iPresageDRAINING Regime: What Happens Next? A Historical Analysis
By iPresage Research · 9 min read · 2025-01-28
Analysis of 83 DRAINING periods over 7 years shows the regime's end predicts above-average returns, with bearish diamond signals winning 58.6% during it.
When the iPresage scanner classifies the market as DRAINING, it indicates a regime characterized by deteriorating options flow, contracting institutional participation, falling EV scores, and weakening pulse readings across sectors. It is, in plain terms, the scanner's way of saying the market is under distribution. But what does this classification actually predict? How long do DRAINING regimes last, what do returns look like during and after them, and how should traders adjust their signal interpretation?
We analyzed every DRAINING regime classification from iPresage's 7-year history, spanning January 2018 through December 2024. The dataset includes 83 distinct DRAINING periods, defined as consecutive trading days where the scanner maintained a DRAINING classification without reverting to NEUTRAL or SURGING for more than one session.
DEFINING THE DRAINING REGIME
The iPresage regime classification is derived from a composite of options flow metrics, including aggregate put-call ratios weighted by premium, institutional versus retail flow proportions, changes in implied volatility term structure, and cross-sector pulse score trends. A DRAINING classification requires that the composite falls below a threshold calibrated against historical distributions.
It is important to understand what DRAINING is not. It is not a measure of price direction -- a market can be DRAINING while prices remain flat or even rise temporarily. It is a measure of flow quality and participation. In practice, however, flow deterioration tends to precede price weakness, which is why the classification has predictive value.
Of the 1,764 trading days in our 7-year sample, 389 were classified as DRAINING (22.1%), 627 as NEUTRAL (35.5%), and 748 as SURGING (42.4%). DRAINING is the least common regime, which makes sense: periods of genuine distribution are typically shorter than accumulation or trend-following phases.
DURATION AND FREQUENCY
The 83 DRAINING periods in our sample had a mean duration of 4.7 trading days and a median duration of 3 days. The distribution is heavily right-skewed: 61 of the 83 periods (73.5%) lasted 5 days or fewer, while only 7 (8.4%) lasted more than 10 days.
The longest DRAINING period was 28 trading days, occurring from September 12 to October 21, 2022, during the aggressive Fed tightening cycle and the gilts crisis. The second longest was 19 days in February-March 2020, immediately preceding the COVID crash. Extended DRAINING regimes are rare but tend to coincide with genuine macro dislocations.
DRAINING periods cluster seasonally. We observed 31.3% more DRAINING days in September-October than in any other two-month period, consistent with the well-documented seasonal weakness in equity markets. January and April showed the fewest DRAINING days, aligning with typical post-holiday and earnings-season optimism.
RETURNS DURING DRAINING REGIMES
SPY returns during DRAINING periods averaged -0.09% per day, compared to +0.02% per day during NEUTRAL and +0.07% per day during SURGING regimes. While the per-day difference seems small, it compounds: a typical 5-day DRAINING period produced a cumulative SPY return of -0.43%, compared to +0.10% during NEUTRAL and +0.37% during SURGING periods of the same length.
The dispersion of returns is equally important. The standard deviation of daily returns during DRAINING was 1.34%, compared to 1.08% during NEUTRAL and 0.96% during SURGING. DRAINING regimes are not just directionally negative on average -- they are more volatile, which has direct implications for options pricing and strategy selection.
Sector performance during DRAINING regimes showed significant dispersion. Defensive sectors outperformed cyclicals consistently. Utilities averaged +0.02% per day during DRAINING periods, while Technology averaged -0.14% per day. This 16 basis points of daily differential creates meaningful divergences over multi-day DRAINING windows.
WHAT HAPPENS AFTER A DRAINING REGIME ENDS
This is the question most traders care about. When the scanner transitions from DRAINING to NEUTRAL or SURGING, what do subsequent returns look like?
We tracked SPY performance over the 5, 10, and 20 trading days following the end of each of the 83 DRAINING periods:
At T+5 (one week after DRAINING ends), the average SPY return was +0.68%. The win rate -- defined as a positive return over the window -- was 62.7% (52 of 83 instances). This is modestly above the baseline T+5 win rate of 57.2% across all periods.
