I have been observing the weekly drafts here at Draftmix for the past couple months, and I’ve noticed a few interesting things. I am going to go over a few strategy considerations, as well as examine some facts about our drafts in particular using aggregate data culled from leagues we’ve run over the last month or so.
There are quite a few differences between the weekly format and the full season format. The biggest two are match-ups, and roster size.
Match-ups are who each athlete is playing against during the coming week. This is probably the most important consideration in a weekly draft. There are some athletes who are going to be near the top in expected value for fantasy points every week, but once you get past those 5 or so there is a large pool of fairly similar players whose value fluctuates greatly depending on their matchup. This is going to be where you win or lose your draft.
The roster size also has a huge effect on the strategy. In full season leagues where people have a bench you won’t be able to get the 20th best running back in the last round, but when everyone can only draft two running backs it becomes very likely. Also, Draftmix teams require three wide receivers, forcing players to draft deep down the list. This adds some strategic considerations by making wide receivers, as well as running backs, and quarterbacks, valid early picks.
Out of curiosity, I wrote a few scripts to examine some of the data from leagues we’ve run and discovered a few interesting things about our format. The first script shows which draft position won the contest (10 or 12 player drafts since we launched our new site only). Most of these drafts were 10 man drafts as we only recently started doing 12 mans.
1st pick: 0 wins
2nd pick: 1 win
3rd pick: 0 wins
4th pick: 1 win
5th pick: 0 wins
6th pick: 3 wins
7th pick: 2 wins
8th pick: 2 wins
9th pick: 4 wins
10th pick: 5 wins
11th pick: 0 wins
12th pick: 2 wins
This is quite unexpected. Nobody who drafted first won, and as you can see a clear trend towards the later players winning. Granted this data comes from only 20 drafts, but is still surprising. I am looking forward to seeing what happens after 40 drafts or 100.
The second script shows who the first pick of the draft was for each winning team. I ordered them by position.
Quarterbacks:
Ben Roethlisberger (x2)
Peyton Manning
Tony Romo
Running Backs:
Joseph Addai (x3)
LaDainian Tomlinson (x2)
Steven Jackson (x2)
Adrian Peterson
Marshawn Lynch
Brian Westbrook
Wide Receivers:
Terrell Owens (x3)
Randy Moss
Braylon Edwards
Tight Ends:
Antonio Gates
Another very interesting output. First there are 4 different positions represented, the most unusual being the one Antonio Gates pick. The other thing I found interesting is that no one who has drafted Tom Brady has won a draft for some reason, despite his consistently excellent performance.
Most of our drafts have been 10 or 12 man drafts, but here are the winning first picks from the two 8 man ones we ran during the period.
Larry Fitzgerald (chosen by the 5th player)
Tom Brady (chosen by the 1st player)
In conclusion the weekly draft format is a very interesting take on fantasy sports. There are a lot of different strategies, and the variety in player value is pretty well balanced.
What do you think?