Is autodraft a necessary evil?

In Michael Stein’s column posted earlier this week, he wrote the following:

It is true that many league Web sites have a draft robot or settings to automatically draft the next-highest-rated player if the time clock expires. However, in a sense, that unfairly prejudices the teams that follow in the draft order because there is no guarantee that the offending team would have taken that player.

This passage reminds me of one of my pet peeves about autodraft or autobid and begs the question of whether these features are necessary evils or just problems in need of creative solutions. First, let me establish my somewhat selfish beef with the autopick feature; many of you will relate.

Every year there are a few good, but most often not great, players about whom the pre-ranking committee on the league provider site and the general public have incongruous opinions. Most often these are players who have “sneaky value,” meaning their production isn’t glamorous in the glory categories, but they help everywhere and have little weakness or are players coming off a breakout that few saw coming and the public is skeptical of the player’s ability repeat. Sometimes I’m on the side of pre-rank, and sometimes I’m on the side of the public.

The thing that’s common among these players is that they often are drafted later than their pre-rank would demand, either because the non-believer doesn’t believe or the believer is trying to “profit” off the pervasive non-belief. Most often, I’ll actually be in the middle of both sides. I often think the player is a reach at his pre-rank but also often perceive the public backlash too strong and see that player as the best available pick a round or two after his pre-rank.

Most who autodraft won’t actually submit their custom pre-ranks for more than the first few rounds worth of picks, if at all. Now, having the robot select the highest pre-ranked player on the board is rarely an issue to me in the first few rounds, as there is less divergence of opinion at the earliest stages of the draft and very little likelihood that a player toward the top of the draft board will last until your pick if it is more than a few selections away.

But in the middle-to-late rounds, this situation becomes more common. Once people start deviating from the pre-ranks more consistently and widely in their selections, it often starts to become obvious who auto-drafting team X is going to wind up with when his pick comes up seven slots from the present.

It’s frustrating when the autodraft robot makes what you think is a savvy selection on behalf of an owner that you don’t think would have been astute enough to trust a correct pre-ranking of an obscure, yet valuable player. I know I’m not the only person who has cursed to myself that I would have wound up with (my desired) Player X if only owner so-and-so was making his own picks.

Another specific scenario in which this may happen is when there had been a run on a specific position or type of player, and you had abstained, and you know that because of the way other teams are constructed, the other teams shouldn’t really be targeting one of the higher-ranked players on the board.

Unable to dynamically filter by the needs of the roster during its construction, the autodraft robot is primed to draft a player that will do its owner little good but will cap the value you, as the minority owner in need of such a commodity, can extract from that player’s draft position.

Of course, autodraft also will make its share of foolish selections, as well, thereby creating opportunities for value to be reaped by the real live humans participating in the draft, but to me this isn’t a you-take-the-good-with-the-bad write-off. And I guess that belief is rooted in a subjective opinion regarding my fantasy sports values.

Some view autodraft as a neutral tool for those unable to participate in their draft or not confident enough in their personal capacities to make wise decisions. If you hold that view, it stands to reason that autodraft should be a neutral application. But, I don’t share that opinion. I take fantasy sports more seriously than that, obviously.

I value active participation in a league, and among the holiest commandment on that list is to participate in your live draft, in person. I know sometimes emergencies happen, but in the absence thereof, figure out a way to make your draft, plan ahead enough to request a rescheduling of the draft, or at the very least try to find a friend to participate for you instead of ceding control to the robot. I’d rather you give your draft over to a savvier human player than to Wilson.

The fact that I hold this value leads me to feel that using the autodraft should be punitive. So it should come as no shock that I get quite peeved in instances where, in my estimation, one’s choice to opt into autodraft actually punishes ME!

So, what should we do about this? Anything? Is this something I just have to suck up and live with, because sometimes emergencies really do arise and, therefore, it’s unfair to punish somebody who is on autodraft because we don’t know the circumstances that led to that outcome?

I’ve been in live drafts where an owner had computer problems, lost his internet connection and faded into autodraft for a few rounds while the problem was resolved. Even if we feel it’s fair to punish the lazy or intellectually timid auto-drafter, should he who is the victim of technical difficulties be subject to the same punishment?

A Hardball Times Update
Goodbye for now.

In Michael’s column, he mentioned that he institutes a penalty for an owner who doesn’t make his selection within the allotted per-pick time limit. That owner is not allowed to autodraft the highest player on the board, but instead gets his pick skipped and is then put back on the clock at the end of the round, forced to pick last. I think that’s a good idea. Frankly, I’d be okay with instituting the same rule for auto-drafters, with some exceptions, forcing them to simply pick last every round.

