Keep the Working Shortlist Small: A Desk Checklist on a Normal UTC Day
A desk checklist for a small crypto shortlist on ranks 51-200: sort Breakout Score™ after the UTC snapshot, require Climb and confidence, and cap every slot.
Updated
A shortlist rarely gets too big in one bad decision. It grows by small yeses: one more star because the chart looked clean, one more alert because the name was close, one more row in the agent universe because it was already on the screen. By Friday the list is a collection, and nobody can say which names the desk actually read. This post is the checklist for the other outcome. It treats the size of the working shortlist as a number you write down before the board opens, then defend on an ordinary UTC day when nothing in the band is forcing your hand.
Key Takeaways
- The working shortlist is the set of names you will read in full tomorrow, so its size is set by your revisit time, not by how many rows look interesting.
- After the UTC snapshot, sort ranks 51-200 by Breakout Score™, then require Climb and confidence to agree before any name gets a star or an agent rule.
- Stars, alerts, and the agent universe all draw on one cap. If each has its own allowance, the real working set is the sum of all three.
- On a normal day, empty slots are fine. A cap is a ceiling, not a quota.
- Board first, agent second: the Breakflare Trading agent (included with Plus, paper mode by default) should inherit a cap you already honor by hand.
- Breakflare is an attention tracker. Core research does not place trades or give price targets.
What job does a small working shortlist do?
It protects the depth of your reading. Breakflare scores CoinGecko ranks 51-200 once per day: hype score for attention level, 3d and 7d momentum for acceleration, Climb for rank movement, Breakout Score™ for the sort, and confidence for coverage and history. Those columns only help if someone reads them together for each finalist. A short list is the budget that makes that reading possible.
Think of it as a promise to your future self. Every name you keep is a name you owe a second look tomorrow, and a third look when its revisit date arrives. If you keep twelve, you will honestly re-read four and skim the rest while telling yourself the full set is under review. The cap is what stops that gap between what the list claims and what the desk really does.
The size limit also keeps the list useful as evidence. A short list with one written reason per name can be graded next week. A long list cannot, because half of it has no reason attached. For the research-quality argument behind a cap, the shortlist size limit post makes the case with slot budgets. This post is narrower: it is the desk routine that enforces whatever number you chose.
How does the cap work on a normal UTC day?
A normal UTC day is one where no sector dominates the band, Bitcoin dominance is not making headlines, and mention volume looks ordinary across mid-caps. That is the day you will run most often, and it is the day the cap is hardest to keep, because nothing external justifies a long list yet the board still has plenty of rows that look half interesting.
The day starts after the UTC snapshot, which is when the whole band is scored on the same cross-section. If you want the batch order itself, see how the daily UTC pipeline is ordered. From there the cap shows up at three moments.
Before the sort, you write the day’s number at the top of your note. Pick it from your capacity, not from the board. If a proper finalist read takes you a couple of minutes per name and you have a short window, that arithmetic is your cap. The two minutes is an example to adjust, not a Breakflare figure.
During the finalist read, the cap acts as a stopping rule. You do not keep reading down the Breakout Score™ sort until something excites you. You read the slice you can finish, decide, and stop.
At the close, you count. If the count is over the number you wrote, something leaves before the note is saved. That one rule, count at the close, does more than any amount of good intention in the morning.
What is the desk checklist, step by step?
Run these in order. The sequence matters because each step removes names, and the cap only holds if the early steps do real work.
- Confirm the board header shows today’s UTC date, then write the day’s slot count.
- Sort ranks 51-200 by Breakout Score™ and take only the top slice you can honestly finish.
- Read 3d against 7d momentum beside the hype score. A high level with a fading 3d reading is a skip candidate, not a finalist.
- Require Climb. If rank has not moved, the story is attention without confirmation, and it belongs in a note, not in a slot.
- Apply your confidence floor. Thin coverage ends the conversation for today, whatever the other columns say.
- For survivors, write one line per name: which columns agreed and which floor it cleared.
- Fill slots from the best agreement downward. Stop when the cap is hit, even if a seventh row looks fine.
- Park overflow in a dated note with an alert at most. No star, no agent rule.
- Close the board. New ideas wait for the next snapshot.
The order mirrors the daily rank 51-200 watchlist workflow, with the cap pinned to steps one and seven. If you want a broader view of how a full UTC day is laid out around the snapshot, the desk checklist for planning around the UTC snapshot covers the clock side of the same routine.
How do stars, alerts, and agent rules share one cap?
They should share it entirely. This is the part most desks miss. A star feels like a commitment, so people cap stars. An alert feels weightless, so people add dozens. An agent universe feels like configuration, so nobody counts it. Each surface adds attention, and attention is the thing the cap is meant to ration.
