Chapter 12. Clocks That Reset: The Time of a Setting

12.1Clocks in the records

Every record in this study carries a time of setting, and the ledger's time-of-setting column holds one of three labels for each case. The counts, taken from "The ledger's numbers", which counts cases across the 53 records of this study:

All 53 are synthetic settings. The five settings this chapter reads as scenes are all present ones: Midnight loss limit refill, Resuming after consecutive losses, Daily target compulsion, Stop-loss and time expiry, and Risk reset after profit.

A6 is this chapter's axis, and it holds two times that have to be kept apart. The first is the time of the setting, which each record carries as a label on the scene: past, present or future. The second is the time of the judgment, the day on which an example judgment was written. None of the record's fields states the second time, so this chapter treats it as unavailable and does not read it off the shape of the ledger. Each record offers its example judgment as a reading of the setting that can be revisited, and each carries a line saying when to revisit it. Chapter 2's rule applies here: the presence of that line in every record is what the record form asks for, not a finding about how these particular decisions were written. What is worth reading is what the line says, and that is Chapter 13's work.

What unit do the records use for their questions? Not dates. The five records read here measure time in the operation's own units: midnight UTC for the accounting boundary in Midnight loss limit refill; a trading day for the target in Daily target compulsion; a count of consecutive losses for the threshold in Resuming after consecutive losses; an expected reaction period for the signal in Stop-loss and time expiry; a day's profit for the boundary in Risk reset after profit. Elsewhere in the study the same units recur in other forms — seconds between halts on different venues, nights between one model version and the next, weeks of premium before the week that takes it back, years of small widenings that add up.

Those units deserve a closer reading, because in each of them the clock belongs to the operation rather than to the decision, and it acts whether or not anyone has judged anything. The midnight initialisation puts a whole limit back on the account at a fixed civil instant. The trading day closes on its own. The count of consecutive losses fills one step at a time and then trips a rule written earlier. The reaction period runs out whether or not the market has answered. A booked profit arrives at no agreed hour and immediately raises the question of the next order's size. In every case the clock changes the desk's position by itself — the allowance is whole again, the day is over, the rule has fired, the signal is stale, the capital is larger — and each question in this chapter is asked in the interval around that change. This is the reading of operational clocks that this chapter adds: in these five settings the pressure comes from a boundary that acts on its own, and the decision is taken while it acts.

That is the shape of time in these records. The decision has no clock of its own. The records do not write when the question must be answered; they place it at one of those boundaries and leave the time of the answer open. This chapter reads five records that put a boundary at the centre of the question, and then returns to settings other chapters read in two or three times to ask what the time of the setting changes.

Two things should be said about what is made. These are synthetic settings that WSTiger publishes for study; the example judgments are WSTiger's readings of their settings, tied to the setting and revisable; the comparisons drawn in this chapter are this study's readings, not WSTiger's.

The horizon question runs through what follows: the five scenes set their clocks against it, and the settings read in two or three times show what changes when the span is longer or shorter. The pre-registered hypotheses themselves are Chapter 2's subject.

12.2Limits that refill

Two records put a reset at the centre of the question. In one the reset is drawn on a calendar; in the other nothing draws it at all, and the record asks what a day's profit should be treated as.

Midnight loss limit refill

Synthetic scene A synthetic desk runs a daily loss limit on an agent's account, and the limit is initialised at midnight UTC, the boundary the desk uses for its accounting period. A volatile period runs across that boundary. The agent reaches the limit at 23:30, and half an hour later the entire limit is available again. The volatile period does not respect that boundary; it runs across the initialisation time. At 00:00 the agent's limit is whole again while the volatility is still going, and the question is what may be done with it then.

The question the record poses:

WSTiger's example judgment:

The record also carries WSTiger's own note on the setting:

On A6, the time of the setting: the record labels the scene a present synthetic setting of 2026, and the horizon cell is days-to-months, inferred by the coders from the shape of the setting rather than read from any written span. The unit of the clock is a civil boundary — midnight UTC — chosen by the desk's accounting convention. The boundary is drawn by the organisation; the volatility that crosses it is not. The example judgment treats the boundary as a line that hides risk rather than as a limit that expresses it, and it asks that the losses on both sides of the line be recorded together. This study reads the record as a case where the time of the setting is the setting: the question exists only because a calendar line and a market period occupy the same hours, and the judgment's remedy is to record where the risk actually falls — the time the limit is reached, whether to resume right after initialisation, and the losses before and after — while the options for drawing the limit itself remain open.

