Fantasy Basketball Decision Making Framework

I’ve been playing fantasy basketball on and off for about 10 years. I really like it because I’m a big fan of the NBA and its very analytical. I’ve decided to share it as I’ll get feedback from other people that will improve my decision making in fantasy basketball.

Fantasy sports is huge, particularly in the US. Some of the fantasy sports games like Draft Kings actually sponsor the real life game. I spend several hours a week playing fantasy basketball and probably an hour a week watching basketball condensed games on league pass.

Its important that I outline what is fantasy basketball and how my particular league works as this dictates my decision making framework.

The gist

In Fantasy basketball, you draft (choose) basketball players to join your team. You compete against other teams on a statistical basis.

It starts with you joining a league with other team managers. A league can be private or public league. A public league is open to anyone to join, whereas a private league is invite only. The leagues I typically join have about 10 – 14 managers. The more managers there are, the deeper you have to go into the draft to find players. If there’s like 6 managers, then you’ll be able to get more starters and superstar players on your team.

In my league, I have to draft for these positions: PG, SG, G, SF, F, PF, C, C, Utility, Utility, and 3 bench players. To provide some colour, PG = point guard, SF = Small Forward, C = Centre. Bench players are not active during the week.

I also play in Yahoo’s fantasy basketball league.

H2H v Rotissere league

My league is a H2H league which means you compete head 2 head against other players. So each week you rotate against a new team. You only play that team for the week.

In a rotissere league, you compete against all players each week. Its based on an accumulation of total stats and its ongoing throughout the season.

The type of league and which categories (“cats”) are selected for the league will dictate your strategy. My league has the following cats:

3PTM (3 pointers made)
PTS (Points)
REB (Rebounds)
AST (Assists)
STL (Steals)
BLK (Blocks)
TO (Turnovers)

The draft

To balance the drafting process, we have a snake draft. This involves a random allocation as to the drafting order. In our league we have 12 managers. I was given #4 spot which means I draft 4th. When everyone has chosen from 1 – 12, the draft then goes in reverse order. So the person with the 12th spot gets the next pick which is the 13th pick. The idea is that it snakes around. This ensures some fairness in the process as you don’t get penalised for drafting last. Since I had the 4th pick, my next pick was #20.

There’s also a dynasty league, where you’ll keep those players from season to season. I have never played dynasty, only in snake draft leagues.

My strategy – efficiency

To recap – I’m playing Yahoo fantasy basketball head to head, 12 managers, in the 9 cats above.

My strategy initially was to draft the best players possible in my first 2 picks, then formulate a strategy from there. My strategy has evolved over time during the course of the league.

An epiphany hit me a few years ago in that in a H2H league with 9 cats, you only need to win 5 out of the 9 cats. If I won 5 and the other person won 4, I would win. So I started changing my strategy to focus on specific categories.

In addition, I noticed that there were two categories that didn’t involve a lot of volume based stats. These were FG% and FT%. I could win even if I took less shots. I had to simply be more efficient. I also worked out that there were some categories that people generally didn’t care a lot about which were steals, blocks, and 3 pointers. It depends what league you are in and how sophisticated the players are.

My point is you need to focus on specific categories & optimise your team for that. Whether its points, steals, blocks, or FG%.

My decision making framework

I used my buddy Daniel Grzelak’s Fasketball website which ranks players using projections over a season. You can also sort by categories. So I used this to help me in the draft.

As with stocks, past success is no guarantee of future success. However, in fantasy basketball I believe there are trends. I do take into account past performance. For example, if for 5 seasons a player has been shooting 39% FG’s, its probably likely they are going to shoot in that range. They are not going to become a 50% FG shooter the next season. 39% is very very low and 50% is very very high. Its possibly, but its unlikely. So I do take that into account.

The cats that I am currently optimised for are …drumroll…..

1. FG% (minimum 45%)
2. FT% (minimum 70%)
3. Steals
4. 3 pointers

To a lesser degree:
5. Points
6. Blocks

So if a player shoots 60% FG and 60% FT, I won’t choose them. Even though 60% FG is incredible, its going to ruin my average FT%.

Something that I have taken into account is that if I have a very high average FG%, I might be able to take a player with slightly less if he’s higher in other areas. But currently in my free agent waiver wire pickups, I don’t accept less than my minimum criteria.

I chose those % as an arbitary measure. But my gut feel from playing fantasy and watching a lot of basketball for a long time are that those are good minimum % to win those particular cats. Anyone shooting less than 45% FG is not a reliable shooter in my opinion. Those minimums also give me a wide enough pool of players to choose from. If I had 50% FG and 85% FT, I would probably only have 4 players to choose from.

I do often choose players that are way over that min. %. For example I have Kevin Durant who is 88% FT shooter (which is 12th in the league). I do tend to get a lot of players in the 80% – 90% FT range.

What this also means is that I’ve opted to tank categories such as turnovers, rebounds, etc… As in I don’t care about those other categories as long as I win the ones I’m focused on. So I do regularly lose in a category such as turnovers.

My watch list

After you have drafted your players for your team you can still alter your team. You can trade players with other teams. You can also drop players from your team and pick up free agents i.e. available players that no one else has picked. In Yahoo, you can “watch” players and add them to a favourite list.

Yahoo also enables you to sort players from lowest to highest in one category. So what I have done is sorted players by FG%. Then I filtered it by manually by selecting players that matched my min. FG% and FT% criteria. This becomes my watch list aka short list.

Drafting a Dud or Dudley? 

Once I had my watch list, I sorted them again by specific categories I wanted to win. In turns out that my sorting has highlighted that Jared Dudley from the Washington Wizards has the highest steals average at 1.0. Its also a category I want to dominate.

