Two articles in the finding has been consistent. We are testing one specific question across different market alignment conditions. When the 7 period RSI lands in a specific 5 point zone does it matter whether it arrived there on a declining RSI or a rising RSI.

The rising RSI entry is the conventional approach. Wait for the RSI to turn back up before buying. That confirmation signal is what most traders are taught to look for. We call it the cross over entry. The declining RSI entry is the unconventional one. The RSI is still falling when you enter. We call it the cross under entry. Both entries find the RSI landing in the same tight zone. The only difference is the direction it was traveling when it got there.

Across two alignment conditions the data has consistently challenged the conventional approach. The cross under entry has outperformed the cross over entry in the majority of zones tested. This article pushes that question into the most hostile environment in the series so far. The 20 day moving average is below the 50. The 50 is below the 200. SPY is below its 200 day moving average. Everything is pointed down.

Most traders would not consider buying weakness here. The data suggests they might want to reconsider. At least in specific zones.

Article 1 - fully bullish alignment

Article 2 - early deterioration

The Test Setup

The methodology is identical to the previous articles. Nasdaq 100 stocks over 25 years. Testing done in Wealth-Lab using WealthData, a clean daily bar data source checked for bad ticks and data oddities. Entry at the open the morning after the signal fires. Fixed 5 day hold.

The same stocks. NVDA, AMZN, TSLA and the rest of the Nasdaq 100. The same 7 period RSI tested in clean 5 point landing zones. For each zone a cross under entry triggers when the RSI crosses down through the upper boundary and closes above the lower boundary. A cross over entry triggers when the RSI crosses up through the lower boundary and closes below the upper boundary. Both entries land in the same tight 5 point zone. The only difference is the direction the RSI was traveling when it got there. Entry is at the open the morning after the signal fires. The position is held for exactly 5 trading days then closed. No stops, no profit targets, no adjustments. A fixed 5 day hold regardless of what price does in between.

The alignment tested here is the most bearish in the series. The 20 day moving average is below the 50. The 50 is below the 200. SPY is below its 200 day moving average. All three moving averages stacked in the opposite order from article 1. The broader market and the individual stock are both in a fully bearish configuration.

One note on sample sizes. The extreme low RSI zones have limited trade counts on both sides. The extreme high RSI zones on the cross over side are also thin. Where sample sizes are limited the findings should be treated with caution. The full data is shown.

What the Data Shows

The core finding of this series continues to hold up. In 13 of 17 RSI zones, a declining RSI entering the zone (the cross under entry) outperformed a rising RSI entering the same zone. Seeing that pattern persist across three very different market environments is significant on its own.

What makes this alignment especially interesting is where the strongest results are appearing. The lower RSI zones are producing the highest quality trades we have seen in the entire series.

When the 7-period RSI falls into the 15 to 20 range on a declining basis, the cross under setup produces an average net profit of 2.06% with a profit factor of 1.65 across 567 trades. In the 20 to 25 range, the setup returns 1.48% with a 1.51 profit factor across 1,376 trades. In the 25 to 30 range, it delivers 1.18% with a 1.42 profit factor across 2,498 trades.

Those are the strongest profit factors we have seen in the lower RSI zones so far. Stronger than the fully bullish alignment. Stronger than the early deterioration alignment. In the most hostile market environment tested, the declining RSI setup in deeply oversold conditions is producing the best trade quality in the data.

There are fewer opportunities because of the alignment conditions, but the setups that do appear are consistently standing out as the highest quality trades.

The average winners and losers are also much larger here than in the fully bullish environment, roughly double in size in some zones. That is simply the nature of bearish markets. Price swings expand in both directions. But larger volatility alone does not weaken the edge. Profit factor already accounts for win rate, average winner, and average loser in a single metric. Even with the wider swings, the lower RSI zones in this bearish alignment are still producing stronger profit factors than the fully bullish environment.

From a practical standpoint, the larger swings are handled through position sizing. The underlying edge itself remains intact. If your average winner and loser are twice as large than our other setups the simple solution is to “normalize” this volatility with smaller position sizes.

What Stands Out Across Three Alignments

Three different alignment conditions have now been tested, all asking the same basic question: does a declining RSI entering a zone outperform a rising RSI entering the same zone? Across all three alignments, the answer has consistently been yes in the majority of cases. That alone is an important result.

But the most surprising finding in the series appears in this latest data set.

In the 20 to 25 and 25 to 30 RSI zones, the cross under entry actually produces a higher profit factor in the fully bearish alignment than it does in the fully bullish one. In the 20 to 25 zone, the fully bearish alignment posts a 1.51 profit factor versus 1.28 in the fully bullish alignment. In other words, the most hostile market environment tested is producing the strongest edge in those specific oversold zones.

The early deterioration alignment tells a different story. That environment, where the 20-day moving average has crossed below the 50-day while SPY itself remains relatively healthy, is where the cross under edge weakens the most in the upper RSI zones. Above the 50 to 55 RSI range, the traditional confirmation entry, the cross over, consistently outperforms the cross under. Interestingly, that pattern does not appear in either the fully bullish or fully bearish alignments.

One thing that remains remarkably consistent across all three environments is the behavior of the lower RSI zones. In deeply oversold conditions, the conventional confirmation entry has never been the superior trade. In both the fully bullish and early deterioration alignments, the cross over entry actually produces negative profit factors in the 15 to 20 and 20 to 25 RSI ranges.

There is also an important nuance in the fully bearish alignment itself. In the fully bullish environment, the gap between cross under and cross over performance in the lower zones was dramatic. Cross over often produced negative profit factors while cross under remained consistently profitable. In the fully bearish alignment, that gap narrows considerably.

In the 15 to 20 RSI zone, cross over actually edges out cross under slightly, with a 1.68 profit factor versus 1.65. In the 20 to 25 and 25 to 30 zones, cross under still performs better, but the margin is much smaller than what was seen in the bullish alignments. In other words, both approaches show edge in the lower RSI zones during fully bearish conditions. The data suggests some normalization between the two entry types, just not enough to fully erase the broader cross under advantage.

Another clear shift in the bearish alignment is the expansion in both average winner and loser size. That is expected in a high-volatility environment. What matters is that the edge, measured through profit factor, actually improves in the lower RSI zones despite the larger swings. The volatility changes the sizing requirements, not the validity of the edge itself.

Article 1 - fully bullish alignment

Article 2 - early deterioration

What This Tells Us

Three articles in. Three alignment conditions tested. The declining RSI entry has outperformed the conventional confirmation entry in the majority of zones across all three. That finding has held in the most bullish environment we tested and in the most bearish. The edge is not a product of favorable market conditions.

The lower RSI zones in the fully bearish alignment are producing the strongest profit factors in the series so far. That is not what most traders would expect. A stock with everything pointed down, generating a declining RSI into the 15 to 30 range, is showing up in the data as one of the better long side setups we have found across 25 years of Nasdaq 100 data.

One important note about what this data represents. Every qualifying trigger across the entire Nasdaq 100 universe is included. This is not filtered through portfolio slots, position limits, or capital allocation rules. Those are separate and important considerations for system design. What this data shows is the raw entry timing question across the complete population of signals. That is what gives the trade counts their statistical meaning.

