Auction Draft Data

[Matthias]

One thing that I've been wondering since my first auction league a few years ago, and I'm sure others have wondered the same thing is: how should I spread my money? Should I sink 80% of it into 2 or 3 guys and then fill out my roster with cheapies at the end? Or should I spend 75% of my budget on 5-8 guys giving me some balance and lowering my risk in case one guy gets injured? Well, this morning while messing around I thought of a way to approximate an answer in one of my auction leagues (spreading the money around was better) but it was only 1 year's worth of data and had the confounding variable of keepers. So.... I'm wondering if anyone wants to help me build a better dataset. To wit, I'm looking for: data on how much each player was drafted for; who drafted them; and how many points each team scored at the end of the year (active roster only). Basically, the results of your auction as well as the results of your league. To keep this data consistent, I'm only looking for 17-player redraft leagues.

If you guys wanna help me out with this, you can send your data (in .xls format preferred) to my new gmail address: ScrubNoScrub@gmail.com. And I'll share whatever results I get with you guys.

Thx.

[CBMGreatOne]

My league just started auctioning this year, so I can't help you with the data, but I think as a general rule there are some specific strategies that you can apply based on league size.

I think in a 10 team league, there is much more incentive to go stars and scrubs. Go ahead and spend 58 on Ray Rice, 38 on Brees, 35 on Reggie Wayne and 30 on Roddy White. Then fill out your roster with the 49 dollars you have left. Less players will be drafted which means more guys ready to be plucked after their big week 1-4 performances. Guys like Bernard Scott, Leon Washington, Mike Bell, Javon Ringer, and Kareem Huggins are just one injury away from being weekly top 25 starters.

In a 12 team league or deeper, most of the above players are going to be drafted and you're going to be grasping at straws to fill your last few roster positions if you invest too heavily in your top picks. In these formats, I like to devalue stud RBs to the point where I don't spend more than 35ish on any one guy. You can still get 2 or 3 guys of the Ryan Mathews, Beanie Wells, Pierre Thomas, Jamaal Charles ilk and fill out with sub 10 dollar guys like Ahmad Bradshaw, Leon Washington, Donald Brown, Fred Jackson etc. Not investing $50+ in your RB1 should free up some cash to land a top QB and at least one top WR. Maybe Romo for 25ish and Wayne for 35?

There are plenty of 1-5 dollar guys that should at least be startable at the WR position in either case--- Jacoby Jones, Bernard Berrian, Nate Burleson, Legedu Naane, and Jericho Cotchery to name a few.

My league is 12 teams, so I'm gonna try to spread it out. Here's hoping the top guys go for inflated prices.

[Matthias]

Yah; my approach has bounced around over the years. I think in fantasy football, generally speaking, studs and scrubs are a viable approach. You can have 50-70% of your points come from 2 guys in any given week. And it's difficult to find a trade partner for your good flex option of Felix Jones when someone feels like they can just pick off the waiver a WR with a good matchup that week for the cost of their least valuable player.

However, in one of my 12-team auctions yesterday I felt I did well by avoiding the early runups and landing: Best, Greene, Spiller & Jackson, Jonathan Stewart, Calvin Johnson, Roddy White, Jermichael Finley, and Eli Manning. And I like data. So I thought I'd take a look at it and see if we can come to any systematic conclusions and not just estimate.

[Goozik]

I would have to say the best way will change a lot from year to year and scoring system to roster set-up.

So getting data from one set is def the way to go on this.

Minor fluctuations really trickle down the Auction draft board though, so not sure any one preset strategy is always possible or the most efficient. But outlining a few that have shown success should help.

Good luck.