A reader opens the trust desk's feed for the first time twelve minutes before a Wednesday weeknight contest locks. The reader sees three notes on the same fixture, a probable XI change, a captain-pool read, and a deck report. The reader has not opened a trust desk note before, has not seen a corrections trail, has not named a source the desk uses, and has not built a habit for filing a note the reader has finished. The reader's first read on the first note is the read the reader will repeat on every later note, and the read the reader repeats is the read the protocol below is built to shape. The protocol is the read the desk recommends on the first note the reader opens, and the protocol is published here because the desk can publish the notes but cannot publish the read on the reader's behalf. Editorial process notes are listed on the trust desk; the protocol below is the reader's first-time half of the same process.
Working line for the reader who wants the short version before reading the longer piece. The first-time reader protocol is seven steps: open the corrections trail first, confirm the source is named, check the timestamp, read the body once top to bottom, name the type, decide whether the note changes the locked XI, and file the note where the reader can find it again. The protocol runs in under three minutes on a single first-time read, and the order is the part that compounds across the season because the order is what the reader will repeat on every later note the desk publishes.
The reader who follows the protocol on the first note will arrive at the second note with a habit the reader has already practiced once. The reader who skips the protocol and reads the body first will arrive at the second note with a habit the reader has to rebuild. The desk treats the first note as the reader's training note, and the training note is the note the protocol is built to walk the reader through. The note the reader opens next is the note the reader has to read on the reader's own.
Step one: open the corrections trail before the body
The first line on a trust desk note is a corrections trail. The corrections trail names the previous version of the note, the source that triggered the correction, and the timestamp of the correction. A first-time reader opens the corrections trail before opening the body because the corrections trail is the line that decides whether the body is the original read or a corrected read.
The corrections trail is one to three short lines that sit at the top of every trust desk note the desk has filed. A note without a corrections trail is the original version of the note; a note with a corrections trail is the desk's later version of a note the desk has already filed. The reader who opens the corrections trail first learns on the first read whether the body is the version the reader is meant to read or the version the reader has to weigh against a previous version the reader has not read. The first-time reader protocol opens the corrections trail first because the corrections trail is the line the body has to be read against.
A first-time reader who skips the corrections trail and reads the body first will read a corrected note as if the corrected note is the original note. The corrected note will look like an original note the desk is publishing for the first time, and the body will read as the desk's first read on the fixture. The reader has missed the seam between the original version and the corrected version, and the seam is the seam the corrections trail is meant to make visible. The reader who misses the seam reads a note the desk has corrected as a note the desk is publishing for the first time, and the read is the read the protocol is built to prevent.
The corrections trail is not a footnote. The corrections trail is the operating input for every later step in the protocol. The body of the note is the read the reader applies the corrections trail to, and the corrections trail is the line the reader applies the body to. The reader who reads the corrections trail first reads the body against the corrections trail, and the reader who reads the body first has nothing to read the corrections trail against. Step one is the cheapest step in the protocol, and the cheapest step is the step that compounds across the season.
Step two: confirm the source is named on the note
A trust desk note names the source the desk has used for every claim the desk has made in the body. The source can be a named account, a named feed, a named broadcaster, a named publication, or a named official statement. A first-time reader confirms the source is named on the first read because a note without a named source is a note the desk treats as advisory and the protocol treats as hedge-by-default.
The named source is the line the reader checks the desk's claim against. A captain pool note that names the source as a public pick pool at a named time is a note the reader can re-check on the same pick pool at a later time. A pitch read note that names the source as the broadcaster's pitch report at a named time is a note the reader can re-check on the same broadcaster's feed. A corrections trail note that names the source as the away captain's statement at a named time is a note the reader can re-check on the same statement at a later time. The named source is the only line on the note the reader can re-check, and the named source is the only line the protocol treats as worth re-checking.
A note that does not name a source is a note the reader cannot re-check. The reader has to take the note on the desk's word, and the reader has no line on the note the reader can re-check the desk's word against. The protocol treats every unnamed-source note as a hedge note by default, and the hedge-by-default treatment is the line that protects the locked XI from a note the reader has no second read on. The reader who confirms the source on the first read can re-check the source on every later read, and the reader who skips the source confirmation has nothing to re-check.