This is raw research. Not a trading system. A complete system requires position sizing, stop methodology, portfolio heat management, and exit optimization. None of that is addressed here. What is addressed is a specific and testable question about entry timing across three different alignment conditions.

Some people will ask what they are supposed to do with this. The answer is straightforward. In the lower RSI zones a declining RSI is consistently better than a rising RSI across every alignment tested. Knowing where that edge lives and where it fades is worth something regardless of what you do with it next.

If you want to see what this kind of research looks like built into complete tradeable systems, that work is on the systems page at TradingTimeMachine.com

Next up: the same test with the stock fully bearish but SPY still above its 200 day moving average. A stock fully bearish against a healthy market. Does the edge survive that condition? The data will tell us. There are so many conditions that can be tested, is there one you the reader would like me test? Let me know!

Dave Johnson

Quant Developer at

TradingTimeMachine.com

Via https://backtest.substack.com/p/bear-market-rsi-entries-the-data

The Edge Shifts: What Happens When the 20 Crosses Below the 50

In the first article, we tested a simple question: in a fully bullish stock alignment, is it better to buy weakness or wait for confirmation?

The answer was pretty clear. Buying on the cross under outperformed waiting for the cross over in 17 of 18 RSI zones across 25 years of Nasdaq 100 data.

A fully bullish alignment means the 20-day moving average is above the 50, the 50 is above the 200, and SPY is trading above its own 200-day moving average. In other words, everything is stacked in bullish order and pointing in the same direction.

This time, we’re testing a different condition.

The 20-day moving average has now crossed below the 50. The 50 is still above the 200, and SPY remains above its 200-day moving average. So the longer-term structure is still intact, and the broader market is still healthy, but early weakness has started to show up in the stock itself.

Same question, different market structure.

What happens now?

Here’s what the data says.

The Test Setup

The methodology here is identical to the first article.

The test uses 25 years of Nasdaq 100 data, run in Wealth-Lab with data from WealthData, a clean daily bar source screened for bad ticks and other data anomalies. Entries are taken at the next day’s open after a signal fires, with a fixed 5-day holding period.

The universe is unchanged: NVDA, AMZN, TSLA, and the rest of the Nasdaq 100.

We again use a 7-period RSI, tested across clean 5-point landing zones. For each zone, a cross under entry occurs when RSI moves down through the upper boundary and closes above the lower boundary. A cross over entry occurs when RSI moves up through the lower boundary and closes below the upper boundary.

Both setups place the entry in the exact same 5-point RSI zone. The only difference is how price got there: momentum fading into the zone versus momentum strengthening into it.

One important note before looking at the results: some zones in this alignment have relatively low trade counts, especially at the highest RSI readings. Sample sizes get thinner at those extremes, so those results should be viewed with a little more caution.

The full dataset is included so you can evaluate the numbers for yourself.

What the Data Shows

Cross under outperformed cross over in 11 of 18 RSI zones. That’s a noticeable drop from the 17 of 18 zones seen in the fully bullish alignment.

The edge is still there, but it’s clearly weaker.

In the lower RSI zones, the pattern largely holds up. In the 15 to 20 zone, cross under produced an average net profit of 0.49%, a 1.36 profit factor, and a 53.00% win rate across 1,202 trades. Cross over in the same zone lost money, with an average net profit of -0.16%, a 0.92 profit factor, and a 50.53% win rate across 281 trades. A profit factor below 1.0 means the strategy lost more than it made.

The same dynamic appears in the 20 to 25 zone. Cross under remained profitable at 0.47% with a 1.34 profit factor, while cross over posted a -0.11% return and a 0.93 profit factor.

The middle RSI zones are where things begin to change.

Unlike the first article, cross over starts to gain an edge around the 50 to 55 zone. By 55 to 60, it leads cross under by 0.17%. In the 60 to 65 zone, the lead is still present, though smaller at 0.08%. As RSI climbs higher within this weaker alignment, the advantage of buying weakness begins to fade.

The 80 to 85 zone is interesting enough to call out separately. Here, cross under generated a 1.07% average net profit, a 2.26 profit factor, and a 59.57% win rate. The catch is sample size: just 141 trades over 25 years. Promising, but not robust.

Beyond that, the data gets too thin to take seriously. The 85 to 90 zone has only 33 cross under trades, and the 90 to 95 zone has exactly one. That’s not analysis anymore, that’s statistical cosplay.

The one real exception in the lower range is the 30 to 35 zone, where cross over edges out cross under by just 0.05%. Given the sample sizes—3,681 cross under trades versus 2,009 cross over trades—that difference is effectively noise. Neither setup shows a meaningful advantage there.

How This Compares to the Fully Bullish Alignment

In the fully bullish alignment, cross under dominated across nearly the entire RSI range. Here, that advantage becomes more selective.

Instead of working almost everywhere, the edge now clusters in the lower RSI zones and gradually disappears as RSI moves higher. A direct comparison makes the shift obvious.

At the lower end, the results are remarkably similar to article 1. In the 20 to 25 zone, the cross under profit factor comes in at 1.34 versus 1.28 in the fully bullish alignment. In the 25 to 30 zone, it’s 1.37 versus 1.38. In other words, very little has changed where the stock is already showing deeper short-term weakness.

Higher up the RSI scale, the behavior changes materially.

In article 1, cross under maintained a clear advantage all the way through the 75 to 80 zone. Under this weaker alignment, cross over begins outperforming as early as the 50 to 55 zone and continues to show strength from there. That is a meaningful change in market behavior.

Win rates tell the same story. In the lower zones, cross under still posts win rates in the 53% to 55% range, broadly in line with the fully bullish test. As RSI rises, that advantage compresses and in several zones reverses entirely.

One additional detail stands out in the comparison: both average winners and average losers are modestly larger across several zones versus article 1.

That’s consistent with what you would expect from a stock showing early structural deterioration. Over a 5-day holding period, price swings widen in both directions.

Frequency and Exposure

The question is unchanged from article 1: does buying weakness (cross under) outperform waiting for confirmation (cross over) within the same RSI zone?

Lower RSI (15–25)

Cross under maintains a clear and consistent edge.

Profit factors remain above 1.30, with positive average returns and win rates in the low-to-mid 50% range. Cross over is negative across the same zones.

This is the most stable part of the distribution and closely matches the fully bullish alignment results.

Middle RSI (25–50)

The edge compresses but does not disappear.

Both approaches remain viable, but cross under holds a modest advantage through roughly the 45–50 zone. Differences are smaller and less consistent than in the lower band.

This is where the first structural shift becomes visible relative to article 1.

Upper RSI (50+)

The edge changes.

Cross over begins to outperform cross under as RSI rises further into upper territory. In this alignment, cross under signals become less frequent and less reliable once RSI reaches elevated levels after the 20/50 crossover has already occurred.

Frequency Considerations

High RSI readings in this alignment are structurally less common. Once the 20 has crossed below the 50 while SPY remains above its 200-day, sustained moves into high RSI zones are rarer.

