Explaining to my parents what I do | Hacker News Reader

Explaining to my parents what I do

neelsomani.com

117 pointsnsomani47 comments

Pricing at marginal cost is the biggest steal in economics.

Just because one folk who contributes 0.1% to the energy mix has double the average cost, everyone gets to sneak in and get free money.

Specially true in the last winter in Texas where we paid billions to the big power generators only to freeze to death.

The best trick that I have seen happening in the industry is suffocating supply so that the new marginal player just comes in, and then you collect all the extra free money, for essentially selling the same volume.

Good times.

For that, I don't blame the generation market as much as I fault legislators for horrible delays in decisions. The market cleared at the $9k/MWh cap for an excessive amount of time.

E: and it was avoidable!

There was huge backlash even when this super high cap was passed. The initial idea was that the market will regulate itself, thus no cap is needed.

The small detail they let out of the equation is that the taxpayers will have to bail out the market if it does not regulate itself.

And boy did we pay for nice ceo bonuses that spring.

I forgot about this hilarious event where a regulator made the mistake of telling utilities they'd get bailed out.

https://en.m.wikipedia.org/wiki/2021_Texas_power_crisis#PUC_...

Yeah I think the theoretical argument is that the uniform clearing price is supposed to incentivize producers to bid at their marginal cost

Which is a great idea if we had a thousand small producers of roughly equal capacity. It’s a horrible idea (for the consumers) when you have 3 huge players and then some tiny participants.

> ... it is a well-known type of problem, called a multiple integer program (MIP)

Is this meant to be "mixed integer program"?

Nice read. I find this especially interesting as I'm familiar with everything you've talked about here except only in Australia (AEMO). What you've written is like a parallel universe to what I know - very similar, but distinctly different.

I don't personally like the pricing model for energy in Australia and I can't wait for the realities of the 5 minute bid and future battery use like FCAS. But that said, even if this was a public utility model like I think it should be, and not this quasi free market (it's a highly regulated and very distorted market at times) price models like this would very probably work, and apply to balancing the energy supply and consumption.

Demand management as a virtual generator is interesting too. Supermarkets and shopping centres have huge HVAC which can bid in and out to consume surplus power into water chilling or heating and use it for its thermal qualities later.

Lol thank you! Evidently I did not have anyone proofread this...

I know close to nothing about economics and was a bit confused by this statement:

"The line sloping downwards represents demand. As quantity increases (bottom axis), the price people are willing to pay also decreases."

This makes it sound like there is an "available demand quantity" that is the independent variable and a "price" that is the dependent variable. But it seems a lot more intuitive to say "demand increases (I want to buy more and more people are willing to buy) if a good is offered at a lower price", where price is the independent variable.

Not sure why you're getting downvoted for asking a question, but hopefully this page helps: https://en.wikipedia.org/wiki/Economic_graph

It wasn't the graph that confused me (I've seen a supply-demand graph before and understood the axes were "reversed" compared to most other disciplines), but the description in the post itself.

Correct. Price is the independent variable, demand and supply are both dependent on price. But for some reason economists decided to put price on the Y axis. Confusing.

However. The demand and supply curves (lines) are dependent on price. But the intersection of those lines determines the optimal price. In that sense, price is dependent on supply and demand.

So 'price' the variable should be on the X axis since it's the independent variable which controls supply & demand. But 'the optimal price' is the dependent variable controlled by supply & demand, and so that should be on the Y axis.

Yeah I agree, I typically think about quantity as a function of price

>Another distinction is that most people will trade the average price of power over a given area called a "zone." But there are thousands of physical busses that people connect to, each of which has its own power price, and you can trade those via nodal trading (with instruments like FTRs and ARRs, again not worth going into).

This is really an interesting concept. It looks like nodal trading is available in the northeast - is it dependent on the RTO and how it is run? I'm passively familiar with some utility policies in Iowa but hadn't heard that brought up on the consumption side.

To the extent you can discuss, I'm curious what your saw during the Texas energy crisis about a year ago?

I spent a couple years on the buying and forecasting side for a very large manufacturing facility mostly on natural gas but heard some of the electricity discussions. Force Majeure conditions could lead to very large penalties if we exceeded our forecasted demand. I also sat in on some meetings discussing the massive electricity consumption and strategies to minimize total cost.

