Whenever you’ve just started playing fantasy NBA or consider some offseason alternatives like fantasy all-time draft one of the first questions you need to ponder is: “how are we going to redistribute players between teams”? And one of the most common answers to this question is “Snake Draft”.
It’s a draft where the positions flip flop every round. For example, if someone drafts first during the very first round, they’ll draft last the following round, then they’ll draft first again in the next round, and so on. This type of draft is the most fair.
This concept is easy to grasp, it was always easy to organize and it just sounds so fair. Well, there’s only one problem with it: in fantasy NBA snake draft is unfair! And it’s not even close to being fair!
How do I know this? Not only it’s essentially the same problem as with competitive balance in the NBA but more importantly we can easily calculate why and how is it unfair.
I’ll use Yahoo’s Default Points Scoring [FGA (-0.45), FGM (1.0), FTA (-0.75), FTM (1.0), 3-pt Made (3.0), Point Scored (0.5), Rebound (1.5), Assist (2.0), Steal (3.0), Turnover (-2.0), Blocked Shot (3.0)] as an example but it really doesn’t matter, you can repeat this exercise for every possible settings with the same results.
| Distribution of picks in a standard 10-team snake draft and a corresponding Total Points Average | Data from 2010/11 NBA Season | Data from 2009/10 NBA Season | Data from 2008/09 NBA Season |
| #1 + #20 + #21 + #40 + #41 + #60 + #61 + #80 + #81 + #100 | 295,81 | 306,18 | 306,93 |
| 42,7 + 33 + 33 + 29,9 + 29,9 + 27,1 + 27 + 25,2 + 25,1 + 23 | 295,81 | 306,18 | 306,93 |
| #2 + #19 + #22 + #39 + #42 + #59 + #62 + #79 + #82 + #99 | 294,68 | 299,10 | 307,35 |
| 41,3 + 33,1 + 32,9 + 30 + 29,8 + 27,3 + 27 + 25,2 + 25,1 + 23,2 | 294,68 | 299,10 | 307,35 |
| #3 + #18 + #23 + #38 + #43 + #58 + #63 + #78 + #83 + #98 | 293,11 | 295,69 | 304,10 |
| 39,8 + 33,3 + 32,8 + 30 + 29,4 + 27,3 + 26,9 + 25,2 + 25 + 23,3 | 293,11 | 295,69 | 304,10 |
| #4 + #17 + #24 + #37 + #44 + #57 + #64 + #77 + #84 + #97 | 293,01 | 294,96 | 298,17 |
| 39,5 + 33,4 + 32,7 + 30,1 + 29,3 + 27,5 + 26,9 + 25,3 + 25 + 23,3 | 293,01 | 294,96 | 298,17 |
| #5 + #16 + #25 + #36 + #45 + #56 + #65 + #76 + #85 + #96 | 293,42 | 294,56 | 297,72 |
| 39,1 + 33,8 + 32,6 + 30,2 + 29,2 + 28 + 26,7 + 25,4 + 24,9 + 23,5 | 293,42 | 294,56 | 297,72 |
| #6 + #15 + #26 + #35 + #46 + #55 + #66 + #75 + #86 + #95 | 291,75 | 293,91 | 296,66 |
| 37,6 + 33,9 + 32,4 + 30,2 + 28,9 + 28,1 + 26,6 + 25,7 + 24,8 + 23,5 | 291,75 | 293,91 | 296,66 |
| #7 + #14 + #27 + #34 + #47 + #54 + #67 + #74 + #87 + #94 | 291,97 | 292,52 | 296,94 |
| 37,3 + 34,7 + 32,3 + 30,4 + 28,7 + 28,2 + 26,5 + 25,8 + 24,6 + 23,6 | 291,97 | 292,52 | 296,94 |
| #8 + #13 + #28 + #33 + #48 + #53 + #68 + #73 + #88 + #93 | 292,40 | 292,41 | 296,92 |
| 36,5 + 34,7 + 32,3 + 31,6 + 28,6 + 28,2 + 26,3 + 26 + 24,6 + 23,7 | 292,40 | 292,41 | 296,92 |
| #9 + #12 + #29 + #32 + #49 + #52 + #69 + #72 + #89 + #92 | 291,69 | 292,42 | 296,67 |
| 35,9 + 34,8 + 32 + 31,6 + 28,5 + 28,4 + 26,3 + 26 + 24,5 + 23,7 | 291,69 | 292,42 | 296,67 |
| #10 + #11 + #30 + #31 + #50 + #51 + #70 + #71 + #90 + #91 | 291,77 | 292,61 | 295,72 |
| 35,7 + 35,1 + 31,9 + 31,9 + 28,4 + 28,4 + 26,3 + 26,1 + 24,2 + 23,8 | 291,77 | 292,61 | 295,72 |
You may scoff here and say “all that fuss about a couple of fantasy points for the whole team?”.