Use one ledger with three columns and one total.
| Surface | What it costs you | Rule on a normal day |
|---|---|---|
| Star | A daily re-read and a written reason | Counts against the cap, one in and one out at the limit |
| Alert | A future interruption that you must triage | Overflow only, with a dated note, and counts as half a slot in your own bookkeeping if you keep it past the revisit date |
| Agent universe | A paper rule that inherits the name | Only names that are already starred and still clear your floors |
The half-slot idea is a personal accounting choice, not a product rule. Pick any weighting you will actually follow. The point is that alerts are not exempt, because an alert that fires on an unstarred name creates the same pull a star would.
Starring itself has its own discipline, covered in starring names and pruning weekly. The cap is the number; starring is what each slot should mean.
What fail modes should the desk name out loud?
Naming them in advance makes each cheaper to catch. These are the ones that tend to show up on mid-cap desks.
The quiet-day fill. Three strong names clear the checklist and the cap is five, so two weak ones get promoted to feel complete. This is the most common failure, and it is invisible because the list looks the same size as yesterday. Fix: write “none” in the empty slots instead of filling them.
The soft overflow. Overflow names go to alerts “just in case,” then never get a revisit date. Within a week the alert list is the real shortlist. Fix: every parked name carries a date, and an undated one is deleted at the close.
The split cap. A meme list, an infrastructure list, and a main list each have their own small cap, so the combined working set is triple the number you wrote. Fix: one shared total across every list.
The agent inheriting the archive. The paper agent’s universe quietly equals everything starred since the summer. Fix: sync the agent universe to the current working list only, and read why low confidence is a skip before widening anything.
The loud-day suspension. The cap is dropped whenever the timeline gets noisy, which is exactly when collection pressure is highest. Fix: on loud days raise the agreement you require, not the number of slots.
How do you turn the cap into a written habit?
Write it once in numbers and nouns, then change it weekly at most. A useful written version answers five questions. What is the normal-day slot count? What is the confidence floor? What is the Climb minimum? What leaves first when a new name arrives at the limit? Who can change any of these? If you work alone, the last question still matters, because it stops you from bending the rule for a coin you already want.
Keep the note short enough to read in under a minute. A page of principles will not get read on a busy morning, and a note you do not read cannot cap anything.
Review it on a fixed day. The Monday reset is a natural place to compare last week’s average slot count with the number you wrote, and the prune rules give you the drop criteria for names that overstayed. If your average count keeps sitting at the ceiling, the cap is not biting, or your confidence floor is too soft to remove anyone.
Then add the agent, second. Breakflare Trading agent is included with Plus and runs in paper mode by default. Paper is where you test whether your written cap, applied mechanically, produces the list you would have chosen by hand. If the agent would hold names you would have dropped, the rule is looser than the habit, and the rule is what needs fixing. The score floor rules post covers how to set floors, and the paper mode walkthrough shows where to start. Live mode uses your own Binance trade-only keys and pay-as-you-go credits at 0.5% of fill notional, and it should wait until paper behavior is dull.
Where does Plus fit?
The columns are the same regardless of plan. Hype, momentum, Climb, Breakout Score™, and confidence mean the same thing wherever you read them. What Plus adds is the live ranks board, whales, trackers, alerts, member chat, 90-day pick history, and the paper Trading agent. That makes a capped list easier to run on current data and easier to audit against your own history.
If you want to run this checklist on the live rank 51-200 board with alerts and the paper agent, you can start a 14-day Plus trial. Plus is $399/mo or $3,990/yr. The research core stays an attention tracker, with no trades placed and no price targets.
Frequently asked questions
Is keeping the working shortlist small a trade signal?
No. It is a rule about how many names you carry into detailed reading. A small list tells you where your attention is committed, not that any coin is likely to move. Breakflare scores attention, and the cap only makes sure you read those scores properly.
What changes on a normal UTC day?
Very little, by design. A normal day has no dominant sector and no dominance shock, so the checklist order stays fixed and the slot count does the work. You will often finish with empty slots, and that is a correct result. On a loud day you tighten agreement requirements rather than loosening the cap.
Should the Trading agent use this as a hard skip?
It should mirror it. If a name falls outside your working list or fails your confidence floor by hand, the paper agent should skip it too. Match the floors in paper mode first, then decide whether a hard skip rule adds anything. Size caps, stops, and the kill switch belong in the same written note.
Does Plus change the columns involved?
No. The same five columns apply. Plus changes how you reach them: live ranks, alerts, whales, trackers, member chat, 90-day pick history, and the paper Trading agent around the same score set.
What if two names tie for the last slot?
Pick the one with the higher confidence, since confidence is the column that tells you how much to trust the others. If they still tie, take the one with the longer written reason, and park the other with a revisit date.
Related reading
Start researching today's shortlist
Pay $1 for 14 days: today's live board, whales, and alerts so you can invest from research, not FOMO. Then $399/mo. Or Core at $99/mo. Cancel anytime.