As a result, upper-zone cross under signals are both lower frequency and thinner in sample size, which naturally reduces the stability of those results.

Takeaway

Where sample size is sufficient, the same core finding from article 1 holds: cross under carries a persistent edge in lower RSI conditions during a healthy broader market. That edge degrades as RSI rises and eventually flips in higher zones under this alignment.

Two articles in. Two alignment conditions tested. The pattern is consistent.

In a fully bullish alignment, cross under outperformed cross over in 17 of 18 RSI zones across the full range. In early deterioration—where the 20 has crossed below the 50 but SPY remains above its 200-day—the edge is reduced but still present, with cross under winning in 11 of 18 zones. The advantage shifts rather than disappears, concentrating in lower RSI conditions and fading above 50.

A note on interpretation.

All results include every qualifying signal across the Nasdaq 100 universe over the full test period. There are no portfolio constraints, position caps, or capital allocation rules applied. This is intentional. The objective here is to isolate entry timing behavior across the full distribution of signals. Portfolio construction is a separate layer.

This is not a complete trading system. It is a controlled test of a single decision point: entry timing across defined RSI zones under different market alignments. Position sizing, risk limits, exits, and portfolio heat are not part of this analysis.

The result itself is consistent across both studies. Buying weakness (cross under) continues to outperform waiting for confirmation in the lower RSI ranges, particularly where signal density is highest. That is where the statistical edge is most stable.

The implication is narrow but clear: under multiple market structures, the distribution of outcomes favors early entry in oversold conditions rather than confirmation-based entry.

If you want to see what this kind of research looks like built into a complete tradeable systems, that work is at TradingTimeMachine.com.

Next up: the same test with the 20 below the 50 and the 50 below the 200. The fully bearish alignment. And this time we look at the long and potentially short side in poor ranges. Does buying down beat waiting for confirmation when everything points down? The data will tell us.

Dave Johnson

Quant Developer at

TradingTimeMachine.com

Via https://backtest.substack.com/p/does-buying-weakness-still-work-when

The Edge Shifts: What Happens When the 20 Crosses Below the 50

In the first article, we tested a simple question: in a fully bullish stock alignment, is it better to buy weakness or wait for confirmation?

The answer was pretty clear. Buying on the cross under outperformed waiting for the cross over in 17 of 18 RSI zones across 25 years of Nasdaq 100 data.

A fully bullish alignment means the 20-day moving average is above the 50, the 50 is above the 200, and SPY is trading above its own 200-day moving average. In other words, everything is stacked in bullish order and pointing in the same direction.

This time, we’re testing a different condition.

The 20-day moving average has now crossed below the 50. The 50 is still above the 200, and SPY remains above its 200-day moving average. So the longer-term structure is still intact, and the broader market is still healthy, but early weakness has started to show up in the stock itself.

Same question, different market structure.

What happens now?

Here’s what the data says.

The Test Setup

The methodology here is identical to the first article.

The test uses 25 years of Nasdaq 100 data, run in Wealth-Lab with data from WealthData, a clean daily bar source screened for bad ticks and other data anomalies. Entries are taken at the next day’s open after a signal fires, with a fixed 5-day holding period.

The universe is unchanged: NVDA, AMZN, TSLA, and the rest of the Nasdaq 100.

We again use a 7-period RSI, tested across clean 5-point landing zones. For each zone, a cross under entry occurs when RSI moves down through the upper boundary and closes above the lower boundary. A cross over entry occurs when RSI moves up through the lower boundary and closes below the upper boundary.

Both setups place the entry in the exact same 5-point RSI zone. The only difference is how price got there: momentum fading into the zone versus momentum strengthening into it.

One important note before looking at the results: some zones in this alignment have relatively low trade counts, especially at the highest RSI readings. Sample sizes get thinner at those extremes, so those results should be viewed with a little more caution.

The full dataset is included so you can evaluate the numbers for yourself.

What the Data Shows

Cross under outperformed cross over in 11 of 18 RSI zones. That’s a noticeable drop from the 17 of 18 zones seen in the fully bullish alignment.

The edge is still there, but it’s clearly weaker.

In the lower RSI zones, the pattern largely holds up. In the 15 to 20 zone, cross under produced an average net profit of 0.49%, a 1.36 profit factor, and a 53.00% win rate across 1,202 trades. Cross over in the same zone lost money, with an average net profit of -0.16%, a 0.92 profit factor, and a 50.53% win rate across 281 trades. A profit factor below 1.0 means the strategy lost more than it made.

The same dynamic appears in the 20 to 25 zone. Cross under remained profitable at 0.47% with a 1.34 profit factor, while cross over posted a -0.11% return and a 0.93 profit factor.

The middle RSI zones are where things begin to change.

Unlike the first article, cross over starts to gain an edge around the 50 to 55 zone. By 55 to 60, it leads cross under by 0.17%. In the 60 to 65 zone, the lead is still present, though smaller at 0.08%. As RSI climbs higher within this weaker alignment, the advantage of buying weakness begins to fade.

The 80 to 85 zone is interesting enough to call out separately. Here, cross under generated a 1.07% average net profit, a 2.26 profit factor, and a 59.57% win rate. The catch is sample size: just 141 trades over 25 years. Promising, but not robust.

Beyond that, the data gets too thin to take seriously. The 85 to 90 zone has only 33 cross under trades, and the 90 to 95 zone has exactly one. That’s not analysis anymore, that’s statistical cosplay.

The one real exception in the lower range is the 30 to 35 zone, where cross over edges out cross under by just 0.05%. Given the sample sizes—3,681 cross under trades versus 2,009 cross over trades—that difference is effectively noise. Neither setup shows a meaningful advantage there.

How This Compares to the Fully Bullish Alignment

In the fully bullish alignment, cross under dominated across nearly the entire RSI range. Here, that advantage becomes more selective.

Instead of working almost everywhere, the edge now clusters in the lower RSI zones and gradually disappears as RSI moves higher. A direct comparison makes the shift obvious.

At the lower end, the results are remarkably similar to article 1. In the 20 to 25 zone, the cross under profit factor comes in at 1.34 versus 1.28 in the fully bullish alignment. In the 25 to 30 zone, it’s 1.37 versus 1.38. In other words, very little has changed where the stock is already showing deeper short-term weakness.

Higher up the RSI scale, the behavior changes materially.

In article 1, cross under maintained a clear advantage all the way through the 75 to 80 zone. Under this weaker alignment, cross over begins outperforming as early as the 50 to 55 zone and continues to show strength from there. That is a meaningful change in market behavior.

Win rates tell the same story. In the lower zones, cross under still posts win rates in the 53% to 55% range, broadly in line with the fully bullish test. As RSI rises, that advantage compresses and in several zones reverses entirely.

One additional detail stands out in the comparison: both average winners and average losers are modestly larger across several zones versus article 1.

That’s consistent with what you would expect from a stock showing early structural deterioration. Over a 5-day holding period, price swings widen in both directions.

Frequency and Exposure

The question is unchanged from article 1: does buying weakness (cross under) outperform waiting for confirmation (cross over) within the same RSI zone?