The source the desk names is not the same as the source the reader would have used. The desk uses a small set of standing sources the desk has already named on the desk's about page, and the small set is the set the reader is meant to re-check against. The reader who wants to re-check the source against a different source has crossed out of the trust desk feed and into the reader's own research, and the cross is a cross the protocol does not try to make. The protocol treats the named source as the only source the reader has to re-check against, and the named source is the only source the desk commits to.
Step three: check the timestamp the desk has named
A trust desk note names the timestamp of every claim the desk has made in the body. The timestamp is the time the desk pulled the source, the time the desk filed the corrections trail, or the time the desk confirmed the change. A first-time reader checks the timestamp on the first read because the timestamp is the line that decides whether the note is fresh or stale.
The timestamp is the line the reader applies the body to. A probable XI note timestamped ten minutes ago is a fresh note the reader can act on; a probable XI note timestamped ninety minutes ago is a note the reader has to weigh against a fresher note the desk has not yet filed. A captain pool note timestamped two hours ago is a fresh note the reader can read on a four-hour-to-lock contest; a captain pool note timestamped ninety minutes before lock is a note the reader has to weigh against the lock screen. The timestamp is the line the reader applies the body to, and the timestamp is the line the reader applies the corrections trail to.
The reader who skips the timestamp check has read a note without knowing how fresh the note is. The body of the note reads the same whether the timestamp is fresh or stale, and the read the body produces is the read the reader has to apply to the lock screen. The protocol treats every stale-timestamp note as a note the reader has to weigh against the lock screen, and the lock-screen weigh is a weigh the reader has to do on every stale-timestamp read. The timestamp check is the step that protects the reader from a body the reader has read as fresh.
A note the desk has timestamped as part of a corrections trail carries two timestamps: the timestamp of the original read and the timestamp of the corrected read. The reader who has skipped step one and has not opened the corrections trail will not see the two timestamps on the first read. The reader who has run step one first will see both timestamps, will weigh both timestamps against the lock screen, and will name the corrections-trail timestamp as the timestamp the body is current to. The protocol treats the corrections-trail timestamp as the timestamp the reader applies to the body, and the corrections-trail timestamp is the timestamp the reader has to apply on every corrected note.
Step four: read the body once, top to bottom, without backtracking
The body of a trust desk note is one read, written to be read in a single pass. A first-time reader reads the body once, top to bottom, without backtracking, because the body is written as a sequence the desk has ordered the reader to read in. The reader who backtracks has read the body twice, and the second read is the read the desk has not written for.
The body of a trust desk note has a structure the desk has used on every note the desk has published: the lead paragraph names the claim, the second paragraph names the source, the third paragraph names the timestamp, the fourth paragraph names the fantasy implication, and the closing paragraph names the corrections-trail pointer. The reader who reads the body once top to bottom reads the body in the order the desk has ordered it. The reader who backtracks has read the lead paragraph twice, the second paragraph twice, and the corrections-trail pointer twice, and the twice-read paragraph is the paragraph the reader has started to weigh against the reader's own read.
The single-pass read is the read the desk has written for. The reader who reads the body once top to bottom takes the body the desk has written, weighs the body against the corrections trail the reader has read in step one, and arrives at step five with a body the reader has read in the order the desk has ordered. The reader who backtracks has read the body in a different order, has applied the corrections trail to a body the reader has read twice, and has arrived at step five with a body the reader has read in the order the reader has chosen. The two reads are different bodies, and the protocol treats the desk's order as the only order the reader should read the body in.
A first-time reader who has read the body once top to bottom is a reader who has time to apply steps five, six and seven. A first-time reader who has backtracked has spent the time the protocol has reserved for steps five, six and seven on a second read of the body, and the reader has arrived at step five with a body the reader has read twice and a locked XI the reader has not yet decided. The protocol treats the single-pass read as the only read the reader has time for on a first-time visit, and the single-pass read is the read that frees the reader to run the rest of the protocol.
