Tracks finalized 30-minute UK grid carbon intensity actuals from NESO, measured in grams of CO₂ per kilowatt-hour.
Click perpetual in the title for more info.
1,000With claude I have at least planned the fixes necessary to make all of the issues that have presented here avoidable, which is good!
My assumption is that even once those things are fixed, this market probably would suck anyway. Any market which is going to end up being about trading the derivative (in this case, trading the residual between the forecast and reality), it's probably better if that is the underlying for the perp instead of expecting other mechanics to make it viable
for example, this market would probably make more sense tracking the actual - NESO forecast instead of the raw intensity.
Either way, I'm going to add some more liquidity and let it rip again so I can keep seeing the interactions. Haven't pushed the fixes that would stop it freezing, bc I need to spend a while making sure they make sense mathematically across all markets, but I'm pretty sure those things are ~solved
Thank you for participating throughout the confusion 😅
@Mochi I believe they're using it as the stress testing ground for perp tech. easily the perp that has the most pains
@Mochi In the abstract it could turn into betting on the variation while the predictable expected movement is zeroed out by the market. But it’s clearly not working like that yet.
@Mochi This was basically the perfect stress test, and represents the most extreme predictable swings you would realistically ever expect from any market.
It helps us very quickly understand how the system needs to be adjusted to make this tradeable. If we can get it to that point, then we know how we need to tweak things so that these less-frequent markets can still be viable.
It was never expected to be a sensible market day 1, but it is perfect in the sense that:
everyone can easily see how to trade on it
when the funding rates were capped, it was literally free mana - people get to experience ADL and we can stress test that side of things with multiple people
this kind of movement is ~desirable for perps, this market just does it extremely quickly / frequently. If it were stretched over 1 month ups and downs, it would be easier to adjust everything, but it would also take way longer for us to figure it out.
as it stands, it roughly works with the extreme funding rates, but is still quite weird (mostly because of the oracle jumps).
We may even totally shut down this market eventually, but it has been extremely helpful in the launch
You really don’t want these issues to only come to light in months when, say, the Trump approval rating perp suddenly has a huge predictable swing. It’s also a nice that it’s not an artificial example; it’s a real quantity where predicting it and trading on the uncertainty are definitely relevant to somebody.
@Simon74fe yeah I'm disappointed that I often can't trade at all, even against the current trend. whatever amm implementation is scaffolded here seems to be failing. that or it's been bled dry
I tried to “refinance” to higher leverage by closing and reopening my position, and got a strange error message even at low numbers and margin:
@traders The market conditions have materially changed. You may want to take another look and consider closing out of any open positions.
We may close this market again at any time, it is being used for experimentation, and we appreciate your patience and support in exploring the edge cases for perps!
This market is broken and will remain closed while we address the specific flaw (it is known, understood, but not obvious how best to handle it).
It happened because this market is quite predictable, and the pool only had a small amount of funding. It is very unlikely to occur in the other live perps!
@moobunny It seems from my armchair that this could only work with much larger possible funding rate numbers, so that, say on the daily upswing, you could choose between paying for the right to go long or being paid to take the short position. In that case you’d functionally be betting on the slope of the curve, with long positions winning if it went up faster than the market expected and short positions winning if it went up slower. As it stands, it’s simply not possible for it to be rational to hold the bad position.
@moobunny yes, I will unstick the market in a few minutes and I am raising the funding cap by nearly 100x. At that level, holding the crowded side into the predictable move costs roughly the expected move, so you're effectively trading the deviation from the expected curve. We'll tune the number as we see how it behaves
The freeze itself was a separate bug that can occur at any funding rate: profit payouts wer partly backed by the other side's unrealized losses, and when the price reverted those losses un-realized while the payouts didn't. Fix for that is coming separately