Lower RSI (15–25)

Cross under maintains a clear and consistent edge.

Profit factors remain above 1.30, with positive average returns and win rates in the low-to-mid 50% range. Cross over is negative across the same zones.

This is the most stable part of the distribution and closely matches the fully bullish alignment results.

Middle RSI (25–50)

The edge compresses but does not disappear.

Both approaches remain viable, but cross under holds a modest advantage through roughly the 45–50 zone. Differences are smaller and less consistent than in the lower band.

This is where the first structural shift becomes visible relative to article 1.

Upper RSI (50+)

The edge changes.

Cross over begins to outperform cross under as RSI rises further into upper territory. In this alignment, cross under signals become less frequent and less reliable once RSI reaches elevated levels after the 20/50 crossover has already occurred.

Frequency Considerations

High RSI readings in this alignment are structurally less common. Once the 20 has crossed below the 50 while SPY remains above its 200-day, sustained moves into high RSI zones are rarer.

As a result, upper-zone cross under signals are both lower frequency and thinner in sample size, which naturally reduces the stability of those results.

Takeaway

Where sample size is sufficient, the same core finding from article 1 holds: cross under carries a persistent edge in lower RSI conditions during a healthy broader market. That edge degrades as RSI rises and eventually flips in higher zones under this alignment.

Two articles in. Two alignment conditions tested. The pattern is consistent.

In a fully bullish alignment, cross under outperformed cross over in 17 of 18 RSI zones across the full range. In early deterioration—where the 20 has crossed below the 50 but SPY remains above its 200-day—the edge is reduced but still present, with cross under winning in 11 of 18 zones. The advantage shifts rather than disappears, concentrating in lower RSI conditions and fading above 50.

A note on interpretation.

All results include every qualifying signal across the Nasdaq 100 universe over the full test period. There are no portfolio constraints, position caps, or capital allocation rules applied. This is intentional. The objective here is to isolate entry timing behavior across the full distribution of signals. Portfolio construction is a separate layer.

This is not a complete trading system. It is a controlled test of a single decision point: entry timing across defined RSI zones under different market alignments. Position sizing, risk limits, exits, and portfolio heat are not part of this analysis.

The result itself is consistent across both studies. Buying weakness (cross under) continues to outperform waiting for confirmation in the lower RSI ranges, particularly where signal density is highest. That is where the statistical edge is most stable.

The implication is narrow but clear: under multiple market structures, the distribution of outcomes favors early entry in oversold conditions rather than confirmation-based entry.

If you want to see what this kind of research looks like built into a complete tradeable systems, that work is at TradingTimeMachine.com.

Next up: the same test with the 20 below the 50 and the 50 below the 200. The fully bearish alignment. And this time we look at the long and potentially short side in poor ranges. Does buying down beat waiting for confirmation when everything points down? The data will tell us.

Dave Johnson

Quant Developer at

TradingTimeMachine.com

via Trading Time Machine https://ift.tt/FXvi1xh

Markov Chains are a foundational tool in probability and quantitative finance, often used to model regime changes such as trending, mean-reverting, high-volatility, and low-volatility market states.

MIT lecture covering the fundamentals of Markov processes and discuss how these concepts can be applied to trading system design, regime classification, and probabilistic forecasting.

Via https://backtest.substack.com/p/using-markov-chains-to-model-market

Markov Chains are a foundational tool in probability and quantitative finance, often used to model regime changes such as trending, mean-reverting, high-volatility, and low-volatility market states.

MIT lecture covering the fundamentals of Markov processes and discuss how these concepts can be applied to trading system design, regime classification, and probabilistic forecasting.

via Trading Time Machine https://ift.tt/9jsxtwA

Most traders wait for the RSI to turn back up before buying. Wait for the bounce. Wait for the confirmation. It sounds reasonable. Twenty five years of data across the entire Nasdaq 100 says it is leaving money on the table in almost every situation you can test.

Waiting for confirmation is one of the most repeated pieces of trading advice out there. The data disagrees with it pretty consistently. But before we get to the numbers, the way you test this question matters. A sloppy comparison produces a misleading answer. So let me walk you through exactly how this was set up and why.

But before we get to the numbers, the way you test this question matters. A sloppy comparison produces a misleading answer. So let me walk you through exactly how this was set up and why.

All testing was done in Wealth-Lab using WealthData, the platform’s built in daily bar data. It is a clean data source checked for bad ticks and data oddities. That matters more than most people realize. Garbage data produces garbage results regardless of how good your system logic is.

The universe is Nasdaq 100 stocks over 25 years. The filter is straightforward. The 20 day moving average above the 50, the 50 above the 200, and SPY above its 200 day moving average. A setup most traders recognize immediately. Entry is at the open the morning after the signal fires. Fixed 5 day hold.

These are not obscure names. We are talking about stocks like NVDA, AMZN, and TSLA. Names that dominate trading forums, social media, and every watchlist worth looking at. The same stocks where most traders are waiting for that RSI crossover confirmation before pulling the trigger.

Now here is where the test design gets important. I used a 7 period RSI and tested every 5 point increment across the full RSI range. For each zone I created two mirror entries that both land in exactly the same place.

Take the 25 to 30 zone as an example. The cross under entry triggers when the RSI crosses down through 30 and closes above 25. The RSI fell into that zone on a down day. The cross over entry triggers when the RSI crosses up through 25 and closes below 30. The RSI rose into that zone on an up day.

Same zone. Same stocks. Same alignment. Same hold period. The only difference is the direction the RSI was traveling when it got there.

That is the cleanest possible apples to apples comparison for this question.

In 17 out of 18 zones tested buying into the falling RSI outperformed waiting for it to turn back up. One exception at the 75 to 80 zone where cross over edges cross under by 0.08%. Both approaches are marginal there anyway. That is not a meaningful win for confirmation. It is noise.

Take the 25 to 30 zone as one example. Buying the cross under produced an average net profit of 0.52%, a profit factor of 1.38, and a win rate of 56.25% across 3,865 trades. Waiting for confirmation in that same zone produced an average net profit of 0.24%, a profit factor of 1.16, and a win rate of 53.19% across 4,189 trades. Both entries profitable but cross under meaningfully better on every metric.

The low RSI zones tell the most dramatic story. In the 15 to 20 zone cross under delivers a profit factor of 1.20 with a win rate of 54.84%. Cross over in the same zone delivers a profit factor of 0.83 and a win rate of 46.89%. Below 1.0 on profit factor means the strategy lost more than it made. In a fully bullish aligned market, waiting for confirmation in a low RSI zone is a losing approach. The data is pretty clear on that.

Think about what these two entries actually look like on a chart. A cross under entry is a down day. Price is falling. It feels uncomfortable. A cross over entry is an up day. Price is rising. It feels safe. That up day is what is eating into your return. By the time the RSI turns back up the stock has already moved. The early money is already in.

Ask yourself why that up day printed. Someone bought the weakness before you did. While most traders were waiting for confirmation, other participants were already entering on the down day. By the time the RSI turns back up those buyers are already sitting on a profit. You are not getting confirmation. You are getting their exit.