First of all it’s a difference between Top100 player and Top70 player.
In other words GMs with top picks get an additional solid player for free.
On a top of that players at the top are easier to predict than guys at the end of first round.
Finally, it’s the tip of an iceberg because the deeper the league is the bigger the problem becomes!
| Distribution of picks via standard snake draft in a 16-team league with 12 players per roster and a corresponding Total Points Average for Top10 in each team |
Data from 2010/11 NBA Season |
Data from 2009/10 NBA Season |
Data from 2008/09 NBA Season |
| #1 #32 #33 #64 #65 #96 #97 #128 #129 #160 #161 #192 | 299,74 | 308,53 | 315,64 |
| #2 #31 #34 #63 #66 #95 #98 #127 #130 #159 #162 #191 | 297,15 | 297,27 | 302,45 |
| #3 #30 #35 #62 #67 #94 #99 #126 #131 #158 #163 #190 | 295,19 | 297,43 | 301,54 |
| #4 #29 #36 #61 #68 #93 #100 #125 #132 #157 #164 #189 | 294,67 | 297,49 | 301,44 |
| #5 #28 #37 #60 #69 #92 #101 #124 #133 #156 #165 #188 | 294,57 | 297,25 | 300,76 |
| #6 #27 #38 #59 #70 #91 #102 #123 #134 #155 #166 #187 | 293,42 | 297,98 | 300,40 |
| #7 #26 #39 #58 #71 #90 #103 #122 #135 #154 #167 #186 | 293,38 | 297,22 | 300,56 |
| #8 #25 #40 #57 #72 #89 #104 #121 #136 #153 #168 #185 | 293,29 | 297,81 | 300,22 |
| #9 #24 #41 #56 #73 #88 #105 #120 #137 #152 #169 #184 | 293,46 | 303,23 | 315,87 |
| #10 #23 #42 #55 #74 #87 #106 #119 #138 #151 #170 #183 | 293,18 | 300,20 | 313,23 |
| #11 #22 #43 #54 #75 #86 #107 #118 #139 #150 #171 #182 | 292,08 | 300,29 | 307,62 |
| #12 #21 #44 #53 #76 #85 #108 #117 #140 #149 #172 #181 | 291,57 | 299,84 | 306,31 |
| #13 #20 #45 #52 #77 #84 #109 #116 #141 #148 #173 #180 | 291,79 | 299,27 | 304,96 |
| #14 #19 #46 #51 #78 #83 #110 #115 #142 #147 #174 #179 | 291,85 | 297,62 | 304,71 |
| #15 #18 #47 #50 #79 #82 #111 #114 #143 #146 #175 #178 | 291,12 | 297,73 | 303,86 |
| #16 #17 #48 #49 #80 #81 #112 #113 #144 #145 #176 #177 | 290,87 | 297,64 | 303,59 |
Do you see the problem now? In a semi-deep league with a shallow benches the difference between top and bottom picks is over 10 fantasy points! Now we are talking about some serious damage here… because even if you pick a player who will out-perform his draft position by 10 fantasy points [for example, 10th best player at #75!]… you will just be on the same level as GMs with top picks!
Let me put this this way: in a competitive environment a snake draft essentially means that your final position in the standings will depend mostly on where you picked in a draft!
Not your skills, not your knowledge but pure dumb luck in a draft order lottery before season even started!
Obviously, there will be some exceptions because it’s not like you can’t pick better than other GMs in your league but it doesn’t change a general rule and by definition all GMs can’t improve their position…
BTW, it also applies to ratings for head-to-head and Rotisserie!
What you can do to prevent this injustice?
1) Try auction draft.
In short, every GM starts a draft with the same amount of virtual money and spend as they like on players until their roster is full or their account is empty. You can read about basic rules for example here:
http://games.espn.go.com/fba/content?page=fbarulesauctionoverview2010
You can also test it for free anytime in a mock auction [so results don’t mean anything]:
http://basketball.fantasysports.yahoo.com/nba/mock_lobby?lobby=auction
Although keep in mind that it’s better to test it with your friends than with strangers who usually leave early and don’t care about anything. But a technical side works just fine in any case.
Seriously, auction is more fun than snake draft and more flexible so you have more control over your team, there are more strategies involved etc. Also thanks to sites like Yahoo, ESPN and others it’s easy to organize. I do not know a single person who tried it and said “nah, thanks, I’ll go back to more exciting snake draft”.
