Volume and Variance in the Long Game
This clip was taken from the Joe Rogan Experience MMA show #32 with Firas Zahabi, the head coach of Tristar Gym and one of George St. Pierre’s main coaches. The main take-away here was Firas’ emphasis on volume when training. In his words:
I’m a big believer in never being sore. You should train, the next day you should wake up feeling good. Okay, now why?
Let’s say I make you do pull-ups. And let’s say the maximum amount of pull-ups you can do is ten… Should I make you do ten pull-ups in our workout? No, I’m going to make you do five. Why? Because I’m setting you up to work the next day. The next day we’re going to do five. And the next day we’re going to do another five. And then we’re going to do six. When six is really easy, we’re going to do seven. Why?
If you did ten pull-ups on Monday, you’re going to be sore until Thursday… So, by Thursday you’ve only done ten pull-ups. Me, I’ve been doing five pull-ups every day. So I’m at twenty pull-ups already, twenty-five pull-ups. I have more volume than you. Now, when you add up at the end of the year who trained more, I’ve trained way more than you.
In the remainder of the clip, he compares how Russian wrestlers train with how American wrestlers train, and makes the point that intensity should only be done every once in a while because by it’s nature, intense workouts can only be done every once in while (otherwise they wouldn’t be intense).
This is great advice for working out, but it’s also a great approach for other things in life. Most people are familiar with the results-oriented mindset, which is one way of saying “optimizing for the results above all else”. What Firas is advocating for here is an application of the volume-oriented mindset, where volume-oriented means “optimizing for volume above all else”.
The results-oriented mindset tracks recent results. How many points did I score last game? How many games did I win in the past month? How many job offers did I get in the past month? So on and so forth. These results are recorded and used to evaluate one’s progress. Good results indicates improvement, and bad results indicate bad luck, stagnation, or regression.
The volume-oriented mindset tracks results, but at a much less granular level. Instead, it focuses more attention on making sure that you can and will continue to play in the future. The results are tracked a less frequently and are used to gauge the direction one is heading in, instead of one’s day-to-day progress.
If you’re the best player on your team and you hog the ball to score all the points, you’re prioritizing short-term results over volume. In the short term, you feel better about yourself because because you scored so much, but at the cost of volume, since you’re decreasing the likelihood that your team members will want to play with you in the future.
The volume-oriented mindset is the most valuable for activities where 1) the main determinant of progress is time spent doing the activity, and 2) there is a large degree of variance in short term results. The mindset is helpful in the first case since it focuses on maximizing time spent, and is helpful in the second case because it prevents you from getting caught up in the short-term results swings that are more indicative of randomness than ability.
For activities like pure mathematics where (allegedly) a large component of one’s ability comes from natural talent, the volume-oriented mindset is unlikely to be of any use. In addition, for one-off or occasional activities whose results are black and white (like pulling your bike out of a river), the volume-oriented mindset is also unlikely to be helpful.
Using the results-oriented approach in the wrong situations will wear you out emotionally thanks to the high variance in short term results. In addition, as time goes by, you’ll see other people that were less-skilled than you start to surpass you in skill level as they rack up more and more hours.
Using the volume-oriented approach in the wrong situations will likely give you an unsatisfactory result. This happens less often since most people default to a results-orientation and only adopt the volume-orientation for certain activities as they see fit.
There is also a difference in the decision-making process between the two mindsets. The results-oriented mindset focuses on positive objectives, where positive in this case means constructive, not happy or good. Positive objectives are objectives that can be progressed towards in steps, like “I want to kick the ball into the goal!”, which would be progressed towards, quite literally, by stepping towards the goal and finally kicking the ball at the goal. The volume-oriented mindset, on the other hand, focuses on negative objectives, where negative means “absence of”. Negative objectives assert something that is to be avoided, and avoidance is important to maximizing volume because there are so many things that can prevent you from doing an activity in the future (injury, sickness, angry teammates, legal troubles, etc). A negative objective might be, “I don’t want to injure my ACL this year”, which is not positive since there are no incremental steps that lead to you immediately achieving this lack of an injury. (This isn’t to say that you can’t take incremental steps in pursuit of a negative goal, it’s just to say that a negative goal is never directly achieved through a series of incremental steps.)
The result-oriented mindset also tends to draw direct relationships between actions and outcomes, in the form of, “when I do this, that good thing happens” and “when I do this, that bad thing happens”. This causal way of thinking works for situations where the variance is low or non-existent (i.e. the consequences of actions are consistent) and the time-frame is short. When you’re trying to pull your bike out of a river, you might observe that pulling it towards the left bank does not move the bike, but pulling it towards the right bank does. Thus, left bank => bad and right bank => good, and two minutes later, your bike is out of the water.
You can go into the volume-oriented mindset with a causal way of thinking, but you’ll quickly realize that it doesn’t work well. Sometimes when I do X, something good happens, and sometimes when I do X, something bad happens. Now whether you should do X is not so clear-cut as before — it depends on the context and how many observations you have of x. This means you have to adopt a probabilistic way of decision-making where each choice you make is a bet, so to speak, and no outcome is guaranteed. The key difference between “thinking in bets” and causal thinking is that when you think in bets, you don’t evaluate the decision based on the quality of the outcome, while in causal thinking, the quality of the decision is the quality of the outcome. Breaking the tie between the outcome and the decision quality is a necessary step that acknowledges the role randomness plays in outcomes. In other words, you can make a good decision and get a bad outcome, and you can make a bad decision and get a good outcome. In order to break this tie, decision quality must be evaluated in the absence of the outcome. You can either evaluate the decision quality before the results come out, or you can tell someone else about your decision, leaving out the resulting outcome.
For situations with discrete (categorical) outcomes (e.g. win or lose), probabilistic decision making sounds like this: if I do X, I believe there is a 70% chance that <good outcome> will happen, so I will do X. For situations with continuous outcomes (e.g. predicted quarterly profit), it sounds like this: if I do X, I am 95% confident that the result will fall between 50 and 100, and if I do Y, I am 95% confident that the result will fall between 55 and 110, so I will do Y.
The volume-oriented mindset is conducive to probabilistic decision making because of larger volume of data that one has to work with, which is necessary for constructing accurate probabilistic mental models. “Thinking in Bets” is a book and a talk by Annie Duke and the source for the parts of this post related to thinking in bets.