This has been a very distracting week, to the point where I don’t have a pure data topic to pen (I’m totally happy to take topic ideas if you have a question you’d like me to explore! Random questions was how the “data cleaning is analysis” post got started.).
This week I’ll unload a bunch of interesting data-adjacent topics that have been occupying my brain cycles.
First, good news! I found an artist to help w/ a logo
As of this writing, we’re still going back and forth with a design. I specified things roughly in line with what I had originally mentioned in last week’s newsletter. This far the rough drafts are running in an unexpected direction, but in a good way. Hopefully I can move on to making something fun for everyone soon. Hopefully.
Finding an artist was surprisingly difficult. I made some tweets and someone eventually hit me up and we decided to work together, but I didn’t have a firm concept in mind, so I didn’t have a good idea of who to search for. Ultimately I’m working with a freelancer with a fun portfolio.
Things learned:
1 - Logos are not cheap, even for budget things that aren’t for big commercial use. I’ve commissioned work twice in my life thus far, both for personal projects where I made it clear it wasn’t for a significant commercial venture, and it still averaging a cost of around $500 each (plus/minus a wide error bar, I don’t want to publish people’s rates w/o consent).
Yes, you can find people who will do it for much cheaper if you have a budget, but it’s only fair they get paid for their time.
2 - I continue to suck at communicating. The thing I described, and the original concept I had floating in my head are COMPLETELY different things. The only overlap are the words “Counting is Hard”, which I explicitly specified. I suppose if my concept was super important to me, I would’ve sketched it out instead of using words. But I’m happier being surprised this way, because I like the WIP result and it’s not something I’m capable of imagining.
Next up is I need to figure out what the heck I want to make, besides stickers. That will let me start narrowing down vendors.
Data Mishap Night is happening!!
https://datamishapsnight.com/
Caitlin Hudon and Laura Ellis are hosting a wonderful little event this Friday, the 5th, at 7pm CST. People will get up and speak for 5 minutes on various mistakes they’ve made over the years in the data world.
I’ll be telling a VERY innocent and cutesy story in the “Too Good To Be True” block. No big explosions or humble retractions here. There may have been some late night booze involved. I’m looking forward to hear all the other speakers.
The event will NOT be recorded! So if you want to hear the stories, please sign up and join!
Geez, what’s with the stonks
Full disclosure, I hadn’t put all my annual bonus money away into various savings vehicles yet, so I put a small chunk of it into GameStop stock somewhat near the start of the frenzy… Now, as of this writing, I’m strapped onto a ridiculous roller coaster, and I should just get off while the getting’s good.
There’s a ton of conflicting news and opinions out there, from accusations of market manipulation by hedge funds and even by retail investors, to people calling this some story about small investors beating Wall Street. I doubt ANY of that is true. A bunch of people took a ridiculously bad bet, it got found out and was punished to the tune of billions, with profits flowing to other big players. The only surprising difference is that small investors played ANY role at all in all this, and they now have a chance at both getting a taste of the action, as well as eating a ton of the future losses.
I grew up in NYC in the 90s, and as with many of my peers, if being a doctor or lawyer wasn’t your fancy, finance was probably near the top of the list of high paying jobs of interest. It didn’t help that Stuyvesant high school was less than 15minutes walk from the financial district. We even had a class dedicated to finance where we studied things like the efficient market hypothesis, topics in behavioral economics, and standard financial instruments.
Looking back, spending a bunch of years learning about squeezing information out of piles of numbers and using accessing horribly ancient data feeds from the government in the last 90s, early 2000s, had probably secretly primed me for my current job. After all, slurping data from various sources and trying to read tea leaves out of tables of numbers is similar across both worlds. Also, finance had been one of the few industries with both the hyper-competitiveness AND ridiculous piles of money to do Big Data before the rest of the world.
So, while I’m significantly out of practice with doing stock-related things outside of passive investing, I got on this ride fully knowing these things:
1- that the situation for GME is really unique with the ridiculous >100% short interest on it that would lead to a short squeeze.
