone is focused on. Whoever you nominate will be drafted by someone, so how do you choose? Nominate someone you don't like to remove money from the board? nominate a lower ranked player to get them as a bargain? Or just grab the guys you like and forget about the rest? While I can't answer all of these questions, today I hope to dig into some of them.
The Data
As with last week's post, I will be using the draft records from ~1500 auction leagues on ESPN. From this, I can know in what order players were nominated, giving me insight who the options were when a player was nominated. Throughout all of this we will be trying to maximize value, specifically did the player go for less than ESPN's suggested price. While ESPN may not be the most accurate prices, it should provide a good baseline for this analysis. As we saw last week, accumulating value is a great way to increase your odds of winning the championship and it should be the goal of your nomination strategy, how do you ensure that you are getting players at value?
I will also talk about tiers in the results to follow. Similar to the work of Boris Chen and Fantasyfootballtiers.com I went and created historical tiers of players based on their ADP in snake drafts and ESPN's suggested auction price. For each year I created 30 tiers that span across the QB, RB, WR, and TE positions. These tiers will be used to look at trends like does the price of a player spike when they are the last of their tier.
Finally, I know some people love to throw $1 players (including Defenses and Kickers) out at the beginning to see if they can get a random bidding war. For this analysis, I will be focused on players with a suggested price over $5. These are really the players that matter in your draft but please keep that in mind while viewing the following analysis.
Nomination Order: Just Follow ESPN?
Line plots of the Price change depending on when the player was nominated. The colored lines represent each year's result while the black line is the average across the three years.
When we look at the nomination order vs ESPN's suggested order, we see an interesting 3rd order fit. Basically, you want to buy players nominated a lot earlier than ESPN would suggest (say 35 spots early or more) or players a little later than ESPN would suggest (say 10-20 slots later). My interpretation is that there are 2 different explanations for this.
First, players nominated really early are not going to be on everyone's mind. People will think that there will be plenty of time to figure out who slots into their team at that price and they are loath to overspend on role players when the stars are still available. In contrast, players nominated only a little early are on every manager's radar and they can more easily figure out how that player would affect the team they are assembling.
Second, players drafted a little late are better values likely because they are what is left. By dropping a little, some of the managers will have priced themselves out of this remaining player and therefore there is a chance for value. However, this is a short window if you wait too long then everyone begins to think they can snag that player for cheap and will start saving money to compete for the "last player of that caliber", taking away all of the value.
What about position specific categories?
Line plots of the price change based on the number of higher tier players available. mean and 95% confidence intervals are shown.
Here we are examining if there are different trends for the 4 main positions. We can see that QBs are never purchased for value, an effect I suspect is due to removing players you cost less than $5 (where most QBs fall). Interestingly, WRs are basically always being drafted for their suggested price, maybe you save $1 on average in the best-case scenario of drafting a WR nominated 12 slots early, but that's it. Tight Ends you want to be drafting the 3rd best TE at the given moment. I think this follows with the idea of nominating a player early that an early TE is outside the current roster calculus managers are doing, particularly if it is not the best TE available. Though similar to QB, there are likely a number of TE's not included here due to the price cutoff.
That leaves RBs, which is the most interesting position. Essentially you really do want to be nominating the RB's you want early with anything from 5-20 RB slots early coming in at value. Not sure why there is such a range here, but I would definitely look to take advantage of this to get the RBs you want in your draft.
What about within a tier?
Binned scatter plot with a linear regression of the change from suggested price vs % of the tier gone. Each dot represents the mean and 95% confidence interval of nearby points.
To control for the size of a tier, I converted the tiers nomination order to percent of the tier (so if there are 3 players in the tier, the first player nominated would be 0%, the second would be 50% and the third would be 100%). As we can see there is a clear upward trend. Basically, the later in a tier a player goes, the more expensive they will be relative to their suggested value. This is unsurprising as people likely get more desperate to ensure they get a player at a specific tier, leading to higher prices. For those curious, I did check this across specific tier ranges and the results are stay roughly the same, earlier in a tier you can secure a player, the better.
Total Money on the Table
Line plots of the price change vs ESPN's suggested price based on the total amount of money left. Mean and 95% confidence intervals for individual years (colors) and
average (black).
I think this is really interesting. The graph argues that the values are at the beginning or at the end of the draft. Once some money starts coming off the board, people will have a sense for how much players should go for and have a sense of where the gaps are on their team, they will get aggressive. if you are trying to accumulate value you want to avoid this and instead wait until the end when people will be out of money to compete.
Limitations
As with the snake draft post, I am not controlling for league settings which may lead to some variation. On top of that, each auction draft will have its own quirks, so it is important to know your league. That being said, these general trends are useful to keep in mind as you try to navigate your draft.
Conclusion:
So, what to do if you want to get players at value:
Nominate the players you want early or a little late compared to
Similarly, aim for grabbing players while higher ranked players are still available at the position (in particular running backs)
Try to get the first player in a tier
the best values are at the start of the draft or at the end of the draft
Questions? Comments? Let me know at ac@alexcates.com. Want to read more breakdowns like this? sign up for my newsletter. Finally, like what I do? Consider supporting me on buy me a coffee or by signing up for fantasyleaguereport.com.
Comments