The charts below show the full picture across all 18 zones. Take a minute to look at them before reading on. The pattern is hard to miss.

A quick note on what you are looking at. Profit factor measures how many dollars were won for every dollar lost across all trades in that zone. Above 1.0 the winning side outweighed the losing side. Below 1.0 means the strategy lost more than it made. That is the number to focus on.

[Visual 1 - avg net profit bar chart]

[Visual 2 - profit factor with breakeven line]

[Visual 3 - win rate]

[Visual 4 - full metrics table PNG]

A few things worth calling out. Cross over produces a profit factor below 1.0 in five zones. That means a losing strategy in five separate RSI zones within a fully bullish aligned market. Cross under produces a profit factor below 1.0 in exactly one zone, the 75 to 80 band, and even there the difference is marginal. Cross under wins 17 of 18 zones. The average win rate advantage across all zones is 2.73 percentage points. In the 25 to 30 and 30 to 35 zones cross under generates roughly twice the per trade return of cross over across thousands of trades. These are not small sample flukes.

Think about what these two entries actually look like on a chart. A cross under entry is a down day. Price is falling. It feels uncomfortable. A cross over entry is an up day. Price is rising. It feels like the right time to buy.

That up day is what is eating into your return. By the time the RSI turns back up the stock has already moved. The early money is already in.

Ask yourself why that up day printed. Someone bought the weakness before you did. While most traders were waiting for confirmation, other participants were already entering on the down day. By the time the RSI turns back up those buyers are already sitting on a profit. You are not getting confirmation. You are getting their exit.

This shows up in the data pretty clearly. In the lower RSI zones the cross over win rate drops below 50% in several bands. That means the confirmation entry is losing more than half its trades in zones where cross under is winning 54 to 56% of the time. Same zone. Same stocks. The difference is who got there first.

One more thing worth noting. I ran the same test using the standard 14 period RSI that most traders and most platforms default to. The result was the same. Cross under outperformed cross over in every comparable zone tested. The finding is not a function of using a faster RSI setting. It holds on the default that most traders are already using. Down is better than up regardless of which RSI period you prefer.

There is another way to think about these zones beyond pure edge quality. Sometimes you want more exposure in a strong market and you need a reason to pull the trigger. A trade with a documented edge in a bullish aligned stock can accomplish both things at once. You are not just taking a trade. You are adding exposure in a disciplined way that the data supports. That is worth something to a trader who wants to be in the market but does not want to chase.

What you have seen here is raw research. Not a trading system. A complete system requires position sizing, stop methodology, how to handle multiple signals on the same day, portfolio heat management, and whether 5 days is actually the right hold or just a reasonable starting point. None of that is resolved here.

What is resolved is a specific question about entry timing that most traders have never tested with clean methodology. In a bullish aligned Nasdaq 100 stock, buying weakness on a down day outperforms waiting for confirmation on an up day in 17 of 18 RSI zones tested. Across 25 years of data. With a methodology designed specifically to make the comparison fair.

The edge is not dramatic on any single trade. But staying consistently on the correct side of a structural finding like this across hundreds or thousands of trades over time adds up. That is how systematic trading works.

Would you rather know this than not? After 25 years of doing this kind of work I have never once wished I knew less.

If you want to see what this kind of research looks like when it has been developed into a complete tradeable system, that work is at tradingtimemachine.com.

Next up: the same test with a partial bullish alignment. The 20 has crossed under the 50 but the 50 is still above the 200. Does the edge survive when conditions start to deteriorate? The data will tell us.

Have a Great Night!

Dave Johnson

TradingTimeMachine.com

P.S. A quick note for those who follow along regularly. The pace of publishing has been slower than usual lately. We are in the middle of selling the house and preparing for a move to Italy. The house has a buyer and things are moving in the right direction. If all goes well I should have feet on the ground in Italy by early July. Looking forward to getting back to a more regular publishing schedule once the dust settles. Appreciate the patience.

Via https://backtest.substack.com/p/why-waiting-for-rsi-confirmation

Most traders wait for the RSI to turn back up before buying. Wait for the bounce. Wait for the confirmation. It sounds reasonable. Twenty five years of data across the entire Nasdaq 100 says it is leaving money on the table in almost every situation you can test.

Waiting for confirmation is one of the most repeated pieces of trading advice out there. The data disagrees with it pretty consistently. But before we get to the numbers, the way you test this question matters. A sloppy comparison produces a misleading answer. So let me walk you through exactly how this was set up and why.

But before we get to the numbers, the way you test this question matters. A sloppy comparison produces a misleading answer. So let me walk you through exactly how this was set up and why.

All testing was done in Wealth-Lab using WealthData, the platform’s built in daily bar data. It is a clean data source checked for bad ticks and data oddities. That matters more than most people realize. Garbage data produces garbage results regardless of how good your system logic is.

The universe is Nasdaq 100 stocks over 25 years. The filter is straightforward. The 20 day moving average above the 50, the 50 above the 200, and SPY above its 200 day moving average. A setup most traders recognize immediately. Entry is at the open the morning after the signal fires. Fixed 5 day hold.

These are not obscure names. We are talking about stocks like NVDA, AMZN, and TSLA. Names that dominate trading forums, social media, and every watchlist worth looking at. The same stocks where most traders are waiting for that RSI crossover confirmation before pulling the trigger.

Now here is where the test design gets important. I used a 7 period RSI and tested every 5 point increment across the full RSI range. For each zone I created two mirror entries that both land in exactly the same place.

Take the 25 to 30 zone as an example. The cross under entry triggers when the RSI crosses down through 30 and closes above 25. The RSI fell into that zone on a down day. The cross over entry triggers when the RSI crosses up through 25 and closes below 30. The RSI rose into that zone on an up day.

Same zone. Same stocks. Same alignment. Same hold period. The only difference is the direction the RSI was traveling when it got there.

That is the cleanest possible apples to apples comparison for this question.

In 17 out of 18 zones tested buying into the falling RSI outperformed waiting for it to turn back up. One exception at the 75 to 80 zone where cross over edges cross under by 0.08%. Both approaches are marginal there anyway. That is not a meaningful win for confirmation. It is noise.

Take the 25 to 30 zone as one example. Buying the cross under produced an average net profit of 0.52%, a profit factor of 1.38, and a win rate of 56.25% across 3,865 trades. Waiting for confirmation in that same zone produced an average net profit of 0.24%, a profit factor of 1.16, and a win rate of 53.19% across 4,189 trades. Both entries profitable but cross under meaningfully better on every metric.

The low RSI zones tell the most dramatic story. In the 15 to 20 zone cross under delivers a profit factor of 1.20 with a win rate of 54.84%. Cross over in the same zone delivers a profit factor of 0.83 and a win rate of 46.89%. Below 1.0 on profit factor means the strategy lost more than it made. In a fully bullish aligned market, waiting for confirmation in a low RSI zone is a losing approach. The data is pretty clear on that.

Think about what these two entries actually look like on a chart. A cross under entry is a down day. Price is falling. It feels uncomfortable. A cross over entry is an up day. Price is rising. It feels safe. That up day is what is eating into your return. By the time the RSI turns back up the stock has already moved. The early money is already in.

