Home · Forward test log · Week 2
Three sessions traded of five. The biggest day in this log so far, and two days later the first losing day in it — which is the more useful of the two.
| Session | Printed | Taken | W / L | Net | Balance |
|---|---|---|---|---|---|
| Mon 17 Aug | 2 | 2 | 2 / 0 | +$160 | $100,717 |
| Tue 18 Aug | 0 | 0 | — / — | $0 | $100,717 |
| Wed 19 Aug | 11 | 8 | 4 / 4 | +$1,997 | $102,714 |
| Thu 20 Aug | — | 0 | — / — | $0 | $102,714 |
| Fri 21 Aug | 6 | 6 | 0 / 6 | −$647 | $102,067 |
| Week | 19 | 16 | 6 / 10 | +$1,510 | $102,067 |
The first losing day on this record. Six trades, no winners at all: −85.76, −212.28, −80.76, −70.76, −65.76, −131.52. It gave back 30% of everything the week had made before it, and it is the largest drawdown of the month so far, arriving inside a single afternoon.
On day one of publishing this I wrote that the red days would go up exactly like the green ones — same layout, same prominence, no softening. This is the day that sentence costs something. There is nothing to add to it, and nothing is being added.
One detail is worth reading even here, because it is the thing that decides whether a run like this ends a week or ends an account. The average losing trade was alive for six seconds, the longest for eleven. Six consecutive losses still came to less than one badly managed trade would have. That is what the stop is for and it is the only part of Friday that went to plan.
Drawdown allowance consumed: $646.84 of $4,000, about 16%. The trailing floor does not move up on a red day, so the whole loss comes straight out of the distance rather than being partly absorbed.
No trades. I also did not screenshot the session, so whether Falcon printed anything on Thursday is unrecorded. The rule here is that setups I did not take get logged with the reason, and I cannot honour that for a session I never captured. This is logged as a gap in the routine rather than as a quiet day, because those are different things and only one of them is my fault.
Read this before the number. Two systems ran in this account on Wednesday. Alongside Falcon · NQ I traded an upgraded build that is not released and is not for sale. Splitting the session by what each one printed, the unreleased build produced roughly $1,005 of the $2,135 gross and Falcon produced the rest, about $1,130. Both ran at three contracts, not one. Neither of those things is available to a subscriber today, so this day is not a result you could have reproduced, and it should not be read as one.
The table above counts the unreleased build's strikes, because those are the ones I can show on the chart: eleven printed, eight taken, four to target and four to the stop. The Falcon trades that made up the larger half of the money are not in those counts. That is an untidy row and I am leaving it untidy rather than merging two systems into one number that would look neater and mean less. When the fills export is reconciled the counts will be restated here, in the open.
The hit rate was fifty per cent and the day still closed up two thousand dollars. That is the whole mechanism in one sentence. Four of the eight taken trades hit the stop. The winners averaged $229.55 and the losers $78.00, so being right half the time was enough — and it is only enough for as long as every losing trade stays cheap. The moment one is allowed to run, the arithmetic inverts and the same fifty per cent becomes a losing day.
The number that explains it is not the P&L, it is the clock. The average winning trade ran 1 minute 33 seconds. The average losing trade lasted 6 seconds, and the longest loser lasted eleven. Largest win $600, largest loss $120.
Three contracts came from the drawdown allowance, not from confidence. A 100K account is allowed $4,000. Wednesday's worst moment consumed $244.56 of it, about six per cent. Size is arithmetic against a loss limit and nothing else — it is not a recommendation, and the day it is treated as one is the day it stops working.
Fees took $138.24 of $2,135 gross — 6.5%. On Monday of last week, on micros, a comparable session gave up 20% of gross to the same costs. This is the first session where the instrument change is visible on real volume rather than on two trades.
Falcon found no qualifying setup in the entire session. Nothing to take, nothing to skip, and the account finished exactly where it started.
This was also the first session on the new build. From today the signals run on NQ rather than MNQ — the instrument changed, the entry logic did not. A blank day is the least informative possible first day for that change, and it is worth saying plainly rather than dressing up: nothing here tests it. The first NQ ticket will.
Sessions like this are in the record for the same reason the losing ones are. A log that only appears on days worth talking about is an advertisement.
Two setups printed, both long, both taken, both paid at target. Nothing went against either entry at any point in the session.
The gap between the ticket and the account is the whole story of this session. Falcon printed $105 on each trade, $209.50 together. The account received $199.50 gross — $10 went to entry slippage before either trade had done anything — and then fees took $39.90 more. Net $159.60.
That fee number deserves saying plainly, because it is not a rounding error: twenty per cent of gross profit, on twenty-one contracts. The trades were correct, the fills were nearly correct, and a fifth of the result still went to the cost of trading. Today's broker report happens to break that down for the first time — commission 41%, exchange 37%, clearing 20%, NFA 2% — which is worth knowing because the broker's own commission line is only two fifths of what actually comes out.
This is also the argument that ended the instrument. A strategy aiming at roughly six points cannot carry that toll on micro contracts, where the fee is charged per contract and it takes many of them to build a position. The same two trades on the full-size contract would have cost about $8 instead of $39.90 — one contract instead of twenty-one, the same points, the same risk. The change is announced in full in the v2.0 update notice, and from here the log runs on it.
Drawdown for this session is zero: the account finished green, so no part of the loss allowance was used. The intraday floor was −$17.05 before the first trade resolved, which is real and is not the same thing as a drawdown against the account.
The first stops in this log arrived on Wednesday — four of them. For eleven trades across two weeks this record showed eleven targets and no full stop, and I said at the time that was the thing most likely to make it look better than it is. Wednesday settled it: half the taken trades hit the stop, and the session still finished green because the losers were small and fast. That is a more useful day than another clean sweep would have been.
Everything above can be checked rather than believed. The historical strikes stay on the chart in TradingView, and bar replay will walk you through any of them candle by candle.
Take the free week and check it yourselfResults shown are from a funded prop account executing in a simulated environment. Hypothetical or simulated performance results have certain limitations. Unlike an actual performance record, simulated results do not represent actual trading. Also, since the trades have not been executed, the results may have under- or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profit or losses similar to those shown. Falcon is a trading tool and does not provide financial advice. Futures trading involves substantial risk of loss.