Step five: name the type the note is on the reader's read
A trust desk note is one of three types on the reader's read: a signal note, a hedge note, or a signal-correction note. A first-time reader names the type on the first read because the type is the line the reader applies the locked XI to. A note without a named type is a note the protocol treats as hedge-by-default, and the hedge-by-default treatment is the line that protects the locked XI from a body the reader has not yet typed.
A signal note is a note the desk publishes as a single observation that should change a single call on the lock screen. A probable XI change, a confirmed toss, a confirmed pitch update, a confirmed weather change. A signal note has one anchor, one magnitude, one source. A hedge note is a note the desk publishes as a probability read on a call the reader has not yet locked. A captain-pool differential, a pitch-read advisory, a captain-pool read on a differential slot. A hedge note has one anchor, one magnitude range, one source, one probability attached to the magnitude range. A signal-correction note is a note the desk publishes as a corrected version of a signal note the desk has already filed. The corrections trail at the top of the note names the original line, the reason for the correction, the timestamp of the correction.
The reader names the type from the corrections trail. A note without a corrections trail is a signal note or a hedge note. A signal note names one anchor, one magnitude, one source. A hedge note names one anchor, one magnitude range, one source, one probability. A note with a corrections trail is a signal-correction note. The reader names the type on the first read of the corrections trail, before reading the body. The reader who names the type after reading the body has named the type through the body's framing of the same call, and the body's framing is the framing the body wants the reader to apply.
The type is the line the reader applies the locked XI to. A signal note changes a locked call the reader has already made. A hedge note leaves a locked call where the locked call is, and the reader files the hedge note as advisory for a later read. A signal-correction note patches a locked call with the corrected magnitude the corrections trail names. The reader who has not named the type has not named the action the protocol recommends, and the reader who has not named the action has not run step six on a type the protocol has confirmed.
Step six: decide whether the note changes the locked XI
A trust desk note either changes a locked call on the lock screen or it does not. A first-time reader decides whether the note changes the locked XI on the first read because the decision is the line the reader applies the body to. A note the reader has decided does not change the locked XI is a note the reader files in step seven as advisory.
A signal note changes a locked call when the locked call is the call the signal is pointing at, and the signal is fresher than the locked call. A hedge note changes a locked call only when the locked call is the call the hedge is pointing at, the hedge carries a probability range, and the probability range is wide enough to disturb the magnitude the reader has locked. A signal-correction note patches a locked call when the locked call is the call the original signal was pointing at, the corrections trail has named a corrected magnitude, and the corrected magnitude is the magnitude the reader has locked. The reader names the locked call on the first read of the note, names the type on step five, and applies the type to the locked call.
The reader who has run steps one through five has the type in mind. Step six is the step the reader applies the type to the locked call. The signal note the reader has named changes the locked call. The hedge note the reader has named leaves the locked call where the locked call is. The signal-correction note the reader has named patches the locked call with the corrected magnitude the corrections trail has named. The decision the reader makes on step six is the decision the reader applies the locked XI to, and the decision is the decision the reader files in step seven.
The reader who decides the note does not change the locked XI is a reader who files the note as advisory. The advisory file is the file the reader opens when the same call recurs on a later contest, and the advisory file is the file the reader weighs the next contest against. The reader who decides the note changes the locked XI is a reader who applies the change to the locked XI and then files the note in the lock-screen file. The lock-screen file is the file the reader opens when the reader has to defend the change to a reader the reader is contesting against. The protocol treats both files as files the reader has to maintain, and the maintenance is the maintenance step seven is built for.
Step seven: file the note where the reader can find it again
A trust desk note the reader has finished is a note the reader files in one of two places: the lock-screen file for notes the reader has applied to the locked XI, or the advisory file for notes the reader has left as advisory. A first-time reader files the note on the first read because the file is the line the reader opens the next time the same call recurs.
The lock-screen file is the file the reader opens when the reader has to defend a change to a locked XI. The advisory file is the file the reader opens when the reader has to weigh a later contest against the note the reader has already read. The protocol treats the lock-screen file as the file the reader maintains as the reader's own reading record, and the protocol treats the advisory file as the file the reader maintains as the reader's research record. The two files are not the same file, and the reader who files both into a single file has filed the reader's reading record and the reader's research record into a single file the reader cannot re-open on a single read.