Ask yourself why that up day printed. Someone bought the weakness before you did. While most traders were waiting for confirmation, other participants were already entering on the down day. By the time the RSI turns back up those buyers are already sitting on a profit. You are not getting confirmation. You are getting their exit.

The charts below show the full picture across all 18 zones. Take a minute to look at them before reading on. The pattern is hard to miss.

A quick note on what you are looking at. Profit factor measures how many dollars were won for every dollar lost across all trades in that zone. Above 1.0 the winning side outweighed the losing side. Below 1.0 means the strategy lost more than it made. That is the number to focus on.

[Visual 1 - avg net profit bar chart]

[Visual 2 - profit factor with breakeven line]

[Visual 3 - win rate]

[Visual 4 - full metrics table PNG]

A few things worth calling out. Cross over produces a profit factor below 1.0 in five zones. That means a losing strategy in five separate RSI zones within a fully bullish aligned market. Cross under produces a profit factor below 1.0 in exactly one zone, the 75 to 80 band, and even there the difference is marginal. Cross under wins 17 of 18 zones. The average win rate advantage across all zones is 2.73 percentage points. In the 25 to 30 and 30 to 35 zones cross under generates roughly twice the per trade return of cross over across thousands of trades. These are not small sample flukes.

Think about what these two entries actually look like on a chart. A cross under entry is a down day. Price is falling. It feels uncomfortable. A cross over entry is an up day. Price is rising. It feels like the right time to buy.

That up day is what is eating into your return. By the time the RSI turns back up the stock has already moved. The early money is already in.

Ask yourself why that up day printed. Someone bought the weakness before you did. While most traders were waiting for confirmation, other participants were already entering on the down day. By the time the RSI turns back up those buyers are already sitting on a profit. You are not getting confirmation. You are getting their exit.

This shows up in the data pretty clearly. In the lower RSI zones the cross over win rate drops below 50% in several bands. That means the confirmation entry is losing more than half its trades in zones where cross under is winning 54 to 56% of the time. Same zone. Same stocks. The difference is who got there first.

One more thing worth noting. I ran the same test using the standard 14 period RSI that most traders and most platforms default to. The result was the same. Cross under outperformed cross over in every comparable zone tested. The finding is not a function of using a faster RSI setting. It holds on the default that most traders are already using. Down is better than up regardless of which RSI period you prefer.

There is another way to think about these zones beyond pure edge quality. Sometimes you want more exposure in a strong market and you need a reason to pull the trigger. A trade with a documented edge in a bullish aligned stock can accomplish both things at once. You are not just taking a trade. You are adding exposure in a disciplined way that the data supports. That is worth something to a trader who wants to be in the market but does not want to chase.

What you have seen here is raw research. Not a trading system. A complete system requires position sizing, stop methodology, how to handle multiple signals on the same day, portfolio heat management, and whether 5 days is actually the right hold or just a reasonable starting point. None of that is resolved here.

What is resolved is a specific question about entry timing that most traders have never tested with clean methodology. In a bullish aligned Nasdaq 100 stock, buying weakness on a down day outperforms waiting for confirmation on an up day in 17 of 18 RSI zones tested. Across 25 years of data. With a methodology designed specifically to make the comparison fair.

The edge is not dramatic on any single trade. But staying consistently on the correct side of a structural finding like this across hundreds or thousands of trades over time adds up. That is how systematic trading works.

Would you rather know this than not? After 25 years of doing this kind of work I have never once wished I knew less.

If you want to see what this kind of research looks like when it has been developed into a complete tradeable system, that work is at tradingtimemachine.com.

Next up: the same test with a partial bullish alignment. The 20 has crossed under the 50 but the 50 is still above the 200. Does the edge survive when conditions start to deteriorate? The data will tell us.

Have a Great Night!

Dave Johnson

TradingTimeMachine.com

P.S. A quick note for those who follow along regularly. The pace of publishing has been slower than usual lately. We are in the middle of selling the house and preparing for a move to Italy. The house has a buyer and things are moving in the right direction. If all goes well I should have feet on the ground in Italy by early July. Looking forward to getting back to a more regular publishing schedule once the dust settles. Appreciate the patience.

via Trading Time Machine https://ift.tt/ndS72pX

New traders tend to obsess over win rate.

It’s usually one of the first metrics people look at when evaluating a trading strategy, and unfortunately, it’s also one of the easiest ways to completely misread a system.

A 70% win rate sounds fantastic. A 45% win rate sounds broken.

That instinct is understandable—but after 25 years of building and backtesting trading systems, I can tell you win rate by itself is one of the least useful standalone metrics in strategy analysis.

Win rate only tells you how often a system is right.

It tells you nothing about how much you make when you’re right, how much you lose when you’re wrong, or what the return path looks like over time. Without those numbers, you’re only seeing a small piece of the picture.

I’ve seen systems with 65% winning trades steadily lose money because the average losing trade was significantly larger than the average winner.

I’ve also seen systems with only 45% winners generate excellent long-term returns because winners were meaningful and losses were tightly controlled.

This is why experienced system developers don’t start with win rate.

They start with expectancy.

Expectancy: The Real Engine of a Trading Edge

Expectancy measures the average expected gain or loss per trade over time.

The formula is simple:

Expectancy = (Win Rate × Average Winner) – (Loss Rate × Average Loser)

If expectancy is positive, the system has a mathematical edge.

That doesn’t mean every week will be profitable, or that every month will look clean. But over a sufficiently large sample of trades, positive expectancy is what gives a trading strategy long-term viability.

This is where many traders make a costly mistake.

They optimize for a prettier win rate while unintentionally destroying expectancy.

A system can feel psychologically better because it wins more often while actually becoming mathematically worse.

That’s like bragging about winning more poker hands while losing money overall.

Profit Factor Matters, But It Is Not the Full Story

Another useful backtesting metric is profit factor:

Profit Factor = Gross Profit / Gross Loss

A profit factor above 1.0 indicates profitability. The higher the number, the more profitable the system has been relative to its losses.

But profit factor alone can be misleading.

A strategy with a 2.3 profit factor sounds outstanding until you discover it only generated 15 trades over the course of a year.

Meanwhile, a system with a 1.35 profit factor producing 400 trades annually may generate far stronger compounded returns.

Trade frequency matters.

A moderate edge repeated many times can outperform an excellent edge with limited opportunity.

This is why system evaluation cannot be reduced to a single metric. Win rate, expectancy, profit factor, and trade frequency all interact.

Annual Return Means Little Without Risk-Adjusted Returns

Even annual return does not tell the full story.

Two systems can both generate 18% annualized returns and be completely different experiences to trade.

One may produce relatively smooth equity growth with manageable drawdowns and low monthly volatility.

The other may arrive at the same return through violent swings, extended drawdowns, and highly uneven performance.

Same annual return. Completely different path.

This is where risk-adjusted returns become critical.

Metrics like Sharpe Ratio, Sortino Ratio, and maximum drawdown help measure how efficiently a system produces returns relative to the volatility and downside risk required to achieve them.

