StrategyAugust 2, 2009

Post to Twitter

Tanking: The Difficult Categories

By Aleksandar Jovanovic

Tanking cats is one of the most popular strategies in fantasy leagues, especially in head-to-head leagues. You have to keep in mind though, while all cats are equal in the sense that they all carry same number of potential wins (H2H) or points (roto), not all of them are tanked as easily.

Regardless of which cat(s) you decide to tank, or punt, the key is to do it as efficiently as possible – have players who produce the minimum of their value in the cats that you decided to tank. I’ve divided the nine standard cats into three groups: difficult to tank (points, rebounds and steals), popular to tank (free throw percentage, field goal percentage and turnovers) and easy to tank (assists, blocks and threes). I’ll be taking a little closer look at every cat and what it means to tank it. In this article, I’ll go over the three difficult to tank cats and in the following two articles I’ll cover the other two groups.

All the research I present in these articles is done using a sample of the 200 best players from the 2008-09 season (per Basketball Monster). My goal is not to go into details too much, but rather to outline important concepts, which is why I chose an arbitrary number of players that should include most fantasy relevant players for standard leagues.

The goal of the basketball in real life is to score points. As a result, almost everybody scores, especially the best players, who provide most of the other contributions as well. This is the main reason why it is not easy to tank points. Among the top 200, the average is 13.8 points per game (PPG) with Wade on top of the league with 30.2 PPG. This means that an average player contributes 45.7% of what the league leader contributes in this category. Same analysis for blocks shows that average player contributes just 21.9% of the league leader’s contribution. So average player contributes much more points than he does blocks, relative to what the elite scorers contribute.

In order to get an idea about how many players are “average, or near-to-average” in PPG, I decided to count everybody who scored within 33% of average player. Again, it is an arbitrary number which will bother some people, but as I already explained this is not hard science, but rather a simple statistical analysis. Thus, an arbitrarily chosen number will do. So if all players who scored between 9.2 (67%) and 18.4 (133%) PPG are near to average, we get 39 players significantly above average, 121 near average and 40 significantly below average. Out of those 40 players significantly below average, only one is ranked in top 50 overall, three more are in the top 100 and nine more are in top 150. In the group of 39 players significantly above average, there are eight of the top nine overall, nineteen more of the top 50, nine more of the top 100 and just two outside of the top 100.

So we can conclude two things that makes tanking points more trouble than it’s worth. One, out of the top 200 players, 160 owe at least part of their value to points. So drafting those players while trying to tank points wastes part of their value (which does not make for efficient tanking). Two, very few major fantasy contributors don’t help significantly in PPG (just one of the top 50) so drafting in the top rounds can become a nightmare with this strategy if you are unable to trade down repeatedly.

The vast majority of big men (and quite a few top rated guards and swingmen) contribute in this category. Dwight Howard led the league with 13.8 rebounds per game (RPG), while the average player from our set grabbed 5.5 RPG, which is 39.6% of the league leader’s total. If we use a similar method to find out how many near-to-average players there are, we get 59 players significantly below average players, 94 near average (between 3.7 and 7.3 RPG) and 47 significantly above average. Out of the 59 players significantly below average, there is just one PF/C eligible player (ranked 189th) and 12 G/F eligible (only two of which are in the top 100). Furthermore, there are no top fifteen players, regardless of position, and only one averaged one or more blocks per game. On the other hand, five of the top ten players are in the significantly above average group.

To sum up, there are three points worth noting. One, standard settings demand that you start two centers and one power forward, which is a third of your roster. It is impossible to find three quality players on those positions who will produce in other categories but not in rebounds, which means that your tanking is not gonna be efficient. Two, very few top players don’t rebound so depending on the pick you get in the first round it might be impossible to pass on a significant contributor in this area. Three, tanking rebounds automatically means you are tanking blocks, which of course is a problem if you were planning on tanking just one cat, or even two because you have no flexibility in choosing your second cat to tank.

Steals are mainly provided by guards, especially the elite ones, but quite a few big man average around a steal a game. CP3 dominated the league with 2.8 steals per game (SPG), while average for the top 200 players was 0.95 SPG, which is 33.9% of the league leader. That number is a little deceiving because, as I said, Paul dominated. If we use average of the runner-up in steals (Wade), we get 43.2%. Per our method, there are 51 players significantly below average players, 112 near average players (between 0.64 and 1.26 SPG) and 37 significantly above average.

As you can see, those numbers are very similar to the numbers we had when we were talking about points (149 near average, or significantly above it) and a similar conclusion stands: a significant number of players owe at least part of their value to steals, so drafting those players while trying to tank steals wastes part of their value, which is not efficient tanking. To be fair, there are more good players in the significantly below average group then there are for points and it is easier to fill out all roster positions while tanking steals then it is with rebounds. Therefore, it is easier to tank steals then either points or rebounds, but there are still too many players that contribute in steals to count it among easily tanked cats.

If you have a choice (and you almost always do), you are better off choosing other cats for tanking than points, rebounds or steals. In upcoming segments, I will analyze the popular to tank categories and the easy to tank categories.

Aleksandar Jovanovic is one of a growing number of fantasy experts who write for the Cafe. You can catch up with Aleksandar in the Cafe's forums where he posts under the name of KalElen.
Rate this article: DreadfulNot goodFairGoodVery good (10 votes, average: 3.20 out of 5)
Loading ... Loading ...

Questions or comments for Aleksandar? Post them in the Cafe Forums!

Want to write for the Cafe? Check out the Cafe's Pencil & Paper section!

Post to Twitter

Related Cafe Articles

• Other articles by Aleksandar Jovanovic
Tanking II: Popular Cats by Aleksandar Jovanovic (posted on 09/19/2009 in Articles)
Opposing Tanking in H2H Drafts by ardilla (posted on 10/14/2009 in Articles)
Down, But Not Out by Aleksandar Jovanovic (posted on 12/15/2008 in Articles)