RotoWire Partners

Nerd Alert: Advanced Stats and Fantasy Hoops

Carson Cistulli

Carson Cistulli

Carson Cistulli writes about fantasy sports for RotoWire.


It's customary, before any fantasy season begins, for an owner to target a couple of players whom he regards as sleeper picks -- nor am I revealing any big secrets by saying so.

For me, one such player -- probably the one at the top of my list -- was Portland's Nicolas Batum. Besides the fact that I thought he'd be available in later rounds, what I liked about Batum was his ability to contribute in almost every category. To my mind, this is a good sort of player for providing a "foundation" to a fantasy team. While someone like Dwight Howard -- that is, one who produces prolifically in one cat while actively hurting owners in another -- while someone like Howard requires extra consideration so far as roster construction goes ("Who do I pair with Howard so I don't get killed in free-throw percentage?"), a player like Batum requires basically zero consideration of roster construction.

A couple weeks ago, I gave the profile of an "average" player (Avg) in a standard 12-team, 13-player league. Here's Batum's RotoWire projection (Pro) relative to those averages I outlined:

Proj: 13.6 PTS, 2.0 3Pt, 5.1 REB, 1.6 AST, 0.9 STL, 0.9 BLK, 51.8 FG%, 83.9 FT%

-Avg: 14.5 PTS, 0.9 3Pt, 5.5 REB, 3.2 AST, 1.0 STL, 0.7 BLK, 46.6 FG%, 78.3 FT%

As you can see, Batum stacks up favorably in basically every categoy but assists -- and, owing to the large standard deviation in that category (because point guards get a bunch of them, most other players get few), Batum's deficit is minimalized.

As it turns out the RotoWire projection for Batum was slightly optimistic. Still, he's produced very similarly to average this season. Regard:

Actual: 12.1 PTS, 1.5 3Pt, 4.8 REB, 1.4 AST, 0.8 STL, 0.6 BLK, 45.0 FG%, 81.9 FT%

Considering all this led me to ask a question: What sort of player ought one to prefer -- those who contribute evenly across all cats (the Even Stevens) or those who put up big numbers in one cat to the detriment of others (the Diverse Dans)? Of course, the question to ask alongside this is: Which players are, in fact, Evens Stevens and which are Diverse Dans?

To look at this, I found z-scores (i.e. standard deviations from the mean) in the eight main fantasy categories for the top-156 players in the league (based off the 12 teams and 13 players each in standard fantasy leagues). Averaging together the z-scores, one can find a ranking of those 156 players where Kevin Durant (average of +1.00 standard deviation) is first overall, Chris Paul (+0.93) is second, LeBron James (+0.81) is third, and so on -- all the way down to Louis Williams (-0.46) in the 156th spot.

Once I did this, I found the absolute value of the difference for every player between each of the eight cats and his overall score. So, for Kevin Durant, I took his point-scoring value (+2.97 standard deviations from the mean) and subtracted from it the +1.00 that is his overall score, getting 1.97. Then I took his three-point value (+1.28) and subtracted +1.00, getting 0.28. Then I took rebounds, assists, steals, etc., and did the same thing. Then I added all the absolute differences to come up with overall variation (Var). For Durant, the total is 7.29.

I went through that process for all the top-156 players. Below are the results, paired with some scattered commentary.

First, the Even Stevens:




Now, the Diverse Dans:




Now, five scattered thoughts:

On average, the most extreme of the Diverse Dans are better fantasy contributors than the most even of the Stevens. The average rank for the Stevens is 111.6; for the Dans, 56.4. What does this tell us? Well, probably that a couple standout tools are necessary for excellence in fantasy basketball. For Durant, that's scoring and free-throw shooting. For Paul, it's assist and steals. For LeBron, it's points and assists.

As one might expect, the Even Stevens are generally players of the wing variety. None of the Stevens are point guards or what one might think of as a "true" center. As one might also expect, the Diverse Dans are generally players on the extremes of the positional spectrum. There are three point guards on this list and four or five true centers (depending on how you feel about Emeka Okafor). In seven of the ten cases (everyone but Paul, Westbrook, and Martin), the huge deficit comes from free-throw shooting.

The first above-average Steven is Andrei Kirilenko, who's 15th-most even and ranked 53rd overall with a +0.07.

Boris Diaw is probably the most average of the the average players in fantasy baskeball. Here are his numbers:

Diaw: 11.4 PTS, 1.0 3Pt, 5.5 REB, 3.8 AST, 1.0 STL, 0.7 BLK, 49.5 FG%, 66.2 FT%

-Avg: 14.5 PTS, 0.9 3Pt, 5.5 REB, 3.2 AST, 1.0 STL, 0.7 BLK, 46.6 FG%, 78.3 FT%

It's actually curious to imagine Diaw as anything like "average," as he possesses one of the more unique skill sets in all the NBA.

Batum really does contribute pretty evenly across cats relative to the rest of the league.