2- that profiting off it amounts to playing a stupid game of chicken with traders and high-frequency trading algorithms with far more experience and information than me, I’m more likely to lose than win
3- but despite #2, this is literally the only time in my life where I’ll be able to have any participation in a ridiculous move like a short squeeze. It’s the sort of market maneuver that you only hear billionaires and big institutions wielding alone. Retail investors don’t normally have a chance to participate before it’s all over. Just that experience alone is worth the full potential loss of my money to me and my younger self.
4- This’ll probably never happen again because it’s very likely legislation will make this unique combination of events impossible in some way. There’ll be a giant swathe of retail investors that’ll get burned by this for a ridiculous amount of money, and legislation or rule changes typically follow such events.
So that’s how I got stuck on this ride for a lark.
Update #2: On Monday, before publishing this post, I closed out my position at a respectable profit, but far from the heady peaks. The main issue for me was I had to devote attention to keep an eye on things before the bottom fell out and I don’t have much attention budget left with everything else going on.
Prisoner’s Dilemma playing out before our eyes
All the stuff that’s happening is actually a fascinating study of the Prisoner’s Dilemma game in that the temptation to sell and run off with the cash is very strong. But because there’s a squeeze on number of available shares to purchase, the internet-squad, loudly making noise and providing cover for the moves of the silent institutions taking advantage of the situation (and likely providing much of the real financial muscle in all this), is trying to hold the shares to bully the short sellers.
Normally, a PD game (and the iterative variant that’s often used to study the evolution of cooperation) assumes that players can’t communicate with each other to communicate. Here, the temptation to walk away with thousands, and potentially millions of dollars in profit are extremely strong.
Sidebar: I do vaguely remember, but can’t find the reference for, a Iterated Prisoner’s Dilemma strategy that barely managed to beat out Tit-for-tat, the usual champion, by encoding a secret handshake so communicate information between the players and one player sacrifices themselves to the benefit of one player, giving it a net edge.
Here we have an entire anonymous mob trying to chaotically communicate out in the open about their strategy, with countless people shouting “hold the line” (aka, cooperate with us). These people dangle the potential for a much bigger profit in the form of short sellers giving in to pressure and being forced to buy sales.
The trick to surviving this mess is to figure out when are those statements true (in that there really is still short sellers trapped into buying scares shares) and when the tide has turned and there’s not enough shorts left but the shouts to cooperate continue in an effort to trick people into holding the bag at the end. Guess who is most likely to have information about this first? It’s not the retail investors.
There’s all sorts of speculation and drama going on as the stock price bounces wildly, putting strain onto the financial system. There’s accusations of market manipulation, charges of false news reports in the media, bots spamming conflicting messages, etc.. Listening in to the tone of the whole conversation, is fascinating, and is the equivalent of watching the dynamics of an untrained mob try to figure out if they want to go charge into battle or not. When the ranks break it won’t be pretty.
I’m not going to bore you w/ breathless recounting of what’s going on. But I did want to highlight this thread that explains a bit of the intricate details involved in the settlement process of a trade (which normally takes 2 days. It’s something that I never put much thought about, after all, a trade “just happens” and is counted.
But the process is fascinating because they don’t “just happen”, and there is actual risk that the 100 shares of stock you bought in a transaction actually “fail to deliver” for various reasons. In such instances, the broker themselves has to absorb the risk and make good on the transaction.
To help mitigate this risk, brokers are required to put up collateral to help guarantee trades go through even if the counterparty to the transaction manages to not live up to their end of the deal. That counterparty risk because ridiculously huge (to the rumored tune of billions) when some stocks go hyper-volatile like we’ve been seeing the past few days.
Who woulda thought counting stuff continues to be hard?
About this newsletter
I’m Randy Au, currently a quantitative UX researcher, former data analyst, and general-purpose data and tech nerd. The Counting Stuff newsletter is a weekly data/tech blog about the less-than-sexy aspects about data science, UX research and tech. With occasional excursions into other fun topics.
Comments and questions are always welcome, they often give me inspiration for new posts. Tweet me. Always feel free to share these free newsletter posts with others.
All photos/drawings used are taken/created by Randy unless otherwise noted.
This was a nice deviation from the usual DS stuff. I liked it!