Seasonality lowers the 2008 projection

An unknown Ken brings up a good point about my annualization of 2008 numbers in an earlier post about exercise. He asks if there is a seasonal effect, for instance, a spike in activity in early January while New Years' resolutions are still fresh. I don't think the particular example is right since I tend to make resolutions like "Spend more time at charity events," or "Save money elsewhere to spend it on occasional fancy groceries," but I acknowledge there may be seasonality.

Other than a consistent spike in July (a beautiful outdoor month in Boston), it's pretty hard to see the pattern in this data.

One thing that is clearly highlighted is my 2007 effort to train for my first marathon (the red line that jumps up in February). I went out hard, and while you'd hope to see that line trending upward as the mid-April race date approached, it actually trended down, revealing the results of my increasing knee pain.

Another big jump comes from a decision to start lifting weights in the spring of 2008. That heightened level of activity may be unsustainable, and points to a regression to pre-2007 levels unless I maintain the weight-lifting goal, or establish a new one.

However, the question remains – is there a difference between the seven months of data I've collected and the rest of the year? This next graph should clarify the issue:

Using this new information, we can prepare new estimates for the 2008 full year. With 205 days on record, we can figure the number of remaining projected workouts by multiplying 60 sessions by (365-205)/205 and then applying a 25% discount and arrive at a total of 35 additional sessions (as opposed to the original estimate of 41).

While this doesn't change the previous conclusions, which were based on year-over-year changes, and not just the 2004-2008 change, it does make the increase in '08 seem more reasonable, although 2008 still projects to be quite an increase over past years:

Read more...

Workout length poised for a regression to pre-graduate school levels

It appears that 2004-2006, I averaged about 50 minutes per session, while in 2007 and 2008 I averaged about 20 minutes more. The most obvious driving force here is that I attended graduate school September 2006-June 2008, which left me with much more free time than when I was working.

So, what caused the longer sessions? Was it simply that I didn't have work waiting for me at home, and that allowed me to run farther? Partially, yes. I began playing squash on weekdays, something that had been strictly a weekend pursuit for me in the past. That sort of gain is likely to be reversed. Then again, having more time off also inspired me to set larger goals. I ran a marathon in the spring of 2007, and started lifting weights in spring of 2008, both of which made me spend more time at the gym. Those gains may be maintained.

A shift in the type of exercise might also help explain the change in workout length, but the exact composition and its change is unclear. It's also unclear why 2006 shows a decrease in workout length when the last 6 months were in graduate school or on vacation.

It seems clear that when predicting the future, the rise in number of workouts could be sustainable, but that the increase in workout length may or may not be in for a regression to the historical mean, depending on whether changes in exercise mix and goals or weekday squash was a larger driver of the 2007-2008 increase.

Of course, there is a significant observer/actor bias, but we'll assume that exercise is driven by stronger forces than graph creation.

Read more...

Number of workouts per week has strongly driven increase in exercise since 2004

Last time I looked at how much I've been exercising over the past five years. It looked as though exercise has been increasing, but I wanted to know why. The first question to address is how more frequent sessions and longer sessions have impacted total exercise. Have totals been padded because I've made exercise a more integrated part of my life or because I've had more time to spend at the gym? First let's look at frequency of workouts:


This shows a very steady growth of ~10 extra sessions/year. I'd expect this data to be pretty noisy since the decision to exercise or not on any given day is based on so many factors, and because the data refers to ~50-70 events per 365 days. Nevertheless, four years of increases of similar magnitude is enough to convince me that something is going on.

I interpret this as a steady integration of exercise into my daily life. While I did occasionally work out in the years before 2004, it was mainly pick-up sports at infrequent intervals. I had never jogged, never used a weight room (with the exception of 3 months in 2000), and never set goals to improve my endurance, strength, flexibility, or agility.

What this means for the future is that I would expect this trend of increasing exercise to continue and gradually level off. This stands in contrast to the other factor responsible for my increase in total exercise time, which I will investigate next: length of workout.

Read more...

Time devoted to exercise has been growing

Since December of 2003 I've kept an exercise log. It started as a way to track how far and how fast I ran. At first glance it looks like I've increased the amount I've worked out pretty steadily over the past five years. The graph raises a number of questions, though. Why has exercise time gone up? Has it been from longer sessions? More frequent workouts? Is it driven by free time or is it independent of free time? Are the averages even over the years or do they represent periods of intense activity separated by periods of sloth?

Clearly, this needs some more attention.

Read more...

Quitting my job to be a poker pro is not a viable option

My first graph takes a look at my success playing poker online over a three year period 2003-2006. It looks like I’m pretty successful at low limit Hold-’em, but not so successful that I could make a living from it. It also appears I’ve hit my limit online, and that increasing the stakes isn’t a viable option.

I played online at PartyPoker and Paradise Poker, usually sitting at two tables at once (this was possible and advisable since there’s some significant waiting time as the other 9 players at the table act. People who do this for a living will often sit at four or more tables at once, but I found that the loss of concentration and fun I suffered made that a bad choice for me). I chose full table games rather than one-on-one or short handed (six seat) games. As you can see, I won fairly consistently at about $20/hour for the entire period I played after the first few sessions while I was getting the hang of it. You can see that of 32 sessions after the ‘learning period,’ fewer than a third were losing sessions, most of those were small, and a losing session was followed by another one only once. All in all, it looks like a pretty appealing way to make money.

On the down side, playing two tables at once for a long time got to be more of a grind than a relaxing activity. With the rise of programs playing tight poker at the lower tables, games got to be much less exciting, combined with the passage of the port security bill in 2005 it seemed like a good time to quit playing online. I quit entirely after a PartyPoker player obviously cheated, admitted to doing so in the chat box. I notified PartyPoker, they reviewed the incident and declined to discipline the player. That was the end for me.

If I’d been able to duplicate my success at higher limit tables, it might have been a good idea to continue playing. However, as you can see in the next graph, moving up to $3-6 and $5-10 did not result in winning play.


One could believe that what you see here is just the “learning period” that I experienced with $2-4, but I think it’s more than that. I needed more than practice to improve at these levels. As with anything, you need to practice, but you also need to develop technique. In poker, developing those skills requires reading books and articles, tracking and analyzing past performance, and probably playing live games. And, as with all sports, this work on technique is much less exciting than actually playing. I decided it wasn’t producing enough fun or profit to do that sort of concentrated work.

Interestingly, I also tried changing from Limit Hold-’em to No Limit, and played about 125 hours online.


While I started with a lucky run, the last 100 hours of play were slightly worse than break-even online. I’m not sure why my bankroll went up, then leveled off – almost like a reverse “learning period.” I guess this is what people call beginner’s luck, and what I call attention-grabbing, but irrelevant statistical noise.

Read more...