The file the reader maintains is not a file the desk has access to. The desk publishes the note, and the reader files the note on the reader's own filing system. A first-time reader who does not have a filing system the reader has used before can adopt a simple filing system: a single bookmark folder named trust desk, a sub-folder named lock-screen, and a sub-folder named advisory. The reader drags the note the reader has applied to the locked XI into lock-screen, drags the note the reader has left as advisory into advisory, and the filing system is the system the reader maintains across the season.
The filing step is the step that compounds across the season. A reader who has filed every note the reader has read has a record the reader can re-open on every later contest the same call recurs. A reader who has not filed the note has a record the reader cannot re-open, and the record the reader cannot re-open is the record the reader has to rebuild on every later contest. The protocol treats the filing step as the most compounding step in the protocol, and the most compounding step is the step the first-time reader has to build a habit for on the first read.
A worked first-time read on a Wednesday weeknight note
The worked example below walks a first-time reader across a single trust desk note on a Wednesday weeknight fixture, from the first corrections-trail line to the filed note. The example is hypothetical, the example is meant to show how the seven steps land on a first-time read, and the example is runnable in the head on the next Wednesday weeknight.
The setup: a Wednesday evening IPL fixture at 7:30 p.m. The reader's contest locks at 7:25 p.m. The reader has not opened a trust desk note before, opens the feed at 6:50 p.m., and finds one note on the desk: a probable XI change on the home side. The reader is an anchor reader, has the XI partly built, has the captain pick locked on the in-form opener, and has at most one bench slot still open. The reader opens the note the reader has never opened before and runs the seven steps.
Step one, the corrections trail. The note has no corrections trail. The reader names the note as a signal note or a hedge note on the first read of the corrections-trail line. The body of the note names one anchor (the new opener), one magnitude (replacing the previous opener), one source (the home captain's pre-match press conference at 6:30 p.m.). The note is a signal note. The reader has named the type on the first read, before reading the rest of the body.
Step two, the source. The note names the home captain's pre-match press conference at 6:30 p.m. as the source the desk has used. The reader has named the source on the first read. The reader can re-check the source on the same press conference at a later time. Step three, the timestamp. The note names the timestamp of the source as 6:30 p.m. and the timestamp of the desk filing as 6:42 p.m. The reader names the timestamp on the first read, and the timestamp is fresh against a 7:25 p.m. lock.
Step four, the body. The reader reads the body once top to bottom: lead paragraph names the claim (opener change), second paragraph names the source (home captain's press conference), third paragraph names the fantasy implication (captain pool shifts, differential slot moves), fourth paragraph names the corrections-trail pointer (none). Step five, the type. The reader has named the type in step one, signal note. The type is the type the reader applies the locked XI to. Step six, the decision. The signal note is pointing at the captain pick. The captain pick is the in-form opener the reader has already locked. The reader has to decide whether to change the captain pick. The signal is fresher than the locked pick. The reader changes the captain pick to the new opener. Step seven, the file. The reader drags the note into the lock-screen file. The reader has finished the first-time read in under three minutes.
At 7:25 p.m. the contest locks. The reader's XI carries the new opener as captain, the differential slot the new opener has opened, and the rest of the XI the reader has not changed. The reader has applied the protocol to one note on a first-time read, and the protocol is the protocol the reader will repeat on every later note the desk publishes. The protocol is the reader's first-time read, and the first-time read is the read the protocol is built to shape.
Three questions first-time readers send in about the protocol
What does the first-time reader do when the corrections trail names a correction the reader cannot verify?
The first-time reader treats the unreadable correction as a hedge note. The reader opens the corrections trail, tries once to name the original line, and names the type hedge if the original line cannot be resolved on the first read. The reader then applies the hedge-by-default treatment to step six: the reader leaves the locked XI where the locked XI is and files the note in the advisory file. The hedge-by-default treatment is the line that protects the locked XI from a corrections trail the reader has not been able to verify, and the verification step can be re-run on a later read when the reader has more time to check the original line against the source.
Does the protocol change when the reader has already run the protocol on a different note?
