Power Trading - How Does Pricing Work In RT / DA?
I have been trying to understand a market on my own but am a bit unclear as to how the pricing works in the DA / RT markets. As I understand it, pricing occurs at resource nodes, hubs, and load zones. LMPs refer to the marginal cost to serve each incremental load at an electrical bus and are used to calculate settlement pricing for DA / RT at the resource (locally) and hub / load zone (regionally) levels.
-
Are the LMPs representative of the bids market players submit during the DA auction and RT market (5-minute level)? And the settlement prices (settled hourly for DA and 15-minutes for RT) are simply a reflection of the clearing prices / awards of these markets, thus, determining the final charges and payments that flow in and out of the grid?
-
Finally, what do resource / hub / load zone pricing actually mean? Is it something like resource node corresponds to the pricing associated with generation (resource offers), load zone is associated with cost of transportation to serve load (buyer bids to purchase energy), and hubs are cost to move to the middle?
I am trying to clear this up, so any help would be great.
To get a RT role, does an applicant need to know this?
I’m not sure but I’m trying to learn as much as I can to show interest
I see. Would you mind saying where you studying this info?
I’ll try to keep this brief, because it can get pretty intricate.
First, you’re generally right about the different types of nodes, but its not universal. A node is just a location on the wholesale grid the ISO can get an electrical reading on. A node could have both generators interconnected at a substation that is also wired to pull power out of the grid, so its not that clean cut. Hubs/zones are just a weighted average of set of nodes that are broadly meaningful to market participants. I’ve never actually looked myself nor heard anybody ever say the load zones are only aggregations of load nodes. I’d imagine its pretty unlikely given how many nodes there are, but at least in my experience nobody really worries about stuff like that. Hubs/zones don’t have anything to do with transmission really (“moving/transporting power” as you put it.) Thats just transmission, which you analyze via congestion. Congestion pricing is one of the 3 components that make up LMP, the other 2 being energy and losses. Majority of the time energy is 95+% of LMP.
For LMP, you’re pretty much spot on. Its all about the bids/offers participants submit to each respective auction. DA clears once a day, for the following day each hour. RT can clear 5, 15 mins (or hourly in some western/southeast regions that don’t have typical deregulated ISO operated markets) day-of. The hub/zone pricing is again just the weighted average of all nodes that make up the zone, for each time interval of each market.
The way to think about DA vs RT is that DA is how things are “supposed” to work, in terms of demand and supply. You submit a bid/offer, and if you clear, you’re obligated to provide or use that amount of power. The RT is then a “true up” for how things actually shake out that day. More/less power than anticipated is used/consumed, and if you cleared DA, you get/pay the RT price for the difference in volume. Which can mean or less money, depending on whether you’re short/long and bidding/offering. Asset optimization, for both generation and load, is mostly about trying to maximize gains and minimize risk by navigating this process. Along with signing longer term hedges to mitigate the risk thats inevitable with trying to forecast power supply and demand.
Hopefully that covers your questions. They’re good questions so happy to try to answer and follow ups.
Thanks for this information. This helped cleared up some things. I just have one question more on settlement point prices (SPPs). Are the SPPs representative of the awards / clearing prices in DA / RT? What is there function? I surmise that they’re probably calculated differently across markets (saw the SPPs formulas for ERCOT for resource / hub / zone) just like LMP calculations are different for my region (ERCOT LMPs exclude the impact of losses versus NE ISOs like PJM / ISO-NE). Again, I appreciate it!
Sure. So as far as I'm aware, settlement point prices only exist in ERCOT. SPP = LMP + ORDC adders. ORDC adders are a unique ERCOT concoction which add $ based on the amount of current deployable reserves in ERCOT.
ERCOT tweaks the ORDC curve regularly. Here's a doc on it if you want to read more. It is one of the more interesting components of ERCOT that distinguish it from the other ISOs. That's a good doc that shows how intricate (or fun, depending on your interests) this stuff can get.
https://www.ercot.com/files/docs/2022/10/31/2022%20Biennial%20ERCOT%20R…
An example of the type of discussion/analysis people do, a big topic is how the ORDC will change given increased battery deployment in ERCOT. Will they serve to increase operating reserves and thus reduce ORDC adders? If so, will ERCOT introduce another type of capacity support mechanism? How should batteries be "accredited" as "reliability/deployable" reserves when they're very fleeting and require charging to be available? I.e. poor battery operation is just as likely to harm reliability by charging at inopportune times as they can help it. To this extent, ERCOT introduced the "ECRS" last year which has reduced ORDC adders by classifying ECRS capacity as reserves but increased LMPs by removing these MW from the on-grid deployment.
This link is more high level, but has a good graph of the ORDC curve.
https://www.enverus.com/blog/ercot-volatility-how-are-the-ordc-changes-…
Stepping back, this is one of the levers ERCOT uses to make up for the fact they don't have a capacity market, where most other ISOs do, to some extent
(CAISO and SPP are more like "grey" capacity markets, while PJM/MISO/ISONE/NYISO have the more typical capacity markets). So functionally, in ERCOT, SPPs are just treated as the de-facto LMPs - the price that participants pay/get paid.
There are situations, mostly bi-lateral, where participants will settle on the LMP instead of the SPP and thus sidestep the ORDC adders, which can get into the $1000s, so its can be a big deal.
As a fun fact, during Winter Storm Uri, when the ORDC was at the cap for nearly days, there were some older contracts before the SPP implementation that just said they settled against the "LMP". Causing one party to save millions and avoid bankruptcy on projects, and other other to lose millions. These went into litigation about whether the LMP should functionally mean the SPP since that is the de-facto pricing index.
There may be instances where the other ISOs use the term settlement point price, but they don't have ORDC adders like ERCOT and are really just LMPs. So its some semantic confusion ERCOT introduced by calling it the SPP.
I have never heard of LMPs not including losses, that would not be an LMP in that case because by definition it must include losses, but generally they're really minor compared to the energy component.
completely unrelated but OP probably knows wayyyyyy more than the avg rt trader on shift. I am for sure certain he knows more than me and I have been doing this for 5 years lol. Good job bro. You have a great future ahead of you
If only that translated to getting a shot at one of these places haha
Eos aspernatur ipsum earum praesentium a. Placeat et rerum magni odit id et. Neque fugit qui eligendi harum.
Autem rerum sed earum. Vel sed est optio modi similique laudantium tempora. Corrupti recusandae delectus ut et vitae recusandae a debitis.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...