Analysing Whether Your Entries Were Any Good
Traders spend most of their effort choosing entries and almost none measuring whether those choices were good, which means the effort never compounds into anything.
What follows is how to analyse entry quality from a record, which fields make it possible, and what the resulting comparisons can and cannot establish.
Outcome Is Not Entry Quality
A trade can be entered well and lose, or entered poorly and win, because the outcome combines entry, exit, sizing, costs and a large amount of variance.
Measuring entries therefore requires isolating them from everything else, which is why the record has to be structured before the analysis is attempted.
The Fields That Make It Possible
The level, the trigger, the time of the trigger, the time of the entry, the price at each, the invalidation and the distance to the next reference.
Those seven allow entry quality to be assessed independently of what happened afterwards, which no shorter record permits.
Measure One: Adverse Movement After Entry
Record how far price moved against the position before it moved in favour, expressed against the distance to the invalidation.
Consistently large adverse movement indicates entries are early or the invalidation is too tight, and the two have different remedies.
Measure Two: Distance From Trigger to Fill
The gap between the price at the trigger and the price actually obtained is a direct measure of execution and patience combined.
Totalling it across a month usually produces a figure larger than any individual loss, and it is entirely addressable, as options intraday tips describes.
Measure Three: Whether the Trigger Occurred
Mark each entry as triggered or anticipated, then compare the two groups, since anticipated entries are the most common quiet failure.
Most records show anticipated entries performing materially worse, which settles an argument that otherwise runs for years.
Measure Four: Participation at Entry
Note whether activity was expanding or fading when the position was taken, then group results by that single observation.
This is one of the few genuinely predictive relationships available intraday, and grouping usually makes the difference obvious quickly.
Measure Five: Which Level Was Used
Group entries by the type of reference: opening range boundary, previous session extreme, round number or something unmarked.
Entries at unmarked levels are frequently the worst group, and they are also the ones traders remember least accurately.
Measure Six: Test Number
Record whether the level was being tested for the first, second or third time when the entry was taken.
Second tests are frequently better placed because the reaction is known, and third attempts are frequently taken from frustration.
Measure Seven: Time of Day
Group entries by session window, since results usually concentrate in one part of the day with the quiet middle contributing costs.
The pattern appears within a few dozen observations and is actionable immediately, as index intraday tips sets out.
Measure Eight: Position in the Expiry Cycle
Entries taken close to expiry behave differently because erosion is severe and positioning distorts behaviour around levels.
Separating them frequently shows a disproportionate share of poor entries concentrated there rather than spread evenly.
Measure Nine: Distance Available at Entry
Record the distance to the next significant level at the moment of entry, expressed as a multiple of the distance to the invalidation.
Entries taken with a poor ratio are usually the ones that were wanted rather than the ones that qualified.
Measure Ten: Whether the Cost Filter Was Applied
Mark whether the expected distance comfortably exceeded the round-trip cost, and group results by that mark.
Marginal entries rarely look wrong individually and collectively account for much of a losing month, which this comparison makes visible.
Compare Against Entries You Declined
Setups meeting the criteria that were not taken form the control group, and without them there is no way to judge whether the filter is helping.
Many traders discover their filtering removes the better trades, which is fixable only once the declined set exists on paper.
Compare Against a Naive Entry
Ask what would have happened taking every qualifying setup mechanically without discretion, over the same period.
Where discretion underperforms the mechanical version, that is a specific and valuable finding rather than an insult.
Separate Entry Quality From Exit Quality
Compute the best available result within the intended holding window, then compare it with what was actually realised.
A large gap indicates an exit problem, while a small favourable excursion in most trades indicates an entry problem, and the two are frequently confused.
Sample Size Before Conclusions
Short runs are dominated by variance, so an entry method judged on a dozen trades has been judged on the period rather than on itself.
Decide the number in advance, since choosing it afterwards guarantees a conclusion selected by whichever grouping looks most convincing.
Look at the Distribution
A single unusually large gain can make an entry group look excellent when the typical instance was unremarkable.
Noting how much of the group’s result came from the largest few trades prevents a conclusion that will not repeat.
Change One Element at a Time
Adjusting the trigger, the confirmation and the level types together makes attribution impossible when the results move afterwards.
Each change deserves its own sample, which is slower and is the only method that produces knowledge, as intraday trading strategies describes.
Beware of Filtering in Hindsight
Excluding the entries that lost, after the fact, produces a description of the past that predicts nothing about future sessions.
Any new condition must be stated in advance and tested on entries that have not yet occurred, or it is not a finding at all.
What the Analysis Cannot Tell You
It cannot say whether the coming period will resemble the last, and it cannot distinguish a method that has stopped working from an ordinary drawdown quickly.
Treating it as a diagnostic rather than a forecast keeps it useful, and the same standard applies to longer horizons, as investment advisory describes.
Making the Analysis Cheap
Seven fields entered at the close and a monthly grouping exercise is enough, and anything more elaborate will be abandoned within a few weeks.
Sustainability matters more than completeness here, as the routine in the intraday trading guide sets out.
Review at a Fixed Interval
An analysis performed only after a bad run is an emotional response wearing the clothes of a process, and it reaches conclusions that match the mood.
A fixed interval, whether monthly or every fifty trades, produces comparable reviews and removes the temptation to look only when something hurts.
What Actually Changes After a Review
A review that ends without a written change to one rule has produced a feeling rather than a finding, and the next review will reach the same place.
The output should be a single sentence naming what will be done differently and the sample over which it will be judged.
Analysis Cannot Replace a Defined Setup
Measuring entries assumes there was a rule to measure against, and discretionary entries produce data describing a mood rather than a method.
Where the setup is undefined, the first task is definition rather than measurement, as nifty intraday tips sets out.
Where Entry Analysis Sits in the Whole Process
Entry quality is one of four things that determine a record, alongside sizing, exits and the decision to trade at all, and it is the smallest of them.
Improving it is worth doing once the other three are stable, and intraday tips for beginners describes the order in which they are usually best addressed.
FAQs
Can entry quality be measured separately from outcome?
Yes, using adverse movement after entry, the gap between trigger and fill, and the distance available relative to the invalidation.
Which comparison is most informative?
Triggered entries against anticipated ones. Most records show anticipated entries performing materially worse.
Why record declined setups?
They are the control group. Without them there is no way to know whether your filtering helps or removes the better trades.
How is an exit problem distinguished from an entry problem?
Compare the best result available within the holding window against what was realised. A large gap points at exits.
How many trades before drawing conclusions?
Enough that variance averages out, with the number decided in advance rather than after the groupings have been seen.
Can conditions be added after the analysis?
Only if stated in advance and tested on entries that have not yet occurred. Excluding past losers describes the past only.
How much record keeping does this need?
Around seven fields per trade and a monthly grouping exercise. Anything more elaborate tends to be abandoned quickly.