CAT
Follow

Track record

Every analysis CAT publishes is graded to a result — a win, a loss, or an expiry — automatically, with no manual grading and nothing cherry-picked. None are removed once published. This page is that record, unfiltered.

Back to the wall

Win rate

not yet shown
building the record2 won · 1 lost

3 of 30 tracked outcomes needed. This is the same rule applied to every other figure on this page — a percentage from a handful of reads says more about luck than about method, so it is shown once there are 30.

Everything published

nothing is removed from these counts
Published
8
analyses on the wall
Graded
3
won or lost
Still tracking
1
Expired
0
undecided at 30 bars

A further 4 are marked read only — reads carrying no target-and-invalidation pair, which this record cannot grade either way. They are published and counted here rather than quietly dropped.

Average R

secondary — reward:risk on won+lost with a known entry
83.56R3 tracked

By method

Elliott Wave + ZTD
building the record — 3 tracked so far

By timeframe

H4
building the record — 3 tracked so far

Methodology

What counts as a win
A read counts as a win when price reaches the target zone — a high/low touch is enough, it doesn't need to close there — before the price closes beyond the invalidation level on that read's own timeframe. The headline win rate above is this single yes/no question, nothing more.
What counts as a loss
A read counts as a loss when a bar CLOSES beyond the invalidation level — a brief wick through it, without closing there, does not count. If price gaps straight through invalidation, the loss is counted at that gap's close, not the wick.
Expiry
A read that stays undecided for 30 bars of its own timeframe is marked expired. It is shown on the wall and published in the counts on this page, but excluded from the headline win rate — an unresolved read is neither a win nor a loss, and hiding it would flatter the record.
Every step recorded
Every step of a read's outcome — each target reached, each invalidation, each expiry — is recorded as a timestamped event, visible on that read's own wall page. The headline stays a simple win/loss ratio deliberately; average reward-to-risk (R) and deeper target hits are secondary figures published alongside it, not folded into it.
Why splits stay hidden below 30
Win rates broken out by engine or timeframe are only shown once that split has at least 30 tracked wins-plus-losses. Smaller samples are too easy to misread as a pattern, so until a split reaches 30 this page simply reports how many outcomes are tracked so far instead of a percentage that could mislead.
Nothing is filtered out
Every analysis this system publishes tracks all the way to a result — win, loss, or expiry — even after a newer read is published for the same symbol. Nothing is filtered out or quietly dropped from this record; the wall links each older read to whatever replaced it.
Reads this record cannot grade
Some reads carry no defined target-and-invalidation pair (certain Elliott Wave reads are description-only). Those are labeled 'read only' on the wall and excluded from every statistic here — that is neither a win nor a loss, simply not a claim this record can grade.
When one bar does both
When a single bar both reaches the target and closes beyond invalidation, it counts as a loss. The order those two things happened within that bar can't be proven from OHLC data alone, and where the sequence can't be proven, this record under-claims rather than assumes the more flattering order.