In this example below, I sorted by steals.

my fantasy basketball watch list
So I sorted again by 3 pointers.

fantasy basketball 3pointers
When I sorted by 3 pointers, Jared Dudley also came out on top as he makes on average 1.5 of his 3 point shots per game.

So I investigated further – who is this unicorn?!

My next step is to find out if this is an anomaly or can he keep this up? I want players that are consistent. Even better if they are trending up. So I initially look at the pattern in the last 10 games or so.

fantasy basketball last 10 games
There are some in-consistencies between games. But on the whole he does fill up the stat sheet on a consistent basis with steals and 3 pointers. I’m looking for wide variances in FG% & FT%  too. I also read the notes on the right hand side which Yahoo summarises. However this is not enough.

I want to look at his split stats over the past few months. Split stats tell you how he is doing month by month. Here we are looking at his performance in each month from October to January. What I am trying to ascertain is whether this is consistent pattern over a number of months, trending upwards or downwards.
split stats in fantasy basketballIn October, which is the first month in the league with a couple of games, the numbers are a little off. I disregard this as the sample size is too small and they’re warming up. I’ve now got a pattern of 40 games over 3 months. Is he trending upwards per month?

I also take into account seasonal averages. I will pull up his career averages. I want to know if these numbers are consistent over his whole career. I’m focusing on the 4 key areas which are FG%, FT%, Steals and 3 pointers. As these are the areas I want to improve with my next free agent pick up.

fantasy basketball career averagesFrom his career stats, his FG% and FT% are somewhat consistent with this season. As are his steals and 3 pointers. Given that he’s a veteran of 10 seasons, its likely that his stats will stay the same season.  He’s in a reserve role mostly and sometimes starts. So he’s not going to breakout and become a superstar this season. But that’s fine. I want him to continue doing what he’s been doing.

I do put a premium on proven players as they are more stable. I also like young players that are improving as the season progresses or over their careers. They have an opportunity to become even better.

I do stay away from rookies because there projections are unknown. I’ve probably missed out on a few awesome rookies like KAT (Karl Anthony Town) and Kristaps. I probably would have made an exception for KAT who’s a big man and is constantly getting double doubles in points and rebounds. I definitely stay away from rookie guards because their FG% is too wild and inconsistent.

My next step is to look at transaction trends and opportunities. Does this player have an opportunity to play? You need to factor in injuries and playing time, not only of the player in question but also their team mates. I know from following injury reports and general news that Bradley Beal who plays shooting guard has a lot of time off due to injuries. When Beal came back, he also had a minutes restriction which he publicly declared. I picked up Dudley before Beal was slated to come back. I also figured that Beal wouldn’t play as much as previously, hence providing more opportunities for Dudley.

In addition, I look at general trends across leagues. Are other managers picking up Dudley as well?

fantasy basketball transaction trendsHere I am trying to use some wisdom from the crowd. When a player is performing well, then there will be more adds then drops. The above means that 121 team managers have picked him up and 115 have dropped him. Its somewhat neutral. It also depends what your strategy is. In my case, he fits the criteria of what I am looking for and will enhance specific cats.

You can also go against the trend as well. If other people are dropping him, then maybe there’s a opportunity for a “buy low” scenario. This is when the analysis into prior seasons and past history becomes important. Perhaps he’s not playing well for the past few weeks and a rush of people are dropping him. But I might believe based on past performance he’ll revert back.

Opportunities – Injuries & games played

When a player is injured and can’t play, it opens up the door for another player. Often when a player is injured, you’ll see a sudden jump in his replacement. You need to factor in how long they are going to play as well. Is the player a replacement for one game, one week or 2 months?

You want to utilise the player in the time that they have that opportunity. A player might have an opportunity to go from a reserve to a starter to replace someone for 1 week. So I’ll pick them up at the start of the week and then drop the at the end of the week. This way I get as much out of them as possible.

You need to factor in the number of games played. If you are targeting volume based stats i.e. accumulation of points or rebounds, then someone that plays more games in a week is more beneficial. In this scenario, a player that plays 4 games is more valuable than someone that plays 2 games. If this is your strategy, you’ll need to be a more active team manager, constantly updating your team to find that edge.

It also depends how much free time I have. If I’m not going to play for a while, I might go for a longer term trend in a player. Someone I believe is going to be consistent over a longer period of time in terms of opportunity and statistics. As I might not have the time to constantly add and drop players every week. So I’ll pick up a player that’s going to stay in my team for next few weeks or few months.

Final thoughts

My strategy has evolved over the season. I drafted players that fit my thesis at the start of the season. Based on how I was doing and also relative to the competition, I continued to focus on specific categories I was consistently winning. I continue to draft free agents that meet my thesis.

One of the biggest mistakes I have made in the past is picking my favourite players. As a huge Bulls fan, I picked Derrick Rose as #8 pick a few years ago. Big mistake as he was coming back from injury. After that experience, I no longer take my favourite players. Fantasy basketball doesn’t care that you like a player. Its about statistics. Winners are grinners 🙂

I also watch a lot of NBA, not as much as I used to. But it does help to provide some colour. I like to watch games with the less popular teams like the Milwaukee Bucks, Utah Jazz or Denver Nuggets. I like watching basketball in general, but I’m looking for lesser known players. That’s how I discovered Gary Harris from the Denver Nuggets. He’s only played a few seasons. He had an opportunity to play in a game I watched and I really liked his energy & activity. He was shooting a lot of 3’s. He looked like a player that would improve over time if he had the opportunity. Harris was also on my radar but that game helped push him over the edge to make me draft him. So I recommend doing the eye test and watching some games as well.

I’m out like the fantasy draft,
Matt Ho

Leave a Reply

Your email address will not be published. Required fields are marked *

CAPTCHA * Time limit is exhausted. Please reload CAPTCHA.