4/17/2009

04-17-09 - Waffling

I'm waffling about what car to get. I really *need* a new car because I've realized that the Qualude is just too small for me. I've been sitting hunched over all my life, and I'm trying to correct that and decompress my spine, and I just can't do it.

I'm waffling about whether to have surgery. Some days my shoulders feel sort of okay and I think "I could live with this" but then other days I'm in a lot of pain and it's just all crunchy and grinding and searing pain and awful. And of course any time I try to do a pushup I think "I need surgery".

I always see both sides of every decision, and it's quite paralyzing. In my opinion, most people, even smart people are very cavalier about ignoring the pros and cons. But they are probably right to do so. It's better to just pick something and go with and do your best with that decision. To put it in programming terms - I would say 90% of the smart programmer that I know are far too opinionated about various style issues, they don't rationally give fair credit to the other side's arguments and see that there are good arguments on both sides. But that's fine. Somebody who just says "low level C-style is the one true way" and runs with it might be wrong, but they can be productive because they have just made a decision and are getting work done. Somebody who sits around all the time thinking "hmm would it be better to do this C-style, or C++ OOP? or maybe need a GC language, or I could use OCaml for this bit.." is wasting way too much time equivocating.

Most successful business men, and just dynamic and fun people that I admire, are quick decision makers that just pick something and stick with it.

A related topic I've been thinking about recently : I've always been very dubious about the idea of learning from people who have been successful. There's this whole cult of worshipping rich people, reading interviews with them, getting their opinions on things, trying to learn what made them successful. I think it's mostly nonsense. The thing is, if you just look at who the biggest earners are, it's almost entirely luck.

Think about it this way - you have a bunch of gamblers. Some of them are very good and will make a steady profit of +10 a year with not much variance. The others suck, but play very wild, so their expected profit is +0, but they have high variance, so it could go between -100 and +100 in any given year.

If you look at the list of who the top winners are in any given year, it will be the people who suck. It will just be the ones that happened to get lucky. If you also look at the biggest losers, they will be using the exact same strategy as the biggest winners.

To make a more concrete example - lets say you have a biased coin you're allowed to flip. It has a 55% chance of being heads and you can bet as much of your bankroll as you want on it. A smart gambler might use a "Kelly" bet size ( see my old blog post ) and make a nice expected profit. A crazy gambler would just bet their whole bankroll on every flip. If you have 1000 people betting smart, and 1000 people betting crazy , after 10 flips all the biggest winners will be people playing the crazy strategy.

Anyway, the point is if you just look at successful business people, they will probably be confident, decisive, risk takers, aggressive at seizing opportunities, aggressive about growing the business quickly, etc. That doesn't mean that those are the right things to do. It just means that those are variance-increasing traits that give them a *chance* to be a big success.

Now, just like in poker, there are times when variance-increasing is the right move. For example, if you believe that you basically suck. If you think you don't have an edge on your opponent, then the best way to beat him is to just play really wild and hope you get lucky. For example, the best way for a novice poker player to have a chance in a tournament is to just do lots of all-in preflop shoving and hope you win the draws. The same thing could certainly be true in capitalism; if you believe you don't really have any great money-making skills, your best way to get rich is to variance it up.

4 comments:

Tom Forsyth said...

Subaru WRX. Seriously - stop fucking dithering and just buy one. They fit me (6-foot 1 with short legs) and my wife (6-foot with long legs), they're fun, decently fast, decently economical, they're completely impervious to rain, ice and snow, plenty of space in the back seat and trunk and they never break down.

In fact if you want to buy mine, that would be just peachy. I need the garage space for a GT-R.

cbloom said...

Meh. I love the WRX and if I could have two cars one of them would be a cheap used WRX for bad weather days and off-roading and when I just want to beat something up.

But I have two big problems with the WRX :

1. Cheap interior, rattly, lots of road noise, crazy stiff ride, just feels like a box of bolts. That's fine for rallying but not what I want in a daily commute.

2. No low end torque. Until the turbo kicks it accelerates the same as my Prelude. I want something that can jump from a stop so I can pull dangerous moves like passing people in intersections on one-lane roads. I'm sick of having a car with a terrible 0-20 time.

Sly said...

For this, one needs a light car, not a powerful one.

Sly said...

or an electric car, like a Tesla or a Fetish. Quite pricy :(

old rants