At T+10 (two weeks), the average return was +1.24% with a win rate of 65.1% (54 of 83). This represents a meaningful improvement over the all-period T+10 win rate of 58.9%.
At T+20 (one month), the average return was +2.17% with a win rate of 68.7% (57 of 83), compared to the all-period T+20 win rate of 61.4%.
The pattern is clear: the end of a DRAINING regime is a statistically significant bullish signal for subsequent returns. The effect is strongest in the first two weeks and remains present but diminishing through the first month.
However, the distribution has a long left tail. In the worst case -- the DRAINING period ending in late February 2020 -- the T+20 return was -26.1% as the COVID crash unfolded. Excluding the three worst outcomes (all associated with the 2020 crash), the T+20 average return improves to +2.89% and the win rate to 71.3%.
DIAMOND SIGNAL PERFORMANCE BY REGIME
This is where the analysis becomes directly actionable for iPresage users. We tracked the win rate and EV score of all diamond signals generated during each regime classification.
During DRAINING regimes, the overall diamond signal win rate was 50.3% with an average EV of +0.8. This is materially below the all-regime average of 54.1% win rate and +2.1 EV. The scanner still generates useful signals during DRAINING, but their reliability is reduced.
The asymmetry is critical. Bullish diamond signals during DRAINING had a win rate of just 44.7% and an average EV of -1.2. These signals are actively harmful on average -- they are worse than random. Bearish diamond signals during DRAINING, by contrast, had a win rate of 58.6% and an average EV of +3.9. The DRAINING regime validates bearish signals and invalidates bullish ones.
This asymmetry is even more pronounced when combined with signal quality. High-quality bearish diamond signals (those in the top quartile by the scanner's internal confidence score) during DRAINING regimes posted a 65.2% win rate and +6.4 average EV across 214 signals over 7 years. This is one of the highest-performing signal subsets in the entire iPresage system.
THE TRANSITION SIGNAL
We identified a particularly powerful signal at the moment the regime transitions from DRAINING to SURGING -- a direct flip without an intermediate NEUTRAL classification. This occurred 18 times in our 7-year sample.
Bullish diamond signals generated within 3 trading days of a DRAINING-to-SURGING transition posted a 72.8% win rate and +8.3 average EV across 94 signals. This transition signal captures the moment when distribution gives way to aggressive accumulation, and the dislocations it creates are among the most profitable in the scanner's history.
The average SPY return in the 10 trading days following a DRAINING-to-SURGING flip was +2.86%, roughly double the average post-DRAINING return when the transition goes through NEUTRAL. The flip signal is less common but substantially more powerful.
VOLATILITY DYNAMICS DURING DRAINING
Implied volatility behaves predictably during DRAINING regimes. The VIX averaged 22.4 during DRAINING periods compared to 17.8 during NEUTRAL and 15.1 during SURGING. The VIX term structure was in backwardation (near-term IV higher than longer-term IV) on 41.6% of DRAINING days, compared to just 8.7% of SURGING days.
For options sellers, DRAINING regimes offer elevated premiums but heightened risk. We tested selling SPY at-the-money straddles at the start of DRAINING periods and closing them when the regime ended. Over 83 instances, this strategy produced a positive return 53.0% of the time with an average return of +0.4% on notional. This is a marginal edge at best, and the tail risk is substantial -- the worst outcome was -14.2%.
A better approach is selling straddles at the regime transition from DRAINING to NEUTRAL or SURGING, capturing the volatility crush that accompanies regime improvement. This variation won 71.1% of the time across 83 instances with an average return of +2.7% on notional and a maximum drawdown of -4.8%.
PRACTICAL FRAMEWORK FOR DRAINING REGIMES
Based on 83 DRAINING periods spanning 7 years, we propose the following framework:
This study is based on historical data and retrospective analysis. Regime classifications are generated by iPresage's proprietary model and may change in methodology over time. All options trading involves risk of loss. Past performance does not guarantee future results.