I’d grant the owner a pass, meaning standard autodraft rules, under any of the following circumstances (and perhaps some others that have slipped my mind):

  • A legitimate emergency.
  • If they actually submit their own, personal list of pre-ranks that differs substantially from the default, or a list from a third-party provider in advance of the draft. An owner doesn’t have to disclose the list to the commissioner, but it will be obvious during the draft if their pre-rank list is indeed distinct.
  • If an owner experiences connection issues mid-draft, or some other technical snafu, I think a grace period of a few rounds to either fix the issue or institute an emergency Plan B would be appropriate. But I would not grant immunity from the autopick penalty for more than, say, a half-dozen rounds because you have to protect against people just making their first few picks and then logging off—the classic mock draft behavior.

So, what do you think? Is my ire and proposed penalty reasonable? Or am I being draconian, and an irrational and petty tyrant?


10 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Brad Johnson
12 years ago

The only league I commission right now is an auction, keeper league. We had two owners autodraft in the first year which had some bad side effects. One owner held a fire sale prior to the season, getting 25 cents on the dollar in March. The other owner quit mid-season.

Last season, we rescheduled the draft half a dozen times in order to accommodate everyone. The auction autodraft can be particularly heinous. It makes the snake autodraft feature look downright intelligent.

I institute a different penalty. If you miss the draft without a legitimate excuse, you better be damned active or you’re not getting invited back in the future.

But this column gives me another idea. I’ve been trying to insert a number of taxes to disincentivize unwanted behavior. A $10 tax for missing the draft seems interesting.

Dylan B
12 years ago

“Unable to dynamically filter by the needs of the roster during its construction, the autodraft robot is primed to draft a player that will do its owner little good “

Does this really happen alot. I typically just use Yahoo for any fantasy league I am in, and the auto draft will fill starting postiions first before any bench spots.* You may see a player drafted to fill their UTL spot, you wont see them draft someone for their bench while they still haven’t drafted a catcher. I’ve been in drafts where there are a couple of auto drafters and have used that to my advantage, where I have a need for a Power corner hitter and a SS, and I know that a couple of the teams between my 2 picks on auto have already picked a SS. I can see the issue about the competiveness of those auto teams, so as long as they keep up during the year I think it is by far the best solution.

*In the past I have adjusted my pre-rankings to exclude most/all players at certain positions; typically in Football(kickers, def team) or Hockey(Defencemen, backup goalies); that I deam “weak”; that way if for some reason I can’t make the draft I didn’t draft the Green Bay defence with my 8th pick when there was a starting RB available. I could see a similar thing with Baseball, removing all but the top couple of catchers, or any RP eligible pitcher that is not a closer.

Derek Ambrosino
12 years ago

Dylan,

I was mainly referring to statistical needs of teams, and not positional needs.

And, frankly, I think there are plenty of people out there who would actually be well served to autodraft because they rely too heavily of name recognition, etc. Like I said, sometimes autodraft does make better picks for some owners than I think they’d make if they were steering the ship.

But, the question this part of the discussion begs is what about something that is essentially a human generated super autodraft app that is truly an example of artificial intelligence.

What if you create a program where you devise target totals for each category, then add player projection data, including positional eligibility and basically write a script that turns each projected stat line, perhaps weighted slightly for positional value, as a “probability” added to amass a roster that comes closest to reaching all benchmarks, as well as track the categorical projection totals of your opponents and react accordingly when picking?

Could you create your own, dynamic, unbiased script that would basically be the Deep Blue of fantasy sports drafting? …And, what will happen to fantasy sports as a mass market commodity once – not if, but when – something like this goes mainstream?…

I was planning to write up some thoughts about the future of autodraft technology and its potential implications on the game. I’ll offer some more thoughts about this in near future, if not next week.

Dylan B
12 years ago

Derek,

Yeah, I agree with you on the statistical need, you typically do see the auto teams with higher “power” numbers as well as high counting stats for pitchers(W and K’s). And I do agree that some people are better off with an auto drafted team, than drafting themselves; but I beleive right now I can draft better than both of them. And while the auto draft might give them a stronger team, it also makes their picks more perdictable; which “should” make my draft even stronger.

As for the creation of program that can do what you, isn’t that esentially the cheat sheet that you create before each draft wink .
I don’t think we are that far off; I had a free trial of Bloomberg Sports for baseball last year, and it seemed close to doing that. It did have some issues with playing time and min inning requirments(especially with negative stat cats), but I think in a few years that or something like it will make its way into the mainstream.

Dale
12 years ago

I see no problems with autodrafters.  Personally, I think you should beat an autodrafted team easily if you stay as active as the other manager does.  While you’re in the draft selecting the players that will perform, fighting all the other savvy managers for those players, this team will be selecting the recently injured, the players coming off of a career year, the players who have possibly fallen off the cliff, and the perenially overranked.  Their squad will have plenty of question marks in the points where the draft matters most.