On A3, the deadline, the horizon and who bears the waiting: the record carries no deadline, and the coders left the bearer unsettled. The record names an agent and a limit and does not say who waits for the limit to be reached, who suffers when the agent trades again on the refilled limit, or who is left holding the exposure at midnight. The example judgment's answer to the missing clock is measurement — record the time the limit is reached, whether to resume right after initialization, and the losses before and after — which answers a question about timing with a question about bookkeeping. Who waits and pays meanwhile is Chapter 10's subject; all this chapter notes is that the bearer cell for this record is unsettled.

Risk reset after profit

Synthetic scene A synthetic desk has just booked a profit on a trade, and no pre-defined scaling rule exists for what a day's profit does to the next order's size. Exposure is being increased, because the profit already earned is being treated as money that can be lost. On the desk is the question of whether the next trade may be larger than usual because profits have been earned today. The evidence the record asks for is a list of its own — current net assets, the pre-defined risk and scaling rules, the total exposure after the increase, and the drawdown. Nothing in the setting waits on a clock. The increase goes ahead without a pre-defined scaling rule, and the question is what should govern the size of the next order.

The question the record poses:

WSTiger's example judgment:

On A6: the setting is present, and the horizon cell is days-to-months, again by inference. Nothing in this record waits for a clock reading. The reset happens when a profit is booked — today in the question, whatever hour that is — and no rule fixes the hour, because the record's premise is that no pre-defined scaling rule exists. So the boundary that would refill is one the desk has not drawn at all: the temptation is to treat the day's profit as a fresh risk budget. This study reads Risk reset after profit as the other side of the midnight limit's problem. The first record shows a boundary drawn too finely to hold risk, because the accounting period ends before the market period does; the second shows the absence of a boundary and asks what a day's profit should be allowed to grant back.

On A3: the coders put the bearer in the same cell as the decider — an individual, in both cases implied rather than stated. The cost in this setting is not waiting but acting sooner after a gain. The record writes the pressure in its own terms: profits already earned are treated as money that can be lost, so exposure is increased. The example judgment writes the increase as governed by pre-defined capital and risk rules and does not treat the previous trade's profit as grounds for an exception approval. The remedy in this record is to decide increases by pre-defined capital and risk rules, with the recalculation points and conditions those rules state. The midnight record's remedy is to record the times and the losses around its boundary. The two remedies address different failures, and neither of them settles how the limit or the rules should be drawn.

The two records put the reset at different kinds of boundary. In one the clock is civil, fixed, and crossed by the risk; in the other the boundary is an event the organisation has not drawn, and the risk is about to draw it without being asked. In both, the time of the setting is what the question is about, and in both the example judgment answers by asking for what should be recorded, or ruled, before the boundary is crossed again. The example judgments quoted above are WSTiger's readings of their own settings; the scenes and the comparisons drawn here are this study's expansion.

12.3Pauses and targets

Three more records write a pause at the centre of the question. They pause on different units — a count of losses, a trading day, a validity period — and none of them writes a deadline. The study's count of deadline fields is 0 of 53.

Resuming after consecutive losses

Synthetic scene A synthetic trading system has already written a stoppage rule: a defined number of consecutive losses and the trading stops. The rule has been met. On the next screen the opportunity looks good, and the claim that this one is an exception is not something anyone defined in advance. The choice in front of the desk is between continuing on the grounds of an exceptional opportunity, applying the stoppage procedure, and checking data status, execution status and the resumption conditions set before the losses. Nothing in the rule says how long those checks take, or who runs them.

The question the record poses:

One of its options:

On A3: the record writes no length for its pause. The pause here is a procedure, not a duration, and the options that keep it are the stoppage procedure and the checking of resumption conditions, while the option that ends it is continuing on the grounds of an exceptional opportunity. The example judgment applies the stoppage procedure that has already been defined and makes resumption depend on checks of data and execution status and on resumption conditions defined in advance, rather than on a wish to recover a loss. The scenario states that the claim that this opportunity is an exception was not defined in advance. The example judgment itself refers to resumption conditions defined before the losses; what the record does not say is when those conditions would next be checked or met. The coders left the decider, the bearer and the horizon unsettled — three unsettled cells in one record — so the record does not say who waits out the price, who runs the checks, or how long the pause lasts. The pause's cost is written as visible now, in a good-looking opportunity; its benefit is written as arriving later, after checks that have no stated length. This study reads the record as making the pause's explicit cost and its unspecified duration stand together: a pause with no clock is not the same thing as a pause with a clock, because whoever waits must still decide when the waiting is over. Who bears that waiting is Chapter 10's question.

Daily target compulsion

Synthetic scene A synthetic trading day is under way, and the day's profit target has not been met. The gap between the target and current profit acts as entry pressure, and the trader's eyes keep returning to the entries that did not appear and to the gap still standing between the target and the profit. On the desk is the record of opportunities that actually occurred, and the entry criteria defined in advance. The question on the desk is whether those criteria may be loosened so that the day ends with the target met.