The biggest thing most consumers don't realize is how variable the production costs are. On the hottest days when everyone is using their air conditioners, special power plants come online to cover the added load. These peaker power plants may only run a handful of days per year, are cheap to build, but are very expensive to operate. Residential consumers often pay the same electricity cost per kWh despite the utility paying a highly variable cost.

Good intro to the problem. I did a fair bit of work on more complex versions of this problem in distribution systems at a past job. Because a distribution system is unbalanced, each phase has to be solved for but the phases interact with each other so the optimization problem becomes ~9x more complex.

Totally, when you start optimizing over multiple time steps it gets way more difficult. You can essentially solve a single time step by hand.

// Albert Einstein said: "If you can't explain it to a six year old, you don't understand it yourself." I like to think this is true about explaining things to my parents

- Isn't this a quote by Richard Feynman?

I think that is probably a "folk quote", the attribution to Feynman is also inaccurate AFAIK. He said something similar though:

> Feynman was a truly great teacher. He prided himself on being able to devise ways to explain even the most profound ideas to beginning students. Once, I said to him, “Dick, explain to me, so that I can understand it, why spin one-half particles obey Fermi-Dirac statistics.” Sizing up his audience perfectly, Feynman said, “I’ll prepare a freshman lecture on it.” But he came back a few days later to say, “I couldn’t do it. I couldn’t reduce it to the freshman level. That means we don’t really understand it.”

– Feynman’s Lost Lecture by David and Judith Goodstein

I like the subtle hidden arrogance that "we don't really understand it" if Feynman can't explain it at freshman level.

It's probably true that if he didn't understand it well enough to explain it at freshman level then nobody did.

I found it ironic that this initial quote seemed to be the intent of the article... which then goes on to display charts and equations that are unlikely to jive with a six year old.

This concept is not about explaining things in terms a 6 year old can understand.

It's about understanding the subject matter well enough that if necessary you could simplify and elaborate on topics to bring your audience to your level of understanding.

(i.e. - understanding your subject vs. being able to recite facts)

You really come to terms with how well you know a subject when you sit down and try to teach it to someone else.

That is a good way to curse your self!

Pretty sure the intent of the article is to explain to his parents (as indicated by the title and sentence after the quote)

Neither Einstein nor Feynman. It's just one of those quote memes that get stuck to famous people because those are the most likely attributions to get repeated.

https://quoteinvestigator.com/2019/10/19/barmaid/

So this is the business Enron was in? I can see why it can be so lucrative now.

Lol. With such volatile prices it can easily blow up, too. But yeah the backlash post-Enron is a large reason why areas like the Pacific Northwest never joined an RTO.

In About the Author section at the bottom, it should at least mention his name.

His name is there, just a couple lines below.

I don't see his name in the about section:

"About the Author

I'm a quantitative researcher in Chicago. I'm interested in computer science, math, philosophy and entrepreneurship. You can follow me on LinkedIn and Twitter.

Very interesting read. 6 year old me (optimistically) followed for about the first half and 40 year old me made it to about 85% through with full understanding.

Thank you for reading it!

Super interesting read. I ended up getting curious as to who he worked for, checked out his LinkedIn and it turns out he works for Citadel.. I swear I've heard of that company somewhere? ;)

He says he works at Citadel in the article itself. It’s a pretty large company in its market, doing rather well.

> I swear I've heard of that company somewhere?

Care to share where?

Google 'meme stocks'

I did, no mention of Citadel. What is this, some fight club rule number 1 and 2 where you gotta be cryptic?

It was Citadel (along with others) who heavily shorted these stocks. And supposedly they are still trying to save their positions by any means, including illegal ones.

I think your use of supposedly is definitely in order. A lot of what I read had connections that were shaky at best.. a lot seemingly read more like Q/conspiracy inspired bullshit to me than cold hard facts (to me).

Awesome work Neel - this is an inspiration to start blogging again!

Insightful thanks

nice read neel

I'm an anti-intellectual ball buster, so I'm triggered by the "ELI5" conceit preceding a verbose summarization of the linear programming problem. A better follow-up to the ELI5 condescension would be, "I exploit discrepancies between what energy prices are and what they ought to be for the benefit of my employer, and thus myself."

“And to the detriment of others”

Given the zero-sum nature of secondary markets, that's tautological and rather uncharitable to the trading profession. Arbitrageurs do serve a real, and given their compensation, quite valuable function of accelerating price convergence to a theoretical optimum.

Reply on news.ycombinator.com