Adam Harstad Profile picture
Jun 7 6 tweets 2 min read
New post. The story of why car washes make me sad and some math about birthdays.

adamharstad.com/thoughts-for-0…

(Love or hate these notifications? Follow or mute #HarstadBlogs as desired.)
I started dating my wife when I was seventeen years old. We ended our second or third date with a moonlit walk along the beach– not so much because I was a romantic as because it was close, it was free, it was private, and I was loathe to let an enjoyable night end.
At some point in the conversation, I picked up a seashell off the beach and grandly declared that I’d keep it forever to remember this night. And then I did, keeping the seashell with me through nine moves in ten years.
I took to keeping it in the coin tray in my car so that I would periodically stumble across it and think back to the early nights of our relationship. The seashell (like the relationship) outlasted two cars with no signs of wear.
Until one day when I took my car in to get it detailed. I drove off only to realize the next day that when cleaning out the trash in the console they had mistakenly thrown away the seashell, too.

And that’s the 100% true story of why to this day car washes make me sad.
(There’s perhaps a parable here about how the physical reminder of an event eventually became more interesting and memorable than the event itself, but I’ll leave that to you to puzzle out. Sometimes a story is just a story.)

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Adam Harstad

Adam Harstad Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @AdamHarstad

Jun 10
Yes. Strongly against normative statements in dynasty.

When you join a league you accept basic obligations to your leaguemates. Basic reciprocity, no undermining the league, etc. But there’s no such obligation to play the way (a subset of) your leaguemates would prefer you play.
The big one I see is a toxic manifestation of the “there is no off-season” mentality. If one of my leaguemates wants to check out from May to July, that’s absolutely fine unless it was strictly spelled out and agreed to before joining. Different people like different things.
Some people like dynasty for the puzzle and persistent rewards, not for the fact that they can obsess about fantasy football for 12 months instead of 6. There needs to be a place for them, too.

My home league is basically on hiatus from February through July. And that’s okay.
Read 8 tweets
Jun 9
My takeaway: there’s no sub-sample size of “games to end the season” that is as predictive as just using the full sample.

And given that full-season stats are easier to generate *in addition* to being more predictive, there’s no benefit to “Last X Games” splits.
Other than the benefit a drunk receives from a lamp post, of course.

Support, not illumination.
Now, the other takeaway is that while “last 11 games of the season” splits (last 8+ games for rookies) aren’t MORE predictive than full-season stats, they aren’t significantly LESS predictive, either.
Read 4 tweets
Jul 23, 2021
With the Olympics set to kick off, I have been positively inundated by messages from people demanding I explain Goodhart's Law in an easy, accessible manner. My pager and telegraph have been beeping off the hook. I had to disconnect my fax machine.
Originally expressed as "Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes", Goodhart's Law is now most often now stated as "When a measure becomes a target, it ceases to be a good measure."
"But Adam, what does any of this have to do with the Olympics," you interject, "and do you really still have a fax machine?"

"I'm glad you asked that, imaginary interlocutor (who totally isn't just a rhetorical foil I invented myself), and also no."
Read 27 tweets
Jul 23, 2021
One of the best lessons I learned is to first have a goal, and then consider whether my actions further that goal, are neutral, or hinder that goal. And then try to do more of the former and less of the latter.

The guy who taught me probably didn’t even realize I was listening.
There are few things that feel as satisfying as responding to someone who is rude with some Sick Burns.

But my goal is to live a drama-free life where I am mostly liked and widely respected. Sick Burns prove orthogonal to that goal. So I just think them and chuckle to myself.
My inner life is hella petty sometimes. Not always, not even usually. But sometimes.

(I’d never think Sick Burns about *YOU* of course.)
Read 7 tweets
Jul 21, 2021
Very very very important point:

A large portfolio full of risky assets is not risky. It is safe. The bigger the portfolio, the safer it is.

*(provided they’re all independent of each other, which for fantasy purposes they usually are.)
*A* rookie pick is risky. You might get Mike Evans. You might get Sammy Watkins.

*SIX* rookie picks are not risky. Some of them will be Mike Evans. Some of them will be Sammy Watkins. By and large, you can estimate how many of each you can expect with reasonable confidence.
If you buy *AN* injured player in dynasty, sometimes you’ll get Demaryius Thomas and sometimes you’ll get David Wilson. It’s risky!

If you buy *A LOT* of injured players in dynasty you’re guaranteed to get plenty of both. Much lower-variance play.
Read 4 tweets
Mar 6, 2021
Volume is not more important than efficiency. Most of the time, volume *is* efficiency.

See also: Harstad’s Razor. (I’m never going to stop trying to make this happen.)

Toy model. Imagine NBA players see 5 “amazing looks” per game, shots that the average player shoots 60% on.

And they see 5 “good looks” (50%) and 5 “okay looks” (40%) and they see as many bad looks (30%) as they want to take.
Read 18 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(