it doesnt play by daily whims it has hardware that last a couple years before upgrading. it has electric contracts of 6-24 months. and so no matter the hashrate they still mine. as there is no cost benefit not to mine if they have contracts of electric for long periods. (they save nothing by stopping, they just waste time by not mining)
I think it's becoming more complicated than that personally, even if the mining industry remains majoritively traditional. Take for example the
hydrodams in Italy that are effectively weather-dependant miners during droughts (including winter), especially when gov subsidiaries are withdrawn. Not forgetting as well general energy demand, whereby if grids have excess electricity they are increasingly putting it into Bitcoin as opposed to expensive storage. While this is still the minority of the mining industry these hybrid models are becoming increasingly popular, so it's not 100% true that disposable miners are permanently used due to their short life span on fixed rate electricity. There's no need for an electricity contract when you own or have control of electricity, likewise if your produce it yourself with renewable energy, the latter we know is a sizeable chunk.
It may seem mad that affiliated industries would only use miners maybe 25-50% of the time, but they've obviously done the math and realised that for example continuing to run a dam over winter, even if only using miners for a few months (rentals maybe?), or otherwise electrical grids also using them, is cost effective for them. So overall, cost of mining is becoming increasingly complex and thus effects hash rate. I see the argument that if say rented miners weren't at a dam over winter, they'd simply be somewhere else, but this doesn't factor in overall business expansion from these modern "mobile" mining companies.
if you can calculate their costs of a 6 month run based on last 6 months average hashrate. and last 6 months of coin rewards. then you can calculate the cost per coin baseline. which should 'step up' incremental each 6-24 months rather then pander to the daily whims
above that base line. then you can start to build more short term indicators
but one thing is for sure $19,20,21 is not a base line.. the line is still near the 15 zone
I also don't think the average mining cost can accurately factor in renewable energy, as that's not really possible to easy to calculate, nor cost for hydrodams and electrical grids contribution, offsetting storage costs vs loss of gov subsidiaries vs operating costs and the rest. It seems to be a simplistic calculation based on average cost of electricity (globally) vs the average mining reward based on (total) hash rate . Clearly, it's not possible to calculate based on the wide range of different electrical costs in different countries based on what hash rate is contributed from which country with x cost, as this data isn't widely available.
It's an extremely vague estimate that probably only applies to the hobby miners you refer to, if they were paying an average electrical cost with an averagely efficient miner - not much else. But whether this estimate is rising or falling is relevant, as for the "part-time miners" that I referenced, increased value of block rewards is an incentive to start mining, continue or reduce mining...
.. right now the most impacting indicator this week is the futures market. where there are alot of deals not wanting to see the price go above $21k before their futures contracts expire.. hence the artificial flatline on the spot market resisting prices going above $21k this week due to whale day trading/arbitraging to keep it down to ensure their larger futures contracts win.
Not convinced about the artificial resistance around $21K. $21.5K is the previous macro lower high for many, so breaking through there would be an argument for the end of the bear market (even if others would claim that $25K remains a lower macro high). The futures market is likely respectful of this macro resistance for various reasons, rather than spot market respecting futures.