TL;DR: Does anybody here have insight into "optimal" strategy for fantasy football auctions? Or how to punish other owners' mistakes or deviations from the optimal strategy?
I have some thoughts myself, but am curious to see if anyone has thought about this issue deeply as well:
The importance of strategy in fantasy football auctions.
Auctions are full of decision points where managers can - and invariably will - make errors from the optimal strategy. Those points come from every nomination, bid, and selection. And the manager with the least impactful errors will end the draft with a huge advantage - I theorize it is often equivalent to the value of a buy-in in most traditional leagues. That is, where players each buy in for $100 and the payouts are $750/350/100 for 1st / 2nd / 3rd, the best drafter’s team may be worth almost $200 immediately after the auction. The worst’s will be worth $50-75, depending on how bad your league mates are.
Anecdote to illustrate a single decision point in a fantasy football auction.
You are in a competitive 12 person PPR auction league. Everyone spends $200 auction bucks to pick their players, and it’s about 33% through - you have $120ish left and most other teams do too. You’ve done your research pre-draft and know you want a balanced team. You especially like Travis Etienne. Online sites like fantasy pros say you can get him for $28. Not in your league. Yahoo!s platform recommends managers bid up to $34 for him. He’s gone for $32 and $45 in your mock auctions online, but you still like him below $40.
Now, it’s your turn to nominate in the fourth round and you’ve got 20 seconds to do so. You’ve already secured Davante Adams for $51, Lamar Jackson for $19. The clock ticks down from 15 seconds. Most of the good QBs, RBs, and TEs have come off the board already, and you’re a bit frustrated you kept getting outbid by one dollar on each elite running back. 10 seconds. But there are a few players left you’d be happy to get. 8 seconds.
Do you nominate Travis Etienne - 6 seconds - your favorite RB - or, uh, 4 seconds - Dalton Schultz, the TE you targeted
3. as undervalued predraft? 2. You click wildly and get your nomination in with 1 second left. You selected Etienne. He’s the last good running back and your favorite player; you have to get him. But your friends in the league know you like him, they have $130 and $150 left to spend, and they too want the last good running back. The bidding war over him skyrockets - at $35, you begin to wish you had just spent an extra $4 on Nick Chubb. At $45 you wish you had taken the flier on Tony Pollard. And when Etienne reaches $50, you can’t stomach clicking the bid button again. You’re nauseated - your rival gets your favorite player, the next best RBs are injured to start the year, and you have to shift your strategy wildly.
What matters in a fantasy auction? And why?
I seek to identify the decision points that matter in an intuitive fashion. I struggled to find articles about optimal strategy or meta strategy. Most statistical analyses - often at rudimentary levels - focus on positional allocation of draft/auction resources (i.e., 0 rbs or wrs within the first 3-4 rounds or below $ bids, the effect qb+wr/te stacks, etc.) While an optimal auction strategy requires too many inputs to accurately statistically model for each individual league and its players, there should be some way to identify the biggest goals to capture what one could call expected value, or EV, during a fantasy football auction.
The importance of expected value (EV).
Imagine a $100 buy-in budget 12 team league that pays $750/350/100 to first/second/third. The EV of a fantasy football league of identically programmed robots would be $100, as the prize pool equals the sum of the buy-ins. Each robot would be equally likely to finish 1st, 2nd, …, 12th as its rivals. It would thus expect 0% return on investment.
Because real humans are not identical robots and we are each fallible, each of us will make mistakes that “leak” EV to the other players. As an extreme, somebody who spends $190 of their auction budget on kickers and $1 fillers at other positions would be drawing very slim to place in the prize-placing positions - his EV might drop, postauction, from $100 to $0.10 or so. (I would make such a player a 1000-1 dog to make the playoffs in a competitive league, much less finish top 3). The $99.90 of EV he leaked would be distributed across the other eleven players in proportion to how successful their own auctions were.
A successful drafter captures more EV from the suboptimal drafters than it loses from its own mistakes. I have no mathematical basis for this belief but the anecdata from the leagues I play in, but in casual leagues like my long-term friend group, I suspect the equity to breakdown in a range from $180 for the best team to $50-70 for the worst while in competitive leagues I notice much more parity amongst long-term results and draft quality*. While draft quality is subjective at the time of the draft, I personally hold a mangers' poll post-auction each year in my leagues, whose consensus often tracks the season-long quality of the uninjured players. The managers' consensus on the worst teams seems to correlate more positively (i.e., accurately) than their consensus on the best teams, although the consensus of the best teams are often more fractured than consensus on the worst.
Defining player valuation.
Accurate player valuations Player valuation is much more nuanced than who will finish with the most fantasy points - the traditional meaning of QB1 is the QB with the most fPTS by the end of the year, but such a metric wholly fails to capture the most valuable fantasy QB. While the highest scorer and most valuable will often align, additional factors must be considered.
Risk.
An important facet of EV maximization will be risk management. The market value of each players identifies this risk (i.e., compare the 2023 auction valuations of Justin Jefferson with Cooper Kupp). Managers who are afraid to take risk when it is improperly valued in their own league will lose EV, but so too will managers who assume too much risk by overpaying for it. And risk compounds against price, such that higher priced players carrying great risk are far more detrimental than lower priced players.
Consistency.
An RB who scores exactly 30ppg may be more valuable than one who scores 300p in one game and 0 across the next nine. Most managers intuitively grasp this concept and it’s baked into “inconsistent” players’ values - in the past, think of Desean Jackson, or now Taysom Hill. This tends to be one of the more easily predictable metrics - players who score a lot of big-play or gadget-play touchdowns are likely going to be inconsistent from game-to-game, while players who get a lot of touches/targets will be more consistent.
Salience.
Next, consider that all games are not created equal. Targeting players with soft early matchups provide great sell-high value, while players with soft playoff schedules can boost your EV well beyond similarly-ranked counterparts. I tend to place a lot of weight in the middle of the season on trading for players with soft playoff matchups, and jettisoning my early overachievers who have rough matchups down the stretch.
Watchability.
All else equal, I suspect that exciting or popular players are overvalued in auction formats.