I really can’t come up with a situation like you’re describing because the provider’s rankings are the defacto standard against which you’re drafting on (although if you’re in a good league you are tempering those rankings with ADP rankings and your gut instincts).  If you think someone is going to overperform by two rounds, be sure to reach at least one round.  If you think he’ll overperform by four rounds, be sure to reach at least two rounds.  The same thing that lets you beat the other owners is going to let you beat the computer. Planning on owners to underdraft seems foolish.  Crying about it afterwards, doubly so.

I’ve done “perfect draft” reconstructions, and you’re aided in getting really great teams because there’s always 1/4 of the “perfect team” that goes undrafted. Clearly they weren’t ranked highly enough to be autodrafted.  Typically you’re moving players that were actually drafted in the 20th rounds into the 12th rounds or so, because the non-obvious picks are so plentiful.  So worrying about a player that’s drafted where he’s projected to draft (and this isn’t the case, as your leaguemates are bound to be reaching, forcing the autodrafter to play catch-up) isn’t productive.

Derek Ambrosino
12 years ago

Thanks for the talk, guys.

As I imply in the article, I think on of the aspects heavily in play amid my feelings here is my values in regard to what it is to be an engaged participant in a fantasy sports league. My temptation to punish people who don’t participate in the draft is less to increase my advantage and more to disincentivize turning what’s likely the single most important aspect of building your team over to a machine. I included the anecdotal example mainly to prove it does exist, but I agree with you all that I should beat those teams regularly, barring catastrophic injury.

Autodraft also bugs me because it gives autodrafters a built-in excuse for mediocrity – well, I didn’t even get to draft my own team. As if that’s the problem… and to a degree it may be, but that excuse ignores the fact that by not prioritizing the drafting of your own team, you’re already not displaying a behavior pretty highly correlated with, you know, actually being good at this thing we call fantasy baseball.

Rick
12 years ago

The problem appears to be one of perception.  Having a player that you want taken by an autodraft is in your face to the point where it appears to happen more often that it does.  However, having a player that you want NOT taken by the autodraft appears as a non-event.  Fractions that complain about the autodraft overlook the selections that they were allowed to make that may have not been available.  I agree with the statement above, that autodrafting “is like drafting against a player who’s showing you his hand.”  Given the default list and the information that the team ahead of you is autodrafting, you can look ahead and have a better idea of who may be available (or in this case not available) with your next pick.  If all the teams in the league are doing this, it will penalize the autodrafter more than they deserve.

Derek Ambrosino
12 years ago

Rick,

I do agree. There’s definitely the air of the bad beat vs lucky win cognitive dissonance at play. …But, given the fact that is opinion is rooted in my sense of values regarding fantasy play and participation, equity isn’t my primary concern. Yes, that person is being “punished” more than I am, that’s certain. My question is whether he should be punished even more. I place “punished” in quotes the first time around because there’s nothing really punitive happening – he’s getting the courtesy of a semi-competent robot most likely doing a decent job drafting his team when he has most likely chosen to put an iota of effort into the process. That seems like as much of a reward as a punishment. …So, it’s less about what’s most advantageous for me, and more about what is sufficiently punitive for him.

This is also another issue of perception – my perception of my leaguemates giving a crud. I don’t care if you are just going to take whoever the highest pre-ranked player is at every single one of your picks. Show that you care by showing up and doing it person. When you autodraft, you create the perception of apathy, so I respond in kind with increased apathy in regard to “fairness” as it relates to you.

WilsonC
12 years ago

The way I look at it is that the autodraft tool, without a modified pre-rank list, is like drafting against a player who’s showing you his hand.  The picks tend to be reasonable for the most part, which is important if you want any chance of that owner being active during the season, but they’re also predictable.  You may not think a player would value a particular player as highly as his pre-rank, but you should know exactly how the autodraft tool values that player, and can either plan accordingly and take him a round before, or you can let that player have him and plan to work out a trade; if you’re right in his valuation of the player, the owner might be looking to move him.  Sure, you sometimes miscalculate, but if I miss a player you like when I can reasonably guess that he’s creeping up on someone’s list (as you can with an autodraft team), I chalk it up to my own error.

On the opposite end of the spectrum, with autodraft teams, you can get a good idea of some of the lower pre-ranked players who aren’t likely to be on the autodrafter’s list.

Jim Detry
12 years ago

If you are the commissioner, you can control who you invite to participate. Since the draft is half the fun of a Fantasy league, I find managers make every effort to attend the draft. I notice a larger problem with managers who give up at some point in the season. Those guys are simply not invited back the following year. You should let people know they are expected to make an effort and those who don’t will be out next year. It won’t take long before you have a league of solid, dedicated managers.

If you can’t beat autodrafters, you shouldn’t be allowed to write a fantasy column. It is easy to do a Yahoo mock draft against a bunch of autodrafters and see how your team stacks up at the end. It becomes much more interesting to do an auction draft, as the autodrafters may end up following an assortment of strategies. As a matter of fact, an auction draft is, in general, a lot more fun than a snake draft as, if you really want a player, you can almost always get him.