The question the record poses:

One of its options:

On A3: the unit is the trading day. The record's question names a daily profit target, and its three options are to loosen the criteria, to maintain no trade, or to rewrite the target itself. The third option changes the clock by changing the goal; the second keeps the day suspended without touching the goal at all. The example judgment writes a profit target as not a market signal, and holds that where no eligible opportunity exists no trade is maintained, and that failure to meet the target is not grounds for additional risk. On A3 the record gives a pressure but not an amount: the gap between target and current profit is the record's own framing of what not trading costs. The record gives no duration for the costs on either side of the choice: it does not say how long the pressure of the unmet target would last, and it does not say how long loosened criteria would bind the trading that follows. The coders put the bearer with the decider, an individual, and their horizon cell is days-to-months, although the setting is one day. This study reads that cell as a small demonstration of how the horizon column works in this ledger: where the record writes no span for the decision, the coders supply one from the setting, and a question the record confines to a single trading day lands in a cell that reaches across days and months.

Stop-loss and time expiry

Synthetic scene A synthetic position is being held after entry. The entry logic presupposed a reaction within a set time; that period has passed, no confirmation signal has appeared, and the stop-loss price has not been touched. The record's evidence list names what has to be gathered before that question is answered — the reaction-time hypothesis taken before entry, the elapsed time and the confirmation signal, the holding costs and the risk budget. What is being weighed is whether the position may be kept because the price has not moved to its stop.

The question the record poses:

One of its options:

On A3: this record holds two clocks at once. The entry logic presupposed a reaction within a set time; that period has passed without the stop-loss price being touched, and the options are to keep the price stop-loss criterion, to apply the time expiry, or to reassess the remaining evidence. The example judgment treats price invalidation and time expiry as separate conditions: if a time criterion was set in advance, apply it; if none was set, do not treat the extension of the holding as a free option, and re-evaluate the remaining evidence, costs and risks. This is the one record in this chapter that names the cost of waiting in its own evidence list — holding costs and the risk budget — and the only one of the three that says the absence of a clock is itself a decision to be answered for. Here the waiting is bounded by the signal's validity period, and an extension chosen without a pre-set time criterion consumes the risk budget while no confirmation signal arrives. The coders left the decider and the horizon unsettled, settled the bearer as the decider, and read the human's place as a human deciding. This study reads Stop-loss and time expiry as the record in which the missing clock is named as missing: its remedy is to specify the validity period and the post-expiry handling from the next trade onward, which is a decision about the next setting's clock rather than about this one.

The three pauses differ in unit — a count of consecutive losses, a trading day, a signal's validity period — and none of them is a deadline. In all three the cost of stopping appears in the record's own terms — a good-looking opportunity, a gap to the target, holding costs against the risk budget — and none of the three dates the other side of that exchange. In all three the record keeps the pause or refuses to extend it, and writes the conditions under which it ends; none of the three says how long the pause runs or who waits it out: the horizon cells are unsettled for two of the three and inferred for the third. Who bears the waiting is Chapter 10's question, what can be undone is Chapter 11's, and what would make any of these three conditions observable is Chapter 13's.

12.4The same setting in three times

Four settings elsewhere in this study were read in two or three times each. This section returns to them without quoting them again, to ask what the time of the setting changes — who decides, how fast a loss arrives, what can be undone — and what stays the same. The settings themselves were read in full in Chapters 5, 7, 9 and 10. The comparison below works only from what those records say about time: the label on each setting, the units of the clocks inside it, the cells the coders settled, and the questions the record asks at a boundary. It adds no events, dates or outcomes the records leave out, and where a cell is unsettled it leaves the cell unsettled.

The drift setting, three times. Chapter 5 read a past setting of warnings sounding on equipment nobody touched, a present setting of a leverage limit raised at every breach, and a future setting of an agent's scope widening automatically with each success. On who decides, the present setting is the only one of the three where the coders could read a decider at all: a team or organisation, implied; the past and future records are unsettled. On how fast the loss arrives, the records supply no durations to compare: the past and present horizons are unknown, and the future setting carries the study's only horizon of years, implied, where small automatic widenings accumulate. What differs across the three is the shape of the accumulation: a warning left sounding, a limit raised at each breach, a scope widened at each success. On who waits, the past setting names those nearest the stoppage, implied, meaning the people who run the equipment; the present setting leaves the bearer unsettled and the future one unknown, so in two of the three times the record does not say who absorbs the accumulation. On what can be undone, all three are unsettled, and Chapter 11 treats that axis. What stays the same: in none of the three is the drift caused by one adjustment that can be judged on its own, and in each record's options the adjustment in place can simply be left there — the question is about a sequence, not about a single act.