The seven steps do not change. The order the reader runs the seven steps in does not change. The reader applies the protocol to the second note the same way the reader applied the protocol to the first note. The only thing that changes is the reader's filing habit: the reader who has filed the first note into the lock-screen file or the advisory file has already built the filing system the second note files into, and the filing system the second note files into is the same system the first note filed into. The protocol treats the second note as a note the reader runs the same seven steps on, and the seven steps are the seven steps the protocol has named.
What does the first-time reader do when the desk has not named a source on the note?
The reader treats the unnamed-source note as a hedge note. The reader runs step one (no corrections trail), runs step two (no named source), runs step three (timestamp may be named), runs step four (body), names the type hedge in step five, applies the hedge-by-default treatment in step six (locked XI untouched), and files the note in the advisory file in step seven. The hedge-by-default treatment is the line the protocol has named for every unnamed-source note, and the hedge-by-default treatment is the only line the protocol applies to a note the desk has not backed with a source the reader can re-check.
Where the protocol stops being a rule
The protocol is a rule for the first-time read the reader runs on a single trust desk note. The protocol is not a rule for the second read the reader runs on the same note, and the protocol is not a rule for the read the reader runs on a note the desk has already corrected with a signal-correction note the reader has not seen. The boundary below is the line the protocol keeps in front of the rule.
The first boundary is the note the reader has not opened. The protocol runs on notes the reader has opened on the lock screen and read past the corrections trail. The protocol does not run on notes the reader has not opened, on notes the reader has archived before reading, or on notes the reader has read past without naming a type. A note the reader has not opened is a note the protocol treats as not yet named, and a not-yet-named note is a note the protocol returns no decision on. The reader who applies the protocol to a note the reader has not opened has applied the protocol to a body the corrections trail has not confirmed, and the body the protocol has run on is a body the reader's bias has named.
The second boundary is the read the reader has run more than once. The protocol runs on the first read the reader has made on a single note. The protocol does not run on the second read the reader has made on the same note, on the read the reader has made on a note the desk has since corrected, or on the read the reader has made on a note the reader has filed in the advisory file and re-opened on a later contest. A second read is a read the reader has run on the reader's own, and the read the reader has run on the reader's own is a read the protocol does not cover. The protocol treats the second read as a read the reader has to build a new protocol for, and the protocol the reader has built for the second read is the protocol the desk does not have a position on.
The third boundary is the source the desk has not named. The protocol runs on notes that name a source the desk can verify. The protocol does not run on notes that do not name a source, on notes the desk has flagged as unverified, or on notes the desk has marked as superseded by a fresher note. The protocol treats every unnamed-source note as a hedge note by default, and the hedge note runs steps five, six and seven on a type the reader has named by default rather than by reading. The named source is the only source the protocol applies to, and the named source is the seam between the desk's commit and the reader's filing habit.
One habit the protocol recommends, one it does not
Two short paragraphs on the single first-time-reader habit the desk thinks will sharpen the next contest the most, and the one habit the protocol would not recommend. The frame is bounded: the protocol can publish seven steps for the first note the reader opens, but the protocol cannot run the seven steps on the reader's behalf, and the seven steps are the steps the locked XI owns.
The habit the protocol recommends is the filing-system habit. The first-time reader builds a filing system the reader will maintain across the season: a single bookmark folder named trust desk, a sub-folder named lock-screen, and a sub-folder named advisory. The reader files every note the reader has applied to the locked XI into lock-screen and every note the reader has left as advisory into advisory. The filing system is the system the reader opens on every later contest the same call recurs, and the filing system is the system the reader can re-open on every contest the reader has to defend a change on. The saving is the difference between a reading record the reader can re-open and a reading record the reader has to rebuild.
The habit the protocol would not recommend is the always-re-read habit. Reading every note the desk has published, weighing every note against the locked XI, and changing the locked XI on the strength of every new read produces a locked XI that has re-read the desk but has not played a contest. The bounded view: the locked XI that follows the protocol on the first read will out-read the locked XI that re-reads every note the desk publishes, and the difference will be visible at the next deadline, not at the next season opener. The protocol's only forecast is that the seven steps compound, and the compounding is the only honest forecast the protocol is willing to make on a first-time reader's habit.