As I wrote in my earlier post on the importance of predicting minutes played in daily fantasy basketball, I’m new to the game, testing out different strategies and doing a bunch of research to see what I can learn.
I never think inexperience is a net positive, but I do believe there are ways to leverage it to help you play daily fantasy sports; mainly, I think the fact that I don’t have any preconceived biases can help me create lineups that aren’t influenced by widespread narratives or ‘groupthink’—something that’s difficult to accomplish when you really become immersed in a sport.
Since I’ve recently decided to test the daily fantasy basketball waters, I thought it would be cool to take you on that journey and write about some of the stuff I’m researching. Some of the numbers I post might fly in the face of conventional wisdom, which I think is a good thing at times, while other data might support the consensus.
Analyzing Aging Curves
No matter the sport, the key to daily fantasy success is making accurate predictions. This game is entirely about making better predictions than the next guy.
Since daily fantasy sports exist as a marketplace—each player has a salary and you’re basically buying stocks in anticipation of price rising—the value of each prediction you make is very much tied to public opinion of a player’s current worth. Our job is to ask, “Is that perception accurate, and how might the future perceived value look different?”
I’m not entirely sure on the general opinion of how age affects NBA players—I’d imagine it’s more important in football—but I think it’s safe to say that players in all sports typically evolve at specific rates; their play improves after coming into the league and ultimately declines as they age. Except in rare cases, that’s the case for everyone.
I think there’s a ton of value in understanding typical aging curves because they give us a really good idea of when players might break out, when they could see a drop in production, when they’re at their career peak, and so on.
We all know that second-year players are typically better than they were as rookies, but how much? Quantifying that improvement, for example, can obviously aid us on the macro level of player evaluations, but it can also help us even on the micro level of daily fantasy basketball.
Typical Aging for NBA Guards
I don’t really know whether or not certain positions age differently than others in the NBA—given that some take more physical abuse on a nightly basis, I’d guess so—but certainly the evolution of particular statistical categories is probably different for guards, forwards, and centers. I wouldn’t expect the career trajectory of assists per game for a guard to resemble that of a center, for example.
For that reason, I’m analyzing career production by age for guards only. I charted the percentage of career peak numbers that guards produce at each age. If a player peaked with an average of 30 minutes per game at age 28 but played 27 minutes per game at age 27, then the age-27 number would be 90 percent of peak production. Here’s a look at the evolution of minutes, points, assists, rebounds, steals, and turnovers for the average NBA guard.
You can see they every stat except for turnovers peaks at ages 28-29. Note that I’m not at all saying that every guard will peak at that age, but just that the late-20s is the most probable timeframe for an NBA guard to be in his career prime.
I think what’s interesting is to look at how each of these stats shifts by age. To do that more easily, I created this graph.
You can see that young guards have the most trouble with assists early in their careers, producing only 62.59 percent of their eventual career peak, on average. I think that makes sense since most assists aren’t really about using pure athleticism, but rather about being smart and knowing a system to find the open man. Meanwhile, young guards are best in terms of rebounding, which I’d argue is more of a reflection of athleticism and explosiveness than assists.
5 Important Points
There’s a lot of interesting stuff going on with this data. Definitely give it a look and see what you can find, but here are the five points I believe to be the most valuable and actionable.
Assists climb slowly, but are more consistent late in guards’ careers.
Again, I think this is due to assists being primarily the result of cerebral play and on-court chemistry that takes time to develop. In the middle of a guard’s career, I think you can generally project the highest jump in assists—a stat that should hold steady for a longer period of time than points scored or steals. That’s great news for aging guards who rely more on assists than steals, for example.
Scoring drops in a hurry as guards age.
Points scored is a unique stat in that I think it’s very much a combination of intelligence and athleticism. Players tend to use more explosiveness to get to the rim when they’re younger and more experience to score later in their careers. Nonetheless, scoring drops pretty quickly after age 30 for the typical guard.
That’s important if you’re considering an aging guard who sees the majority of his fantasy points come via scoring. It also might mean that a player like Rajon Rondo is more likely than the average guard to maintain a high level of play into his 30s.
Turnovers peak early in guards’ careers.
Turnovers are influenced heavily by luck, but they’re also the result of inexperience or being careless with the ball. It’s not surprising to see that young guards turn the ball over more than older ones. Turnovers aren’t extremely important in daily fantasy basketball, but this supports the idea that “experience stats” take longer to develop than “athleticism stats.”
At ages 24-25, every stat exceeds minutes played.
This is a potential value opportunity because—as is the case in pretty much every sport—coaches seem to wait too long to play the young guns. I have more work to do on efficiency, but the fact that all of the bulk stats surpass minutes played suggests that guards are highly efficient at ages 24-25.
The average guard of that age, however, sees just around 86 percent of the playing time as he will at his career peak. When you’re talking about players in that age range that you know are going to see an elite rate of minutes on the court, that’s probably their new career peak where opportunities and efficiency are both maximized.
Consider buying low on guards in their mid-20s and selling high as they reach their 30s.
Even in daily fantasy basketball, it helps to know which players are likely to improve or decline from the prior season. In most cases, age 30 is about the time that most guards tend to fall off of a cliff. Some of that has to do with a decrease in minutes, but a couple years later the efficiency really drops as well. Especially early in the NBA season, it might make sense to be a little more bullish on players in their mid-20s and perhaps fade players as they reach age 30 and beyond, all else being equal. That’s especially true if you suspect a coach will limit an aging player’s time on the court to keep him fresh.