The Pontifications of Lord Zola: The A is for Average
Lord Zola explains the fallacy of using average draft position and suggests a simple alternative.
What if I told you I had some data for you to use in your drafts and it was generated by a bunch of losers?
Well, that’s exactly what an ADP (average draft position) report is. Think about it; one person wins each league while the rest don’t. But yet we’re supposed to glean some tangible information from where a bunch of non-winners chose to draft the players that resulted in their losing teams?
OK, this is a bit of hyperbole but it does help frame the ensuing discussion. Using ADP data is nice but it’s not the be-all-end-all elixir to a successful snake draft. It’s a tool, nothing more.
You’ve no doubt read this before but I’m going to present the argument from a different angle. Others can warn you about using ADPs generated from outdated information, different leagues, autopicks etc. The implication is if those are the issues then if they didn’t exist the ADP was useful. Yeah, maybe a little more useful, but not to the extent most think.
Let’s jump to the end. An ADP is useful to help time some picks to maximize the potential you put on your roster. You get an idea where the player may be drafted so you know to take him earlier to make sure you get him. But there’s a big problem with that. To help explain the reason, let’s think of things in terms of auction dynamics.
In an auction, you no doubt have a list of bid values. These are either directly based on your expectations for each player and are generated by some formula typed into a spreadsheet or they’re intuitive based on years of experience. Either way, the values represent how much you feel the player is worth. If you’re able to purchase the player for less than that value you have a potentially better return on your investment. Some will contend there are reasons to pay more than that value (which will be a topic for another day) but the bottom line is if you pay more, you minimize if not mitigate your potential return on investment.
Let’s focus on that last thought. Overpaying takes away your potential return on investment. In an auction, you drop out of the bidding when it gets too high. If you think paying that much hurts and not helps your team, you, albeit reluctantly, let someone else purchase that player.
THE SAME THINKING SHOULD BE APPLIED IN A SNAKE DRAFT. Yes, I am yelling.
Jumping a player on the ADP is the same as overpaying in an auction. The difference is in an auction, the overpayment may be a buck or two, sometimes a little more. In a draft, it can be several draft spots which if converted to dollar values, far exceeds a buck or two depending on when the jump is made.
Chances are, the reason is we have a concrete number on front of us in an auction to guide our bidding. This number is based on our expectations. Mentally, it is a whole lot easier justifying whether or not to bid based on the bid value. We aren’t comparing our bid value to the auction equivalent of an ADP. Yes, they do exist by averaging the winning bids of several auctions. Actually, think about how silly that sounds - judging a bid on the average that others bid.
BUT THAT’S EXACTLY WHAT MANY DO IN A DRAFT. Sorry, I’ll go back to my inside voice.
In a draft we have a ranking or cheat sheet, along with an ADP. The thought process isn’t to acquire players we like under the bid value. For too many, the process is trying to figure out when to draft guys we like based on the ADP. For some reason, having a player ranked 75 and taking him with pick 65 doesn’t trigger the same “NO” that bidding $24 for someone valued at $20 does. Mentally, we are fine with dropping out of the bidding in an auction but we feel compelled to take a player early in a draft. More confounding is in an auction, the overpayment may only be a buck or two but in a draft, it could equate to more. Why?
One reason is in an auction, if you drop out on a player you still have the ability to get in on everyone left in the player pool but in a draft, you are limited to who is left at your turn so the inventory may be limited. This makes some sense, but an overpayment is still an overpayment. The idea isn’t to balance or justify overpayments with picks you made under value. That would result in a team nestled in the middle. You know, average.
While taking a player earlier than your rank should trigger the same “NO” that going another dollar in an auction does, it simply isn’t the case. Just think of how many times you hear someone allude to an ADP and justify a pick because they knew the player wasn’t getting back to them. We need a better means of triggering “NO”.
Every pick in a draft has an expectation. Given that there is upside reward and downside risk factoring into our expectation, we should not be making picks if the player does not have a plausible chance to meet or beat the expectation correlated with that draft spot. It’s really as simple as that. If jumping a player up the ADP means he doesn’t have a reasonable chance to be worthy of that spot, JUST SAY NO.
As suggested, comparing your rank to the draft spot doesn’t suffice as a guide. Perhaps we should borrow from our auction thought processes and attach a dollar value to each spot in the draft and instead of a cheat sheet, have a dollar value with each player. Or at minimum, add a dollar value to the cheat sheet.
What follows is a table showing how much each draft pick is worth in terms of dollar values. This is applicable for all sizes and formats. That is, the numbers may not be exact, but they are certainly close enough to be useful.
|1||48 to 30|
|2||30 to 25|
|3||25 to 22|
|4||21 to 19|
|5||19 to 17|
|6||17 to 15|
|7||15 to 14|
|8||14 to 13|
|9||13 to 12|
|10||12 to 11|
|11||11 to 10|
|12||10 to 10|
|13||10 to 9|
|14||9 to 9|
|15||9 to 8|
|16||8 to 7|
|17||7 to 6|
|18||6 to 5|
|19||5 to 4|
|20||4 to 3|
|21||3 to 2|
|22||2 to 1|
Note how much bigger the difference is between players at the top of the draft. Jumping a player up the ADP a round or two in the early portion of the draft is the same as overpaying four or five bucks in an auction. In the middle it’s not so bad.
But the point is, you don’t need an ADP at all if you simply follow the dollar per round guideline and compare that to your cheat sheet expected value earned. Drafting a player you peg to earn $12 in the sixth or seventh round is the equivalent of overpaying $3 or $4.
That said, on occasion an ADP is useful to help gauge the market if you’re trying to acquire two players and you want to decide which to take first to maximize the chances to get both. But it isn’t useful to help decide how far you need to boost up someone like Bryce Harper, Billy Hamilton or anyone else. Boost them too far and they do more harm than good.