2) If you want to stick with a snake draft force GMs with top few selections to forfeit one of their picks in the middle round.
As I’ve shown above you can easily calculate which pick should it be in your league.
Mostly something around #70 and #80 but this detail really depends on all your settings.
Important note! Changing order in one of the rounds won’t be enough to make it fair! That’s because in the middle rounds differences between players are very small so even if you propose an order where in every round outside of first it goes from last to first it won’t be enough!
Are YOU convinced?
If not… it’s your loss! Well, unless you were lucky to win a draft order lottery… ;-)
Weak Side Awareness 
karl
December 1, 2011 at 05:20
I don’t want to be moaning but I will try to give you some feedback on this one. It is too late to argue with your arguments so I will just focus on appearance. Two things:
First, you have mentioned the auction draft as a better proposal for a fantasy draft. Ok, that’s fine but please give us some explanation of what it is. At least some links. You have done it well with the “snake draft” so why you changed it? :)
The other one is (I know, it might be too challenging but still) that you could give us exact calculations for each team’s total fantasy points. How did players combine each number? If you have an excel file, perhaps you could upload it? It would be always easier to argue with you :)
Anyway, good to see you back!
wiLQ
December 1, 2011 at 14:23
“It is too late to argue with your arguments”
Wait, what? It’s never too late for it and that’s the main point of comments!
I encourage you to argue with my arguments if in your opinion they are wrong.
Hell, if you can prove that any of my posts [even old ones] didn’t have any sense, please do it ;-)
“First, you have mentioned the auction draft as a better proposal for a fantasy draft. Ok, that’s fine but please give us some explanation of what it is. At least some links. ”
Oh, I’m sorry, I thought that part was obvious but you are right.
You can read about basic rules for example here:
http://games.espn.go.com/fba/content?page=fbarulesauctionoverview2010
You can also test it for free anytime in a mock auction [so results don’t mean anything]:
http://basketball.fantasysports.yahoo.com/nba/mock_lobby?lobby=auction
Although keep in mind that it’s better to test it with your friends than with strangers who usually leave early and don’t care about anything. But a technical side works just fine in any case.
“The other one is (I know, it might be too challenging but still) that you could give us exact calculations for each team’s total fantasy points. How did players combine each number?”
Sure, I’ll do that [probably in the evening].
“Anyway, good to see you back!”
Thanks, and thank you for your feedback!
Leszczur
December 1, 2011 at 12:26
I absolutely agree that snake draft will put you in a competitive disadvatage in sigular value scoring systems if you are unlucky enough not to land the top pick. It’s been shown numerous times in many leagues we played in, and you show numbers that support it very clearly.
Although I wonder what made you draw similar conclusion for rotisserie (H2H I won’t talk about since I’m no expert by any means on that subject)?
In rotisserie – singular value for players has much lesser meaning since the competition revolves around 8 or 9 separate statistical categories. So how can you estimate impact as “bigger” than in fantasy points since you do not really calculate anything here?
If you put gun to my head – I would say that in rotisserie it’s easier to erase some of that competitive advantage of Top5 pick by applying strategy in creation of your roster. At lease there is a chance to do something while in fanasy points – there’s nothing you can do (short of flying to US and applying a baseball bat to somebody’s knee cap).
wiLQ
December 1, 2011 at 14:46
“Although I wonder what made you draw similar conclusion for rotisserie”
Because the difference between top picks and picks at the end of first round exists in a same way for pretty much every possible reasonable rating system.
“So how can you estimate impact as “bigger” than in fantasy points since you do not really calculate anything here?”
You are right, I’ll remove this comment. I added it too casually, I didn’t want to step on my future posts too much and nuances of rotisserie is a totally different issue than unfairness of snake draft.
“If you put gun to my head – I would say that in rotisserie it’s easier to erase some of that competitive advantage of Top5 pick by applying strategy in creation of your roster. ”
Easier to erase via strategy – I would agree if only because there is more strategy possible in rotisserie than in fantasy points – but competitive advantage of Top picks still exists.
Matt Buser
December 5, 2011 at 18:47
Interesting theory, let me know if you are able to verify with actual results. Contacting me through my feedback – I had saved your original email but somehow it got lost in the shuffle – is probably the best way to go.
wiLQ
December 5, 2011 at 21:11
Sure, I’ll let you know via twitter and thanks again for pointing out that source at Yahoo.
With so much information in one place I’m pretty sure I’ll check many more things ;-)