At a high level, annual return is the numerator. Volatility is part of the denominator.

A system producing 20% annually with severe drawdowns and unstable equity swings is often less attractive than a system generating 14% with smoother behavior and lower portfolio stress.

In many cases, that smoother 14% system is actually more valuable because it can potentially be scaled or modestly levered to target higher returns while still maintaining a more tolerable drawdown profile than the naturally volatile 20% system.

That is a concept many traders overlook.

Higher raw return does not automatically mean a superior strategy if it already comes packaged with excessive volatility.

A smoother equity curve gives you more flexibility.

This is not just about statistics.

The path of returns directly impacts whether a trader can realistically stick with a strategy in live trading.

A mathematically sound system is worthless if its volatility profile causes you to abandon it during a completely normal drawdown.

This is one of the biggest disconnects between backtesting and live execution.

Backtesting is easy when you are looking backward.

Living through a six-month drawdown is something else entirely.

Why Traders Get Trapped by High Win Rates

The attraction to high win rate systems is mostly psychological.

Losses feel bad.

A system with an 80% win rate creates fewer losing streaks, fewer red days, and less emotional discomfort. It simply feels easier to trade.

That feeling often gets confused with quality.

But comfort is not edge.

Many traders unknowingly optimize their systems for emotional comfort rather than mathematical robustness.

This happens constantly in system development.

Backtests showing 45% to 50% win rates often get dismissed immediately, even when expectancy and risk-adjusted returns are excellent.

Meanwhile, high win rate systems that quietly erode capital often get a free pass because nobody looked beyond the headline metric.

What Backtesters Should Actually Analyze

Before getting excited about win rate, evaluate these metrics first:

• Average winning trade

• Average losing trade

• Expectancy

• Profit factor

• Maximum drawdown

• Sharpe or Sortino ratio

• Number of trades per year

These metrics tell you far more about whether a strategy has a durable edge.

Win rate is simply one input.

Test Your Assumptions With Monte Carlo Simulation

A trading edge often looks very different once randomness enters the equation.

That is why I built a free Monte Carlo simulator on this site. (That can be found here) - scroll down the Trader Tools page to find the simulator.

You can plug in your own:

• win rate

• average winner

• average loser

• number of trades

Then simulate hundreds of potential equity curves to see how the same statistical edge can produce dramatically different return paths over time.

For many traders, this is where the light bulb finally goes on.

They stop obsessing over win rate and start understanding the full statistical structure of a trading system.

And that is when strategy development starts becoming much more interesting.

Have a Great Night!

Dave Johnson

TradingTimeMachine.com

via Trading Time Machine https://ift.tt/ZF5JIoE

New traders tend to obsess over win rate.

It’s usually one of the first metrics people look at when evaluating a trading strategy, and unfortunately, it’s also one of the easiest ways to completely misread a system.

A 70% win rate sounds fantastic. A 45% win rate sounds broken.

That instinct is understandable—but after 25 years of building and backtesting trading systems, I can tell you win rate by itself is one of the least useful standalone metrics in strategy analysis.

Win rate only tells you how often a system is right.

It tells you nothing about how much you make when you’re right, how much you lose when you’re wrong, or what the return path looks like over time. Without those numbers, you’re only seeing a small piece of the picture.

I’ve seen systems with 65% winning trades steadily lose money because the average losing trade was significantly larger than the average winner.

I’ve also seen systems with only 45% winners generate excellent long-term returns because winners were meaningful and losses were tightly controlled.

This is why experienced system developers don’t start with win rate.

They start with expectancy.

Expectancy: The Real Engine of a Trading Edge

Expectancy measures the average expected gain or loss per trade over time.

The formula is simple:

Expectancy = (Win Rate × Average Winner) – (Loss Rate × Average Loser)

If expectancy is positive, the system has a mathematical edge.

That doesn’t mean every week will be profitable, or that every month will look clean. But over a sufficiently large sample of trades, positive expectancy is what gives a trading strategy long-term viability.

This is where many traders make a costly mistake.

They optimize for a prettier win rate while unintentionally destroying expectancy.

A system can feel psychologically better because it wins more often while actually becoming mathematically worse.

That’s like bragging about winning more poker hands while losing money overall.

Profit Factor Matters, But It Is Not the Full Story

Another useful backtesting metric is profit factor:

Profit Factor = Gross Profit / Gross Loss

A profit factor above 1.0 indicates profitability. The higher the number, the more profitable the system has been relative to its losses.

But profit factor alone can be misleading.

A strategy with a 2.3 profit factor sounds outstanding until you discover it only generated 15 trades over the course of a year.

Meanwhile, a system with a 1.35 profit factor producing 400 trades annually may generate far stronger compounded returns.

Trade frequency matters.

A moderate edge repeated many times can outperform an excellent edge with limited opportunity.

This is why system evaluation cannot be reduced to a single metric. Win rate, expectancy, profit factor, and trade frequency all interact.

Annual Return Means Little Without Risk-Adjusted Returns

Even annual return does not tell the full story.

Two systems can both generate 18% annualized returns and be completely different experiences to trade.

One may produce relatively smooth equity growth with manageable drawdowns and low monthly volatility.

The other may arrive at the same return through violent swings, extended drawdowns, and highly uneven performance.

Same annual return. Completely different path.

This is where risk-adjusted returns become critical.

Metrics like Sharpe Ratio, Sortino Ratio, and maximum drawdown help measure how efficiently a system produces returns relative to the volatility and downside risk required to achieve them.

At a high level, annual return is the numerator. Volatility is part of the denominator.

A system producing 20% annually with severe drawdowns and unstable equity swings is often less attractive than a system generating 14% with smoother behavior and lower portfolio stress.

In many cases, that smoother 14% system is actually more valuable because it can potentially be scaled or modestly levered to target higher returns while still maintaining a more tolerable drawdown profile than the naturally volatile 20% system.

That is a concept many traders overlook.

Higher raw return does not automatically mean a superior strategy if it already comes packaged with excessive volatility.

A smoother equity curve gives you more flexibility.

This is not just about statistics.

The path of returns directly impacts whether a trader can realistically stick with a strategy in live trading.

A mathematically sound system is worthless if its volatility profile causes you to abandon it during a completely normal drawdown.

This is one of the biggest disconnects between backtesting and live execution.

Backtesting is easy when you are looking backward.

Living through a six-month drawdown is something else entirely.

Why Traders Get Trapped by High Win Rates

The attraction to high win rate systems is mostly psychological.

Losses feel bad.

A system with an 80% win rate creates fewer losing streaks, fewer red days, and less emotional discomfort. It simply feels easier to trade.

That feeling often gets confused with quality.

But comfort is not edge.

Many traders unknowingly optimize their systems for emotional comfort rather than mathematical robustness.

This happens constantly in system development.

Backtests showing 45% to 50% win rates often get dismissed immediately, even when expectancy and risk-adjusted returns are excellent.

Meanwhile, high win rate systems that quietly erode capital often get a free pass because nobody looked beyond the headline metric.

