The On-Chain tab of the altcoin scorecard opens with a health score of 84, labelled Very Strong, and a line saying it is based on hash rate, active addresses, transaction volume and miner revenue trends. Underneath are four bars. Hash Rate reads 100. Tx Volume reads 100. Miner Rev. reads 100. Active Addr. reads 36.
Sit with that for a second, because it contains the whole lesson of this piece. Three of the four inputs to the score are at the top of their scale. Whatever happens to them next week, they cannot go higher. The only bar in the panel with room to move in both directions is the one reading 36. So of four metrics feeding a headline number, exactly one is currently capable of changing that number in either direction, and it is not one of the three that people quote.
One test, applied to every number on the screen
The test is a single question and it is deliberately harsh. If this number moved significantly tomorrow, would I hold a different amount of crypto than I hold today?
Not "would it be interesting". Not "would it tell me something about the network". Would it change the size of your position. If the answer is no, the number is not part of your process, whatever else it is, and every minute you spend looking at it is a minute you are not spending on something that is.
Most on-chain readouts fail this test for most people, and they fail it for an honest reason rather than a stupid one. They are measurements of a network's operation, and you are not making a decision about a network's operation. You are making a decision about how much of a volatile asset to own. Those two things are connected, but loosely and with long and variable lags, and a metric has to survive that gap before it earns a place in a decision.
A bar already at 100 cannot change your mind
This is a point about resolution, not about truth. Hash rate at the top of its range is a real fact and a good one. What it is not is a source of new information, because a metric sitting on its ceiling has almost no capacity to surprise you upward and has to travel a long way before it surprises you downward.
The panel does not explain how a raw measurement becomes a 0 to 100 bar, so read the bar as a position within some normalised range rather than as a level with units. Whichever normalisation is being used, the practical consequence is the same. Three bars are compressed against the top of their scale and one is in the middle where it can actually vary.
There is a version of this that catches people out in the other direction. A pinned metric is not worthless, it is just a different instrument. It is an alarm rather than a dial. Hash rate at 100 tells you nothing today, and hash rate falling off 100 would tell you a great deal, because that only happens for a small number of reasons and all of them matter. So the correct treatment is not to delete it, it is to stop reading it weekly and set a condition under which you want to hear about it.
Keep, watch, ignore
Sort every readout on the tab into three piles.
Keep is the short list of numbers that have room to move and that map onto demand for the asset rather than onto the operation of the network. On this panel, Active Addr. at 36 is the only component that qualifies on the first criterion, which is a striking thing to be able to say about a four-input score.
Watch is for the pinned metrics. Do not check them on a schedule. Decide the level at which each one becomes news, write that level down, and until it happens treat the bar as background. That is the honest treatment of hash rate, transaction volume and miner revenue when all three are at the top of their range.
Ignore is for anything you have never once acted on. Be strict here and count backwards. If you have looked at a metric for six months and it has never once been the reason you bought, sold, or resized, it is decoration. Keeping it on your screen has a cost, which is that it produces a feeling of thoroughness that is not backed by a decision.
The one bar with room to move, and what it does not mean
Active Addr. at 36 while volume and revenue sit at 100 is the interesting reading on this screen, and it is worth being careful about what to conclude from it.
What it says structurally is that value moving through the network and the revenue that movement generates are near the top of their range while the count of distinct participating addresses is not. Activity is concentrated in fewer hands than the throughput alone would suggest. That is the observation.
What it does not say is why. Fewer addresses moving more value is consistent with larger and more professional participants dominating flow, which is one story. It is also consistent with a small number of automated actors generating most of the throughput, which is a completely different story with a completely different implication for whether the activity persists. The panel does not distinguish between them and neither can I from this screen. If you want to know which it is, the follow up is to look at the distribution of transaction sizes and the concentration of the top addresses, not at another aggregate.
One further caveat worth carrying. Two of the four inputs, hash rate and miner revenue, are quantities specific to proof of work chains. That means this health read is shaped around a mining network, and it sits inside a scorecard about altcoins, many of which do not have miners at all. It does not make the score wrong. It does mean you should not read it as a statement about the health of the tokens you actually hold.
Write your two lines before you open the tab again
Here is the thing to do this week, and it takes ten minutes with a text file.
Write down the two conditions under which you would reduce your alt exposure, in numbers, with the metric named. Then open the On-Chain tab and check whether the panel contains a number that could ever trigger either of them. If it does, you now have a reason to look at that number and a threshold that makes looking meaningful. If it does not, you have learned something more valuable, which is that this tab is context for you rather than an input, and you can read it once a month instead of once a day.
The failure mode this prevents is the common one. A panel of green readings produces a general feeling of confidence, that feeling gets converted into slightly larger positions without anybody writing down a reason, and when the market turns there is no documented condition to check because there never was one. A score of 84 labelled Very Strong is precisely the kind of reading that does that, and the four bars underneath it are the reason to be sceptical of the headline rather than reassured by it.