A falling equity curve can make closing positions and waiting for a new high feel sensible. Oliver Sparing and Sven Otremba discuss how traders deal with such periods in the Trading Podcast. Sven tested an exit rule using the Equity Growth System in TradeNeon's Titan module. The comparison describes what happened under that rule in the historical test; it cannot predict the next drawdown.

Basis for every figure below: historical Equity Growth System backtestIn a backtest, a trading strategy is applied to historical price data to check how it would have performed in the past. using daily data from 3 January 2000 to 8 September 2026; Sven's analysis dated 13 September 2026. The results are hypothetical, not actual trading results. Past backtests are not reliable indicators of future results. Trading stocks can result in the loss of invested capital.

01

Which exit rule did Sven test?

A drawdown is the decline of an equity curve from its most recent high. In Sven's calculation, a follower initially mirrors the model portfolio. If that portfolio's curve falls five or ten percent below its previous high, the follower closes all positions that day and holds cash. Trading resumes only when the model portfolio reaches a new high.

This is not a stop-loss on an individual trade. It is an additional rule for the whole account. The model portfolio continues trading during the pause; the follower's account does not participate in any recovery. Once the curves diverge, a five-percent trigger on the model portfolio cannot guarantee a five-percent loss limit in the follower's account.

Sven compared the variants using the same daily returns as the original backtest. This is a historical rule test. It does not reproduce a particular follower's execution, different fill prices (slippageSlippage is the difference between the expected execution price of an order and the price at which it is actually filled.) or missed trades as a personal account statement.

02

What changed in the backtest?

The three Equity Growth System sizes differ in their strategies, stock universes and maximum number of simultaneous positions. For each size, Sven compared continuous participation with pauses triggered at five and ten percent.

Sven's comparison · historical backtest
How a pause changes the curve
ContinuousPause at 10 %Pause at 5 %
Historical EGS equity curves, Size SLogarithmic scale. Three sets of sampled points from Sven's curves: continuous participation, pause at ten percent and pause at five percent, 2000 to 2026. Final values: 306.8, 67.0 and 15.1 times the starting value.125102050100200500200020052010201520202025Continuous: Final value 306.8 times the starting valuePause at 10 %: Final value 67.0 times the starting valuePause at 5 %: Final value 15.1 times the starting value
Continuous: 306.8×Pause at 10 %: 67.0×Pause at 5 %: 15.1×
Indexed equity curve, logarithmic scale. Half-year points and the final weekly point are sampled from Sven's weekly curves; straight lines between them do not show every fluctuation. Daily backtest: 3 January 2000 to 8 September 2026; final weekly label: 11 September 2026. Source: Sven Otremba, analysis dated 13 September 2026. Hypothetical results, not a forecast.

Historical backtest: 3 January 2000 to 8 September 2026; analysis: 13 September 2026. Annualised return and deepest drawdown refer to the full simulated curve of each variant.

Size Measure Continuous Pause at 5% Pause at 10%
S Annualised backtest return 23.9% 10.7% 17.1%
S Deepest backtest drawdown −17.9% −19.8% −16.3%
M Annualised backtest return 25.5% 11.8% 19.6%
M Deepest backtest drawdown −19.4% −18.0% −16.6%
L Annualised backtest return 26.8% 12.4% 20.4%
L Deepest backtest drawdown −18.3% −20.4% −14.6%

With the five-percent rule, annualised backtest returns fell to less than half for all three sizes. The deepest drawdown became larger for S and L and slightly smaller for M. The ten-percent rule also reduced returns across all three sizes. It reduced the deepest decline relative to continuous participation in this test, but did not cap it at ten percent.

Time changed too. For size S, the five-percent rule extended the longest period between equity highs from 7.4 to 49.7 months in the backtest. This variant held cash for 29 percent of the time. A pause can therefore mean a long departure from the system originally tested.

03

Why can a pause change so much?

Exiting does not remove the decline that triggered the exit. While the follower holds cash, they do not participate in a possible recovery of the model portfolio. Under the tested rule, they return only after that portfolio exceeds its previous high. Until then, the follower's account follows a different path.

This explains why the pause was costly in Sven's retrospective calculation. It does not establish a universal rule that every exit is harmful. An exit for another reason, or a different re-entry rule, would be a different test. Sven compared this precise combination: pause at a five- or ten-percent decline and return at the model portfolio's next high.

04

The harder part: time below the last high

A drawdown need not be deep to become difficult to live with. In Sven's backtest, each of the three original EGS curves spent 73 to 74 percent of the time below its previous high. A long period without a new high can create doubt even when the strategy still follows its rules. Loss aversionLoss aversion describes the fact that losses weigh more heavily psychologically than gains of the same size. can strengthen the impulse to react immediately to a visible decline. Skipping trades or changing size from day to day means following a different variant from the one previously tested.

In Trading Podcast #157, “Drawdown in Trading: Why Taking a Break Can Hurt”, Oliver and Sven discuss the rule using Equity Growth System curves. From 18:07, they compare the five- and ten-percent pauses with continuous participation. The conversation is in German; YouTube's English subtitles may help. They also discuss how drawdowns feel and why past lows do not set a ceiling on future losses.

05

Decide before the first trade

Sven's comparison suggests a practical question before following a system: what decline and how long without a new high could your chosen position size mean for your account? The deepest historical drawdown is an observation, not a limit. Future losses can be larger.

The trader chooses the capital committed and the risk taken. A smaller position changes both potential monetary losses and potential gains; it does not replace checking whether the rules still apply or deciding what to do if a meaningful assumption breaks. A divergence between a personal account and the model portfolio may also deserve investigation. This backtest cannot establish a blanket obligation to keep trading through every decline.

For context on the Equity Growth System, read “Stock Trading Systems” for rules and backtest basics. The Equity Growth System page describes the sizes, requirements and current backtest presentation. Its figures have a more recent data cut-off; do not combine them with Sven's September 2026 table as one time series.

06

Frequently asked questions

Does exiting at five percent limit my loss to five percent?

No. In Sven's tested rule, a decline in the model portfolio triggers the pause. Your account can differ; five percent is not a guaranteed loss limit. For S and L, the deepest historical backtest drawdown with this pause was greater than with continuous participation.

Must I always keep trading through a drawdown?

No. The calculation evaluates one specific pause and re-entry rule in historical data. It does not replace checking whether the strategy still applies, whether orders are executed correctly and whether your chosen risk remains bearable.

Are the table values actual trading results?

No. They are hypothetical backtest results for 3 January 2000 to 8 September 2026, analysed by Sven Otremba on 13 September 2026. Actual execution may differ. Historical results do not reliably predict the future, and stock trading can lead to loss of invested capital.


Sources and context: Sven Otremba, “Exiting During a Drawdown” (analysis dated 13 September 2026) and notes for “Titan Pitfalls”, episode 2 (13 September 2026); TradeNeon Titan and Lab of Trades product masters. On hypothetical backtests, see the SEC Investor Bulletin on Performance Claims.

Oliver Sparing
Oliver Sparing
Founder & Head of Trading · TradeNeon
Oliver Sparing has been actively trading the financial markets since 2011. His professional roots lie in precious metals trading in Hamburg; in 2017 he learned at a London trading floor. At TradeNeon he leads the day trading and swing trading programmes and is responsible for the trading methodology, with a focus on order flow, market structure and fundamental market data. Level of Interest, Futures Insights and Lab of Trades grew out of this work. Platforms such as ATAS, Bookmap and WH SelfInvest regularly invite him as a guest. All guest appearances