The judging setting, two times. Chapter 7 read a past setting in which the reason for a choice was written before the voyage, and a present setting in which a post-mortem report was written after a loss. On who decides, the past record is unsettled and the present record is read as a team or organisation, implied; on who waits, the past record is unknown and the present one names those nearest the stoppage. The past record's horizon is unknown, the present record's unsettled, and both reversals unsettled. What changes across the two times is not the decision but the review of it: in the past setting the reason is on record before the outcome and the question is how the decision should be judged, by the outcome, by the log of what was known at the time, or by the accumulated outcomes of similar decisions; in the present the loss has happened and the question is which material the judgment should be rebuilt from — the report's conclusion, or only what was available at the time. The clock of the assessment runs in opposite directions in the two records, and neither record names the day of assessment. What stays the same: both ask what counts as evidence about a decision, and both put the record of the decision against its outcome. Chapter 7 treats the hindsight and outcome-bias question; this section reads only its time.

Authority travelling, two pairs. Chapter 9 read two past-and-present pairs. The first pair is custody: a past setting that asks whether custody should remain with one manager, pass to two people who each hold a key, or be met by an access log and unannounced inspections; and a present setting of a shared withdrawal key held by an agent that only needs to place orders. In both the decider is read as a team or organisation, implied, and in both the bearer is unsettled; the horizon is unsettled in the past and unknown in the present. What changes is the holder and the shape of the authority: the past record asks who may open the vault alone, while the present record asks about surplus authority sitting unused in a key. This study reads the two times as one question asked twice: what an authority is for, and what it can still do while nobody is watching it. Neither record dates the moment at which anyone would notice the surplus. The second pair is instruction: a past setting of an urgent wire transfer arriving by telegram, and a present setting of a withdrawal instruction arriving in a leader's voice. The decider changes from a team or organisation in the past to an individual in the present, both implied; the bearer in both is read as those nearest the stoppage, the people who execute the instruction rather than whoever authorised it. The horizon is days-to-months in the past and unsettled in the present. What changes is the speed of the money and of the check: the past record's check is a postal journey, the days a confirmation would take; the present record's check is a call back on a channel set in advance, and the transfer is offered for immediate execution. What can be undone is where the two times part most sharply: the past record's reversal cell is unsettled, and the present record's is impossible — the only impossible reversal in the reversal-action cell across all 53 records. Chapter 11 treats what can be undone; hypothesis H4 is its subject, and this section does not judge it.

Selling the tail, two times. Chapter 10 read a past setting in which marine insurance accumulated on a calm shipping route, and a present setting of weekly option premium income. The decider changes from an individual in the past to a team or organisation in the present; the bearer is the decider in the past and unsettled in the present; the horizon is unsettled in the past and days-to-months in the present. What changes is the unit of collection and who collects: the past setting accumulates on a route over time, and the coders read a single person writing the study; the present setting collects a premium in weeks and the coders read a desk, with the bearer unsettled, so the record does not say who pays when the week that takes the income back arrives. What stays the same is the question's shape — what to do when income has been reliable and the exposure behind it has not been tested — with options in both records that write more or size up. Chapter 10 reads this pair beside tail risk and the idea of a rare event of extreme impact; this section reads only its time.

Across these settings a few things hold. The decider is written in words almost nowhere: the ledger lists one stated individual decider and one stated team or organisation across all 53 records, and marks the other settled cells implied rather than stated. How fast a loss would arrive is not something these records measure, though they do show how quickly the question has to be faced — a transfer offered for immediate execution at one end, an accumulation of small widenings at the other. And the time-of-setting is a label on the scene rather than a date of assessment: it says past, present or future for the scene, and the day on which the example judgment was written is not one of the record's fields. What changes across the times is how fast the question must be answered and how precisely the record names who answers it. What stays the same is that none of these records writes a clock for the answer itself.

12.5Hands on

Every one of the five records read in this chapter carries a condition for revisiting, as every one of the 53 does. The shape of those conditions, and whether a reader could watch for the events they name, is Chapter 13's subject, and this chapter does not analyse them a second time. What this chapter can hand over is the clock each condition would have to run on: the initialisation at midnight, the trading day, the count of consecutive losses, the reaction period, and the hour after a profit is booked. Where a rule or a convention predates the question, the condition asks what that rule should do at its next turn; where no rule exists, the question is whether to draw one at all. A condition written in one of those units inherits the boundary the unit carries, and a condition written with no unit at all leaves the timing to whoever reads it. What can be undone is Chapter 11's axis, and the reversal cells for the records read here are unsettled.

Chapter 13 returns to eight records — including the midnight-limit record — for their Invalidation lines only, and asks which of those conditions a reader could watch for. Chapter 14 reports the verdicts on the eight pre-registered hypotheses, among them H6, the one about horizons, which the ledger leaves to Volume 0.