What Backtesters Should Actually Analyze

Before getting excited about win rate, evaluate these metrics first:

• Average winning trade

• Average losing trade

• Expectancy

• Profit factor

• Maximum drawdown

• Sharpe or Sortino ratio

• Number of trades per year

These metrics tell you far more about whether a strategy has a durable edge.

Win rate is simply one input.

Test Your Assumptions With Monte Carlo Simulation

A trading edge often looks very different once randomness enters the equation.

That is why I built a free Monte Carlo simulator on this site. (That can be found here) - scroll down the Trader Tools page to find the simulator.

You can plug in your own:

• win rate

• average winner

• average loser

• number of trades

Then simulate hundreds of potential equity curves to see how the same statistical edge can produce dramatically different return paths over time.

For many traders, this is where the light bulb finally goes on.

They stop obsessing over win rate and start understanding the full statistical structure of a trading system.

And that is when strategy development starts becoming much more interesting.

Have a Great Night!

Dave Johnson

TradingTimeMachine.com

Via https://backtest.substack.com/p/why-win-rate-is-one-of-the-most-misunderstood

So you want to build a trading system. Maybe you’ve already tried. You found an indicator that looked great on NVDA or TSLA, ran a backtest, got excited about the results, and then watched it fall apart the moment you traded it live or tried it on anything else.

That’s not bad luck. That’s a process problem.

Whether you’re trying to build a mechanical trading system for the first time or you’ve been down this road before, the mistakes are almost always the same. People skip the boring foundational stuff and jump straight to the fun part, which is running backtests and watching equity curves go up. It feels like progress. But you’re basically building a house without a foundation.

Algorithmic trading and system building look complicated from the outside, but the core process is actually pretty straightforward. There are five steps, and most people only follow two or three of them. Here’s what the full process actually looks like.

Step 1: Sit With the Data Before You Do Anything Else

Most people come into this with an idea already in their head. They read something online, heard about a setup that “always works,” and now they want to test it. So they go hunting through the data for proof that they’re right.

That’s backwards.

Before you form any strong opinions, just look at the data. Pull up charts across different years, different market conditions, calm periods and crazy ones. Watch how prices move, where volume spikes, what happens after big drops. Don’t try to find anything yet. Just look.

One tool I use for this is the Genetic Evolver in Wealth-Lab. It’s essentially a way to let the software loose on a large dataset and surface small edges you might never think to look for on your own. It’s not about handing the wheel over to an algorithm. It’s more like using it as a research assistant to point you toward areas worth digging into. You still have to make sense of what it finds.

This is the part nobody wants to do because it feels like you’re not making progress. But what you’re really doing is building intuition. You’re learning what the market actually does, not what you think it does. That’s the whole game.

Step 2: Come Up With a Simple Hypothesis

After spending real time with the data, something usually starts to jump out. A pattern that repeats. A behavior that shows up consistently. A tendency.

That’s your hypothesis.

It doesn’t need to be complicated. In fact, the simpler the better. Something like: “When the market sells off hard on big volume, it tends to bounce within a few days.” Or: “When volatility is low and the market starts trending, it tends to keep going.”

The key word there is “tends.” You’re not looking for something that works every single time. Nothing does. You’re just looking for a repeatable edge that you can actually explain in plain English.

Here’s a useful test: try explaining your hypothesis to someone who doesn’t trade. If you need a chart open to get through it, the idea is probably either too complicated or not fully formed yet. That matters, because if you can’t explain why something works, you won’t know when it stops working.

Step 3: Write Down the Rules Before You Touch Your Software

This is where things get real, and where a lot of people get sloppy.

Take your hypothesis and turn it into an actual set of rules. When do you get in? When do you get out? How do you size the position? What happens if it goes against you right away?

Be specific. The goal is to write it out clearly enough that someone else could follow your rules and make the exact same decisions you would. No guessing, no interpreting.

And keep it simple. Seriously. Every extra rule you tack on is another way for the system to fit the past instead of capturing something that will actually repeat. Simple systems are almost always more durable than complex ones.

The habit worth building here: write the rules out in plain language before you ever open your backtesting platform. If you can’t describe the logic without a screen in front of you, go back and think it through some more.

Step 4: Test It to Learn, Not to Confirm

Here’s where most people go off the rails.

They run a backtest hoping to see a beautiful equity curve. If it looks good, they declare victory. If it doesn’t, they start tweaking things until it does. A rule gets adjusted here, a parameter gets tuned there, and suddenly the backtest looks great.

But now you haven’t tested a hypothesis. You’ve just reverse-engineered the past. That’s called curve fitting, and it’s one of the most common reasons systems fall apart in live markets.

The real purpose of a backtest is to understand how the system behaves, not to prove it works. You want to know where it struggles. What does a bad stretch look like? How deep do the drawdowns get? Does it hold up across different years or is all the performance stacked in one particular period?

A system that only shined during the 2020 crash or the 2013 to 2019 bull run isn’t a robust system. It’s a system that got lucky in a specific environment.

One thing that helps: write down what you expect to see before you run the test. If your hypothesis is about mean reversion after big drops and the backtest shows it only works in trending markets, something doesn’t add up. Either the rules don’t match the idea, or the idea isn’t what you thought it was.

Step 5: Honestly Evaluate What You’ve Got

You’ve got results. Now comes the most important part of the whole process, which is being honest about what you actually have.

Start by widening the scope. If you built and tested this on one symbol, take it out for a longer walk. Does the same logic hold up on other ETFs, other markets, other sectors? A system that only works on one ticker is a fragile thing. Maybe it’s capturing something real, or maybe it just fit that one ticker’s history. The only way to find out is to test it more broadly.

Then ask some harder questions. Is the performance consistent year over year, or is it basically one or two great years carrying the whole backtest? How bad do the drawdowns get, and how long do they last before recovering?

And if you already have other systems running, here’s a question worth thinking about: does this new one behave differently from what you already have? Pull the equity curves up side by side. Do they tend to hit rough patches at the same time, or do they struggle independently of each other? You’re not chasing perfect non-correlation, that’s basically impossible to achieve in practice. You’re just looking for something that adds a different flavor to what you already have. Two systems that don’t both blow up at the same time is a genuinely good thing.

If the system holds up across all of that, it earns a spot in your research. If it doesn’t, you go back to step one. No shame in that. Most ideas don’t survive this process, and that’s exactly the point.

Over time you’ll build a core set of principles that tend to be advantageous across symbols and timeframes. These can become plug and play into future data sets and watchlists.

This Is a Loop, Not a Checklist

The thing people don’t tell you when you start down this road is that you never really finish. The process doesn’t end when you find something that works.

Markets change. What worked three years ago might quietly stop working tomorrow. The traders who stick around are the ones who treat this as an ongoing practice, not a one-time project. You’re always somewhere in the loop.

With the systems I provide publicly, that’s the way I think about it. Every system goes through this process. No shortcuts, no skipping the boring parts.

If you’re just getting started, that’s actually the best possible position to be in. You get to build the right habits from the beginning. Start with the process, and the rest gets a lot easier from there.

Have a Great Night!

Dave Johnson

TradingTimeMachine.com

via Trading Time Machine https://ift.tt/B7zw3bL